Dynamic Heatmaps — The New E-commerce Data Gathering Weapon

Dynamic heatmaps give e-commerce companies the leverage of study real time customer behavior on pages dyanamic URLs

The post Dynamic Heatmaps — The New E-commerce Data Gathering Weapon appeared first on Blog.

E-commerce is a pure service business that demands utmost precision, patience, and persistence to survive the industry’s ever growing competition. The goal is not just to help shoppers find good products, but also provide an exceptional customer experience that forces them to convert into repeat buyers.

Also, customer actions serve as a valuable source of data at every stage of an e-commerce’s funnel. What you do with these customer insights to improve the overall shopping experience and conversion rate, is typically what makes or breaks a deal for your business.  

Most marketers suggest using multiple qualitative and quantitative analytical tools to do so. But, not all of them have the prowess to give you the kind of information you would need to study user experience.

This is where heatmaps come into play. Let’s see how heatmaps can improve your E-commerce business!


What are heatmaps?

Heatmaps serve as one of the best qualitative tools to collect relevant customer data, especially in terms of understanding their actual behavior across your web platform.  

Think of heatmaps as X-Ray films. They show a detailed picture (in the form of graphical representation) of how users interact with your site or store. You can observe how far users scroll, where they click the most, and the products or pages they like or dislike.

Such data is precisely what you need to make your platform more user-friendly, drive more traffic and improve your conversion optimization strategies.

But, if you think that any heatmap would work wonders for your e-commerce site, you’re absolutely wrong!

E-commerce websites are highly dynamic in nature. They have more interactive elements and “behind login page” elements such orders, cart page, etc. than any other site. Such pages shed visitor interaction and information, which typically serves as the primary data for businesses to use to draw page performance conclusions, find elemental distractions, and improve overall customer experience.    

This is where Dynamic Heatmaps save the day!

What are dynamic heatmaps?

Unlike static heatmaps which can only be plotted on static web pages such as the Home Page, Product Pages, Landing Pages, Category Pages, etc., dynamic heatmaps give you the leverage of studying real time customer behavior on pages which are beyond the scope of static heatmaps. In other words, dynamic heatmaps can be easily plotted on live websites with dynamic URLs such as My Profile, Orders, Cart, Account Settings, etc. to gather in-depth customer activity data.

In general, a typical dynamic heatmap offers four primary features, namely, click maps, scroll maps, click areas, and element list. Each of these allow you to look at a web page’s hot spot areas in a detailed manner.     

Let’s now understand the scope of dynamic heatmaps for e-commmerce using some examples!

Studying Cart Page Insights Using Dynamic Heatmaps

The image below gives us an insight a dynamic heatmap plotted on an e-commerce site’s Cart page. It shows that in general, most people, after adding products to their cart, click on the product image, cancel button, postcode section, ‘Go To Checkout’ icon and discount codes.

Such information helps you draw multiple conclusions. Some of which are as follows:

  • Customers might want to see the product images once more before proceeding to final payment gateway.
  • They may not like their chosen product(s) and hence, remove the item(s) from the cart. Or, since they’re not able to view the product image (in zoom) on the cart page, they abandon their cart.
  • They’re most interested in availing discounts. Therefore, the area is hotter than other page elements.

Furthermore, such information also serves helpful in drawing hypothesis on how to improve the performance of various page elements with dynamic URLs, meanwhile find the right means and ways to enhance customer experience and increase your conversions.

“VWO’s dynamic heatmaps allow you to access information of those web pages which are, in general, very difficult to access using regular heatmaps.”

Studying Order Page Insights Using Dynamic Heatmaps

The “My Orders” page of an e-commerce site is an important page that is lesser explored in terms of gathering customer behavior data.

The page allows users to look at their current and previous orders, check delivery status, seek assistance, and even browse through their history. As an e-commerce company, plotting heatmaps on such pages and mapping number of clicks can significantly help you study which page elements are fetching you maximum customer attention.  

For example, the plotted above shows that maximum people are clicking on the “Track” icon followed by “Need Help” and product images. They’re hardly clicking on the “Rate & Review Product” icon, which, according to your platform, can be an important form page to seek product reviews and other essential information. You can further use such qualitative data to make necessary amendments and compel customers to fill the form, like:

  • adding product review pop-ups or call-to-action buttons on the Order page
  • making the section omnipresent
  • giving rewards for adding reviews

So now that you know what your static heatmap tools are lacking, it’s time to upgrade your e-commerce platform with VWO’s dynamic heatmaps and make the most of them.  

The post Dynamic Heatmaps — The New E-commerce Data Gathering Weapon appeared first on Blog.

16 AdWords (Google Ads) eCommerce Strategies to Maximize ROAS

Discover our favorite strategies for improving AdWords (Google Ads) eCommerce performance covering Google Shopping Ads and more. Find out the exact strategy used to maximize ROAS for our clients.

In this piece we’re going to take a systematic look at how we optimize Google Ads (formerly known as AdWords) eCommerce campaigns to find the best performing strategies.

We’ll be covering a total of 16 different strategies that apply for Google Shopping ads and Smart Shopping Campaigns, Dynamic Remarketing, Search Ads, In-Market Audiences and finally Dynamic Search Ads.

For each Google Ad type, we’ll be covering:

  • The core strategies we use to get the best return on ad spend (ROAS)
  • Use-case scenarios where we’ve seen good results from implementing these strategies.
  • Some of the newest features and product offerings — and how to make the most of them.

Because these strategies all fall inside two of Google’s main product offerings, we’ve split this guide up into two parts:

Part 1: Google Shopping ads

We’ll walk you through 7 different strategies that bring the best ROAS for eCommerce companies. We’ve gone into a lot of depth explaining how you should be building out and optimizing your shopping ads from scratch.

Part 2: Google Search ads

Here we cover 9 of our favorite internal strategies for maximizing ROAS from search ads. We’ll also answer the question: does my company need to run search ads in tandem with Shopping Ads?

  1. Google Shopping Campaigns (or Product Listing Ads/PLA’s)
  2. Google Search ads

First we’ll start with a quick review of the different Google Ads platforms and their use cases. If you know these already, feel free to skip ahead using the links above to get straight to the strategies of choice.

(Who We Are: At Inflow, we work with dozens of eCommerce companies to increase traffic, conversions, and sales. You can talk to one of our Google Ads specialists to see how we can help you increase ROAS. Get started now.)

What’s the Best Google Ads Platform for Selling My Products?

You’re probably already using Google Ads (formerly known as Google AdWords) to sell your products, but if you’re not, it’s important to have a bit of background info on the Google Ads ecosystem.

Google Shopping ads (also known as Google Product Listing Ads, or PLA’s) are probably the best fit if you’re a B2C selling products online.

eCommerce Ads Strategy: Google Shopping ads are useful if you're a B2C selling products online.

All you need to participate here is a product feed and an eCommerce website.

Google Search ads is perhaps the most well-known of the Ads products due to its longevity and it’s where your text ads are displayed when they match keywords you specify.

Search Ads is different to Shopping Ads because of the way the platform operates. Search Ads gives you more control over keywords, and in turn, shows text ads within your Google search results.

Google Search ads give more control over keywords and are displayed with text ads.

Shopping Ads, on the other hand, are based upon your Product Feed — which needs careful optimization to target the right searches. The product feed contains all the necessary information relating to your product: brands, quantities, sizes, colors, and so on.

This renders a shopping-product ad within the Google SERPS, as well as the relevant pricing and review information.

If you’re a big online retailer, you’ll probably be investing most of your paid ad spend on a mixture of Google Shopping ads and Google Search ads. Participating in both platforms often leads to enhanced product visibility across the customer buyer’s entire journey, from research through to purchase.

This is key when you consider how the customer’s research and purchase journey spans across multiple devices and is made up of many micro-moments — even if you’re only running search ads to cover branded queries.

Our Tried-and-Tested Google Shopping Strategies for Maximizing ROAS

Here at Inflow, we’ve helped hundreds of online retailers from all kinds of industries to maximize their return on ad-spend (ROAS). This is the number one goal afterall.

What follows are 8 of the strategies that have worked best for us on Google Shopping ads over the years.

Note: Do You Trust the Reliability of Your Google Analytics Data?

It goes without saying that you need to be 100% sure of the accuracy of any analytics data before you start investing in Google Ads — or in any other kind of marketing activity.

We always spend time digging into Google Analytics, or whichever reporting tool is being used, to ensure the historic data looks clean and there are no nasty surprises.

If we can’t measure performance to a good degree of certainty, we can’t measure the growth we’re about to deliver. (Then we can’t create case studies like this one.)

Having a reporting tool is not a prerequisite, but we love it when a client comes to us with at least 6 months historic data to delve through. (A full 12 months of data is even better when you’re in a seasonal industry.)

1 – Our 3 Tiered PLA Structure: To Bid More Aggressively on Specific Search Phrases to Maximize ROAS

This is by far our favorite way to achieve the best ROAS from a PLA campaign.

In a nutshell, the 3-tiered campaign structure allows you to focus your spend on the search phrases that drive your sales — something which isn’t as easy as it sounds.

There’s no point wasting spend on non-performers: you want to focus on Search Queries that drive the highest ROAS.

Below we’ve listed the 3 step approach we take when building out campaigns using the “Negative Keyword Waterfall” approach:

Step 1: Review Your Historic Search Volume to Find the Search Terms That Drive Your Revenue  

You need to find the search query themes that generate the most transactions for your company.  

The goal here is to create two groups; one that has the high converting search terms, and another with the low – medium converting search terms.

The best performing search queries group will likely contain several branded terms, with specific model names, SKU codes, part numbers, and other searches that show a high purchase intent.

Step 2: Begin with 3 Shopping Campaigns — Tiers 1, 2 & 3 with the Required Priority Levels

We will setup these 3 campaigns with a shared budget, and each with a different priority level.

(Note: When learning this strategy, remember that in this context, priority setting doesn’t reflect the group’s priority, it’s just the order in which Google will cycle-through the campaigns).

Tier 1 will have the highest campaign priority setting, which indicates to Google that search queries should start here.

This tier, like all the others, contains every product available on the site but with many negative search phrases applied (which we’ll come to shortly).

In tier 2 (or campaign 2), we’ll adjust the priority setting to medium and this is where the average to medium performing search terms live.

In tier 3 (or campaign 3), we’ll adjust the priority setting to “low” and this is where the best converting search terms exist.

There are 3 tiers: tier one (high priority, low bid), tier 2 (medium priority, medium bid), and tier 3 (low priority, high bid).

Step 3: Build and Apply the Negative Keyword Lists

In tier 1, we will be applying negative keywords based on the search queries we want active in tier 2 and 3.

By adding these negative keywords to tier 1, it prevents them from landing in tier 1 and pushes them to the next tier in the funnel — Tier 2.

At tier 2, we again add negative keywords but this time only the best performing search terms.

At tier 3, we don’t need any negative keywords applied as any of those lower value search phrases should have been filtered out already from tier 1 or 2. This was the ultimate goal —  to be able to exclude those lesser converting searches and to bid more aggressively on all searches making it to tier 3.

(Note: We’ve seen some agencies get a little zealous with the amount of tiers they create and it becomes very difficult to maintain — the smallest change in any campaign can completely wreck the entire system. For that reason, we typically keep it to a 3 tier setup).

2 – Identify Your Store’s Best Sellers So You Know Where to Prioritize Your Budget

One of the easiest ways to grow your ROAS is to do a deep audit of your Google Ad campaign to identify historic best sellers — then bid higher accordingly.

You can combine this strategy with the tiered approach outlined above in strategy #1, where you bid your best sellers up on the product or product group level.

Take a look at your historic data (if it’s available) and identify those top selling products.

You can also use the likes of Google Analytics to find best sellers and eCommerce conversion rates, plus the relevant ROAS/ROI metrics.

3 – Regularly Audit and Optimize Your Google Product Feed for Better Overall Performance

If you take away just one thing from this piece, it should be this: your Google Product Feed is essential for succeeding with Google Shopping ads.

Auditing your feed is one of the first things we do when working with a client. This is a vital part of the campaign as it’s a vital cog in the Shopping ads algorithm, so it deserves a lot of attention.

We recently wrote a more detailed piece on how to optimize your Google Product Feed that explains exactly why and how you should be optimizing your feed and covers how to setup and execute your campaigns.

In short, you need to ensure your feed contains all the required product information. If not, you risk a) not showing up when people are searching for your products, and b) being charged a higher CPC to show your ads.

Be sure to keep your product title names relevant without keyword-stuffing.

It’s also vital to keep product titles relevant without keyword-stuffing. This not only helps to enhance visibility for those high intent searches, but it also helps to boost CTR from the ads.

4 – Our Mobile-Optimized Strategy for Improved eCommerce ROAS

Our own research has confirmed that mobile shoppers behave differently than desktop shoppers — no surprises there. The actual queries that convert on mobile aren’t necessarily the same ones people use from desktop.

But many eCommerce teams don’t have a specific approach when it comes to mobile users, aside from reducing mobile bids — which can be a wasteful approach.

eCommerce Ads Strategy: Segment mobile customers

We repeat the search query analysis as mentioned in strategy #1 to segment mobile customers so we can determine a historic mobile-only ROAS.

If we find a gap here, we setup our own tiers with the appropriate negative keywords for mobile users. In the end, we may have 6 tiers setup for a client; 3 for desktop and 3 for mobile.

This beats simply adjusting bids on mobile or desktop. It’s a more holistic and strategic approach to optimising for the customer’s device at that moment, and for the entirety of the customer’s journey.

We find it pays to go into much depth with this.

5 – Seasonality Is Often an Untapped Opportunity (Find Those Crucial Periods for Your Client & Bid Aggressively)

If our client’s in a very seasonal industry, it’s crucial to keep time of year in mind and bid on products/product groups within the tiered structure accordingly.

For example, you should bid higher on flip flops in the summer and snow boots in the winter—but the tiers will remain the same.

By bidding higher on the best converting products in summer, you can maximize ROAS during these peak months when there’s a bigger search demand.

And during those quieter periods in winter, we’ll reduce bids but ensure we’ve still got a good presence — so we often find ourselves bidding differently depending on the time of year.

Take one of our wholesale retailers as an example. They operate in the back to school vertical, so understandably they get peak traffic ahead of the new school year. Search behavior changes in the lead up to these months.

We want to ensure we’re visible during periods of high search activity, whilst ensuring budget for the year isn’t exhausted.

We’ve seen this work across a range of seasonal verticals, and it isn’t something that we’ve noticed many other agencies or in-house teams doing.

(Note: It’s important to fully understand the 3 tiered approach before diving into this strategy or the mobile specific one. The inventory listed will need to be the same across tiers, otherwise you may experience leakage!)

6 – Let Google Optimize (With Supervision!) When There’s No Obvious Tiered Structure or Search Query Tiers

Smart Shopping Campaigns utilize a mix of your product feed and Google’s machine learning to take care of campaigns on your behalf.

We like to bundle any products into a Smart Shopping campaign when they don’t necessarily belong to another tier as explained in strategy #1 above.

These are the smaller products, maybe those lower-priced accessories, which can be left for Google to deal with. Google will then optimize for best-fit based on transaction history: but it doesn’t always mean Google will do the best possible job.

We’ve witnessed occasions where there wasn’t much transaction data for Google to use. Then when we saw a couple of mobile transactions occur Google went very aggressive on mobile — and caused ROAS to plummet.

In this case it would’ve been best to wait until there was more significant transaction data available before leaving it in the hands of them to automate.

7 – Using Dynamic Retargeting to Boost Your PLA’s eCommerce ROAS

As annoying as some people might find them, retargeting ads do convert, and extremely well.

Over time they’ve been shown to have a great ROI so they really work well to supplement your search marketing strategy.

The dynamic retargeting feature from Google Shopping enables you to automatically show ads to people who came to your site without completing a purchase.

It makes use of your product feed to determine the products they display and can intelligently group these together based on what’s likely to convert best.

Using dynamic remarketing is a fairly straightforward strategy to skyrocket eCommerce performance — it’s a must for any online retailer.

eCommerce Ads Strategy: Setup dynamic retargeting ads to close the deal
An example of a dynamic retargeting ad

The Best eCommerce Strategies for Google Search Ads

We’ve covered 7 Google Shopping Campaign strategies above in quite some depth.

But whilst it’s easy to forget, running Google Search ads in tandem with Shopping Ads is a good strategy to cover all your bases.

Here are 9 more essential strategies we use to maximize eCommerce brands ROAS from search ads:

8 – How to Best Structure Your Google Ads Account for More Granular Control (And a Better ROAS!)

Often the best way to setup your client’s account is to actually just mimic their own navigation menu.

They’ve gone through the effort to build it out the way they have — so it’s probably been made that way for good reason.

eCommerce Ads Strategy: Structure your Google ads account in a similar way to how they setup their own website navigation.

If they’ve got a top-level page that contains a category of products (eg shoes) and then sub-categories that contain brands (adidas, nike) then it probably makes sense to have a shoe category, and individual brand-specific ad groups within your Google Ads account.

Setting up in the way above saves a bit of time when it comes to structuring the account, and will make budget control easy too.

With this approach you can also get as granular as you like when it comes to ad group and keyword grouping.

It will also help when other people on your team need to dive in to manage the account, as well as keeping things clean for the reporting team later.

9 – Deep Link to Best-Sellers Within Text Ads & Make It Easier for Your Customers to Checkout

Often within your store’s categories, there will be a handful of outstanding, top selling products.

Instead of directing customers to an individual category page it will often make sense to take them direct to the best selling product page instead — that’s usually where they end up anyway.

You could easily have a few text ads setup on rotate which deep-link to a handful of the top-selling products, and simply monitor which ads bring in the most conversions.

You can run A/B tests on this in the background and keep a close eye on the products that really push the needle on your ROI.

This makes the path-to-purchase cleaner for the customer, and helps to improve your Google Quality Score, too.

In this instance, the keyword/search intent, ad text, and landing page experience is all well aligned and optimized.

In the cases where there is no clear best-seller, it would make good sense to direct the customer to the most relevant category page instead. This approach is often used when bidding on the less specific, short-tail keywords.

10 – Have an Industry-Specific (But Agile!) eCommerce Approach and Test Everything

The strategy we use at Inflow for running Google Search Ads will ultimately depend on the industry the client operates in, and of course the eCommerce platform they use.

We’re forced to tailor our approach to suit our clients (and more pertinent: their shoppers) so when it comes to strategy we’re always flexible to changing our tactics to suit what works best.

It’s important to have an agile approach when it comes to eCommerce marketing since things change quickly and the search landscape is constantly evolving. You need to always be open to new opportunities and test everything!

We like to use Google Experiments within Google Ads to test how variations of campaign setups perform versus our original campaign, helping to shape our ongoing strategies.

11 – Don’t Neglect Any Google Ad Extensions, Especially Price Extensions

We always ensure that every possible extension has been built out when a campaign goes live. Setting up all eligible extensions will give you a better Quality Score on your account and enhances your chances of taking up more valuable real estate within the SERPs.

The obvious choice when it comes to eCommerce clients has to be the Price Extension. This will highlight the product price within the text ad when someone’s shopping for your product.

eCommerce Ads Strategy: Be sure to utilize the 'price' extension because it adds more value to your site.

Ensure your account has the following extensions active and optimised:

  • Callout extensions
  • Structured Snippets
  • Promotion Snippets (essential for Black Friday and other sales)
  • Sitelink Extensions

12 – Product SKU’s, Part-Numbers, and Model Numbers Are a PPC Specialist’s Dream

When running search ads, you’ll want to bid heavily on product SKU and other identifier codes, model numbers, replacement part keywords, and so on.

Whilst these might not have a huge search volume when compared to some other non-brand search queries, they’re going to have a super high conversion rate.

Someone entering a search query for “washing machine” or even “best washing machine” is probably going to be fairly high in the purchase process. They’re probably still shopping around and trying to come to a decision about the particular model they want.

But what about someone searching for a specific washing machine model, like “Samsung WW70K5413UX”? You should be throwing your money at Google for that type of search query.

We often scrape our clients product feeds to get a list of these numbers or SKU’s before using Dynamic Insertion within the Headline of the text ad and the Display URL to help boost CTR, as well as using keyword level Final URLs to send the user to the exact product they are searching for.  

13 – Ongoing Scheduled Maintenance and Optimization Is a Must for All Google Ad Campaigns

Let’s say your ROAS is ticking over nicely at 300% each month. While that’s great, it’s not to say it’s bringing in the most possible revenue.

You shouldn’t neglect campaigns even if they’ve got a great ROAS (that 300% could all be based on a few branded search terms and nothing else that’s going to drive your sales).

Ongoing scheduled maintenance and optimization is vital to ensuring your search strategy doesn’t get left to stagnate.

From regularly reviewing the ‘Search Insights’ report and checking in on the ‘Search Impression Share’, while verifying that rogue searches aren’t eating up your budget — there’s always plenty to do.

eCommerce Ads Strategy: Regularly review your reports for accurate information on how your ads are running.

This goes back to strategies #4 and #5: mobile search behavior is different than desktop search, and seasonality is an important factor to consider too.

Checking how those two variants might affect your search campaigns on an ongoing basis wouldn’t go amiss — especially if you’ve got an old account that has gone a bit stale.

14 – In-Market Audiences Offers Keyword Flexibility (And More Aggressive Bidding Strategies)

In-market audiences can be used within your Search Campaigns to ensure your ads are being seen by a wider audience than normal with a different matching criteria applied.

The in-market audiences to choose from can be viewed from this downloadable Google resource and consists of groups of people who are supposedly in the market for a particular product or service.

Let’s take an example of an online retailer of car wheels and accessories. If somebody searches for a product that is similar to one that the car retailer sells, Google will place them in a particular audience group.

Be sure to target a specific audience with your ads.

As an advertiser, you can then choose to target that particular audience group with your own search ads.

It sometimes makes sense to adjust bids according to your audiences, upping them when they match a particular, high-intent group.

It can also give you flexibility when it comes to your keyword strategy — you don’t need to be quite as granular, because you know this person is (in theory) already interested in what you’re selling.

15 – Dynamic Search Ads Suit Clients with a Huge Product Inventory or No Accessible Product Feed

Dynamic Search Ads are perfect for those who have a lot of inventory but not a lot of time to list everything.

The penultimate Google Ads type we’ll share is our strategy for is Dynamic Search Ads.

This is Google’s offering for those with a massive inventory of products to sell, perhaps without the time required to list individual ads for each product.

Using this method, Google’s ad crawler will scan your entire website and create ads automatically based upon what it finds.

That might be a bit too much control for some, but for others it can be a real time-saver.

For us, we love to use it as a tool to create keyword lists in the unlikely case we may have missed some good search terms during our own keyword discovery phase.

(Note: If you do opt to make use of a Dynamic Search Campaign, you’ll want to ensure you’ve added your normal search keywords as negatives to ensure there’s no overlap.)

Ultimately, Dynamic Ads can be a good low-budget and minimal fuss campaign set to run in the background with low ongoing maintenance required.

16 – Single Keyword Ad Groups (SKAG’s) Can Reduce CPC and Grow ROAS

Last but not least, Single Keyword Ad Groups (SKAG’s) can be a really simple but incredibly effective way to push out the ROAS.

Using historic conversion data, you can quite quickly identify the few keywords which are bringing in most of your revenue (see the 80/20 rule).

These keywords should be listed in their own individual ad groups (1 keyword per group often gives the best results here) with exact match targeting enabled.

Ad texts are then optimised to match that specific keyword, which helps give a great quality score to Google, and hopefully generates a good CTR too.

In your other Ad groups, you’ll want to ensure exact match negative keywords are applied for the terms above to ensure there’s no overlap which may lead to bidding against yourself in the Ads Auction.

Often by making use of the SKAG strategy, you will see a reduction in CPC, and over time you can push more and more spend into the campaign; you’ll want to cover as much of the search demand as you can.

As with all of the strategies listed here, you’ll want to play around and test the above before you jump right in and make any drastic, lasting changes.

Meet Rover: The chatbot that scaled CRO through automation

As the scope of CRO projects increases, so does the difficulty of management and the tediousness of the manual efforts required. To overcome these pain points, digital marketing agency Oogst – a Merkle company collaborated with HEMA, a leading Dutch retail chain and created a chatbot by bringing together the power of the VWO REST API and Slack, to scale CRO efforts at HEMA.

The post Meet Rover: The chatbot that scaled CRO through automation appeared first on Blog.

Continuous experimentation and testing increases conversions at high growth and data-driven organizations. Scaling a conversion rate optimization (CRO) program requires not only building a long-term testing roadmap but also building a culture of experimentation involving multiple teams and stakeholders across the company. As the scope of CRO projects increases, so does the difficulty of management and the tediousness of the manual efforts required. To overcome these pain points,  digital marketing agency Oogst – a Merkle company collaborated with HEMA, a leading Dutch retail chain and created a chatbot  by bringing together the power of the VWO REST API and Slack, to scale CRO efforts at HEMA.

This blog post shall delve deeper into why and how we leveraged the power of VWO API and Slack to scale our CRO program and help in building a culture of experimentation.

The challenges of scaling CRO

For Oogst and HEMA, CRO involved  constant juggling of interests between its CRO specialists, analysts, clients, website developers and other relevant stakeholders. Each decision and action affects many people, which is why the process needs to go as smoothly as possible.

For example, whenever a test is activated or stopped, many  people need to be informed at the very moment it happens without leaving  anyone out. When you scale CRO program at your organization, informing everybody manually becomes a hassle, especially with new tests being turned on and off frequently. This process needed to be automated to bring efficiency and agility to the CRO initiatives.

The Birth of Rover

We discovered that the VWO REST API makes it easy to monitor tests for changes in status, and that we can use this information to make announcements easily. As a result of these discoveries, the idea for a chatbot was born. By linking the VWO API, Slack webhooks and a local database through Python, we were able to notify the right people about test status changes the moment they occurred. We named him Rover: he would be a dog/bot hybrid that barks at us when it needs our attention. Rover also sends notifications to relevant parties about when to check the preliminary results of a test. A huge burden was lifted from our shoulders, leaving us with more time to think about the next set of strategic experiments.

Growing the bot

As these simple adjustments to our progress proved to be quite helpful to us and HEMA, we began exploring the VWO API to see what else we could do. We came up with a number of areas where the bot could help us with:

Quality control

We run a high number of tests and it is of utmost importance to us and HEMA that everything goes according to plan at all times. We have set up quality control checks internally to limit the risks. There are processes we follow before publishing a test to make sure it’s safe to go live and we double check right after publication in case anything was missed.

For example, some of the things we check for on the VWO side include if the traffic has been set to 100%, if GTM integration has been enabled, and whether the campaign’s name follows our naming conventions. Luckily, the VWO API made it possible for us to automate the checking of settings like this, so that we can be more certain a test is ready to be published, while taking us much less effort.

Monitoring live A/B tests

After publishing any test, we monitor events in Google Analytics to make sure we didn’t miss anything and the test is, in fact, running correctly. Although it allows us to maintain the level of quality we desire, this process is also very time consuming, tedious and prone to human error, much like the announcements.

To deal with this, we added the Google Analytics API into the mix to get Rover to check for the amount of VWO events for a particular test and notify us about its findings. The absence of VWO events likely meant the test wasn’t running (correctly), which is something we always had to check manually before.

Conversational A/B test management

However, at this point, Rover would only transmit messages, he could not process received messages. If we wanted a user to intuitively tell Rover to check a test, we’d have to include an element of interactivity.

That’s why we hooked up our bot to artificial intelligence using IBM Watson, which allows it to naturally process language. With this integration in place, a user is able to ask “Rover, is VWO test 244 ready to go live?” and it will perform the checks. Then, the user can simply publish the test by saying “Rover, publish 244”.

Right now, Rover has truly become part of the team, both of Oogst and HEMA. His contribution to the speed and quality of our process is well recognized. We included even more functionalities such as the ability to pause the manual activation tags we built in Google Tag Manager to trigger VWO tests, thereby linking VWO and GTM together.

But we’re not finished yet!

Scoring points: Building a CRO culture through gamification

Although the power and effectiveness of CRO and A/B testing is well established, building support for it throughout the organization still remains a difficulty.

To combat this, we found yet another role for Rover: quizmaster!

First, we pull basic information about the test and its variants from VWO. We then use this data to extract more detailed information about the experiment from our project management board. This includes background information, hypotheses, descriptions of the control and variants as well as screenshots.

From this knowledge, we dynamically generate polls on Slack where users vote on the landing page variants they think will outperform the rest. These users, generally members of the client organization, then battle each other in a CRO tournament where it is determined who knows most about the website visitor.

The introduction of a gamification element has made CRO more exciting to those not generally involved in it, it has made our efforts more visible throughout the organization and also highlights the successes we have. Furthermore, it builds team spirit and generates more ideas for future testing. All because of our data-driven quizmaster: Rover!

To sum up:

Here’s a quick summary of the benefits we have already achieved:

·         Notify stakeholders of starting/stopping of tests;

·         Notify stakeholders to check preliminary results;

·         Check Google Analytics for presence of VWO events;

·         Perform quality control checks;

·         Pause Manual Activation tags in GTM;

·         Start/stop tests through chat;

·        Perform the role of quizmaster, allowing stakeholders to vote on which variant they expect to outperform the others.

More to come!

Future plans

We’re in touch with VWO to expand the abilities of the REST API even further to make Rover more powerful. This collaboration is a very exciting one to us. The API has already proven to be invaluable in our current operations which are now running more smoothly than ever before.

This bot is the collaborative effort of Gino Renardus, Martijn Heerema and Thom van Engelenburg (consultants at Oogst, a Merkle company) and Floor Hickmann and Raun Sips (UX at HEMA).

About Oogst

Oogst, A Merkle Company, is the leading digital marketing, analytics and optimization agency in the Netherlands. Based in Amsterdam, its team of over 70 experts provide leading brands with digital marketing services aimed at utilizing customer data in the best possible way to achieve the highest returns.

The company has partnered with VWO since 2012 to test and optimize some of  the most popular websites of the country for their clients. Their extensive knowledge of, and experience with the VWO platform has led to these operations continually growing in size. Moreover, this has also led to increased stakeholder involvement and more intensive test management. In order to better deal with this increased scope, the Oogst team was recently joined by a new member: a CRO Chatbot that operates through the VWO Application Programming Interface (API). This chatbot is able to assist in many of the activities surrounding the CRO process.
Do you want to meet Rover or get to know more? Get in touch with Oogst’s Data, Tech and Optimization team! Reach out via gino@oogstonline.nl.

About HEMA

Since 1926, HEMA has made the everyday life of its customers easier and more fun through products that positively stand out: due to their quality, design and price. HEMA offers over 30,000 of its own products and services, has over 750 stores in nine countries and 19,000 employees. As consumers move more towards online, HEMA recognizes the importance of developing a digital strategy as progressive as the brand is.

HEMA’s webshop is widely recognized to be one of the best of The Netherlands, with a top position in the Twinkle 100 and by frequently receiving awards such as Best Department Store Webshop. HEMA and Oogst collaborate on CRO to ensure its Dutch and international webshops remain among the best.

The post Meet Rover: The chatbot that scaled CRO through automation appeared first on Blog.

Google Shopping Product Feed Optimization: How to Optimize Your Data Feed for Better Ad Performance

Our guide to optimizing your eCommerce product or data feed for Google Shopping ads to get the lowest costs per click and highest ROAS. We’ve curated these lessons from working on hundreds of eCommerce shopping ad campaigns.

When most people think about optimizing their Google Shopping ads, they focus on optimizing their campaigns, bid strategy, and execution.

While this is important, we’d argue that the more significant wins come from optimizing your data feed.

Your data feed is a subset of all of your products. It’s a digital product catalog that contains attributes like product title, description, GTIN, category, etc. This is what’s used to determine how often and for what keywords you show up for.

Google Shopping consists of two products – Google Ads (formerly AdWords) and Google Merchant Center. Google Merchant Center is where your data feed lives and Google Ads is where you can optimize your campaigns, set your budget and adjust your bid strategy.

We’ve divided this guide into two parts:

Part 1: Data Feed Optimization: Getting your data feed optimized is the foundation of successful Google Shopping Ad campaigns for eCommerce companies. Most guides on this topic don’t cover this in-depth but we’ve dedicated an entire part to this step. We’ll outline our experience in how to properly manage Google Shopping ads for hundreds of eCommerce companies.

Part 2: Campaign Setup and Execution: Here we’ll teach you the nuts and bolts of setting up, optimizing, and executing your Google Shopping ads campaigns. (Note, this will be published soon, and then we’ll link to it here.)

In this post, we’ll walk you through these four steps for how to optimize your data feed:

  1. Optimize These Five Attributes (At a Minimum)  
  2. Monitor Google Merchant Center Regularly
  3. Update Your Feed Regularly
  4. Analyze and Cross-Reference Ad Performance Data

Why Does Data Feed Optimization Matter?

The more information that you have in your data feed, the more relevant you are for people’s searches. In turn, Google will reward you by showing your ads more frequently and for less money.

Optimizing your data feed is like building a house. The data feed is the foundation on where you build all of your Google Shopping campaigns. If the foundation is off or your data is bad,  you will have a hard time performing well on Google Shopping.

Here are a few things that happen when your feed isn’t optimized.

  • You won’t show up for certain searches.
  • It will cost you more money for your ads to show
  • Your competitors will take up more space than you.
  • You won’t get correct visibility which could lead to overspend.

For example, say you sell Nike shoes. If your titles only contained basic information such as, “Nike red shoes” and didn’t contain size information, color information, or model information in your product title, then you would not show up for some of the highest intent searches.

If you did end up showing up, you would be spending a lot more for that ad than competitors who implement a feed optimization strategy.

Instead of "Nike red shoes", being more specific is better in the long-run.

Optimize These Five Most Important Attributes


The GTIN is the UPC of the online world. When you’re a reseller or a manufacturer of goods, you use that information to track your products across Google’s Surfaces and to keep an eye on MAP pricing through the Manufacturer Center.

The only time that you don’t need a GTIN is if you’re selling one-off goods—meaning you sell one-off antique furniture, used products, or when you’re selling things as a bundle. When you have the GTIN, it’s important to include it because of all the information it provides.

Product Titles

One of the biggest mistakes that we see brands making is keyword stuffing their titles—and it ends up backfiring. Google will penalize you for long titles (i.e., anything over 75 characters). If your title is over 150 characters, they won’t approve your ad.

Product titles are important because Google Shopping is all broad match. There’s no phrase or exact match. The algorithm uses the title to identify what the person is actually searching for and matches those keywords with the feed.

The key is to be precise and include the most relevant information in your titles. However, it is important not to keyword stuff your titles. Your title should reflect how your customers or prospects are already talking about your product.  

Data feed optimization: Instead of "men's jeans" you're better off getting more specific like "Men's Levi’s Boss Jeans Size 34"

For example, if you sell jeans, you need to include all of the main keywords in the titles, such as: ‘Men’s Levi’s Boss Jeans Size 34’. The more relevant information that the user includes in their searches, the more likely they are to buy.

If your product title is only “men’s jeans” you’ll appear on all of the general searches, but you’ll miss out on the higher-intent searches that are the most likely to result in a purchase.

This can happen when you directly import the titles from your Shopify store directly into Google Shopping without optimizing them.  

If you are having issues optimizing your feed or just want a second pair of expert eyes on it, you can reach out to us and one of our Google Shopping experts will help you out.

Product Type

In our experience, Google is increasingly putting more relevance to the product types. They contribute to the overall “quality score” of your data feed.

We recommend you follow the structure/breadcrumbs on your website. This is the best way to get proper visibility over all your products.

Product Category

The product category is how Google categorizes products inside Google Shopping itself.

Because there are over 6,000 default categories, we recommend you get as granular as possible. Some of these categories can go down five tiers or more. You will want to use a categorization tool to help you with this. Here is a full list of categories to help with this process.

For example, if you sell reading glasses and you haven’t optimized your product categories, your products might be showing up under drinking glasses instead of reading glasses.

Product Description

The description carries less weight than the other four attributes above because it takes about two clicks to see (as opposed to the title, price, and brand, that show up immediately in the shopping results—without any additional clicks from the user). However, this is still worth optimizing because of its use of added relevancy.

Here are a few tips for optimizing your description:

  • Put your product title at the end of your description.
  • Add high relevance keywords into your description
  • Identify added HTML that was not removed. There are many times that these pieces of code will get caught in the description.

Understanding Google Merchant Center

With text ads, you can see a quality score, such as an 8 out of 10. However, with Google Shopping campaigns, you don’t get a numerical score. Instead, you have the Google Merchant Center.

The Merchant Center acts as a QA source to make sure your data is clean and accurate. When you upload your data feed into Google Merchant Center, you’ll see errors, warnings, and disapprovals.

While Google will still show your ads for certain searches, it will end up costing you more money than your competitors who have optimized feeds. So, when you start seeing these errors come in, you want to clean them up ASAP.

There are three different levels of issues in the Merchant Center—by account item, feed, and account level (with account level being the worst).

Account Level Warnings

Account level warnings can lead to larger issues such as Google shutting down your Merchant Center. If you get an Account Level Warning, such as a verification issue, you should fix it immediately so that you can keep your ads running.

Feed Level Warnings

There’s also a feed check, where they’ll go through your data feed and make sure it’s delimiting correctly, and all the files are set up and readable.

Item Level Warnings

SKU or item level warnings are at the product level, meaning that they’ll go through all of your products. They’ll let you know if you are not including specific attributes, and give you a warning or a recommendation. This is where most issues are found and will need to be fixed in the data feed.

Update Your Data Feed Regularly

To improve your campaigns, get in the habit of updating your feed anytime your product prices, availability, or inventory levels change.

For example, if you sold 500 products last month and now you are only selling 350, you need to update your data feed. Otherwise, you’re paying to market 150 products that you aren’t currently selling. Worse, they’ll 404 in the Merchant Center and bring down the quality of your feed.

In addition, when working with our clients, we ask for a dynamic file meaning that we pull in your new file of all your products every single day. We optimize those products through a custom tool and then upload it in Google Merchant Center.

This should be done by someone who has feed optimization experience because many times when you try to do this on your own (or use one of the $50/per month automated app optimization platforms), you won’t have an analyst looking over the feed. This results in random keywords being added into your titles with inaccurate attributes. You’ll end up with a feed that isn’t fully optimized, and it could cost you more money in the long run.

At Inflow, we do all of these optimizations in-house, which means that we don’t make cookie cutter optimizations like most of the tools out there.

Analyze Ad Performance Data

You should also cross-reference Google Shopping data with search volume and other ad data sources. Some key things to look at are conversions, search volume, and negative keywords for your Shopping ads.

This is a great place to start looking for additional optimizations to add back into the feed. By analyzing the feed and campaigns from this level, it should help you steadily increase performance over a longer period of time.


Optimizing your data feed is a powerful way to improve Google Shopping ads performance, lower cost per click, and generate more sales.

We’ll be sharing more details on Campaign Setup and Execution in Part 2 of our guide—stay tuned!  

Death by CRO: 4 Common & Deadly CRO traps to avoid (includes free survival tips)

Experimentation has been the cornerstone of all successful organizations. But many fall prey to the obvious yet deadly traps when running or building a CRO programme in their organizations. Find out how you can avoid these traps in this blog.

The post Death by CRO: 4 Common & Deadly CRO traps to avoid (includes free survival tips) appeared first on Blog.

Experimentation has always been the driving force when challenging the status quo – whether it’s the battlefield where a change in the strategy has altered the course of history or a product change which separates successful products from thousands of failures. For online businesses, this has translated into improving customer experiences thus leading to increase in conversions at the lowest risk possible. Rome wasn’t built in a day and your CRO programme too will not be.

So before you get going to prepare for the ultimate showdown using CRO in your organization; let me help you navigate the various challenges you might face during this journey.

  1. Don’t break the news, already!

Case 1: Omg! Just 3 days and my variation is performing way better than the control. Let me ring my CEO and tell her that I was right about this change.

If this is you, congratulations you just got killed by CRO. One of the most important ground rules for testing is to be patient. Initial test results might excite you to go out there and proclaim victory but wait for the test to conclude to clearly state it. Setting high expectations for the success of an experiment after seeing initial traction may do more harm than good.

Expectation setting may not directly be linked to website optimization but trust me when the results don’t come out as well as expected (thanks to your initial excitement), you wont get team buy-ins for bigger experiments.

Case 2: Damn! It’s been 5 days and there is no conclusion I can draw from this A/B Test. What will I tell my CMO if he asks me how the test is coming along?

Initial test result might put you on the backfoot if you see no or minute movement in your conversion graph to justify the CRO efforts. The answer to all worries is patience. Big changes or small, it takes some time for your results to reach statistical significance given a variety of factors such as the number of visitors being tested, number of variations, etc.

To help you not get excited or demotivated before time, we have built a calculator to help you determine the duration for your A/B tests here.

2. ‘This isn’t working’ syndrome

5 tests. But no major change in conversions. But company X whose case study I read did 2x better in conversions. What am I missing?

Let’s assume you ran an on-page survey for an ecommerce site and figured that people who were not completing the purchase were skeptical about the security of the checkout page (even though it may actually be safe). This stopped them from putting in their card details and abandoning their carts. As an obvious next step you form a hypothesis backed by solid data and create a test variation with more security certification badges, testimonials, etc. The result- no difference at all!

So, does that mean you crafted a wrong hypothesis?

The answer is maybe. But take a step back and think about how many ways can you improve security perception of your checkout page? Or make people trust your payment processes?

Answer: More than we care to count.

And this is true for your first successful test which may have got you a 10% lift as well. You still have to think of ways of improving that number. There are better alternatives out there. You just need to keep testing.

Your optimization army should be inspired by the losses to dig deeper and find richer insights to create that one victory which will change the course of your business. Ask yourself how many iterations did you try before arriving at the conclusion that ‘it isn’t working’!

3. Monkey see, monkey do

When our neighbors fought their first conversion battle, they just changed their website CTA color to green and camouflaged their way to get better conversion rates! Let us paint our own checkout CTA green!

They say ‘Imitation is the sincerest form of flattery’, but not when it comes to CRO.  ‘Best practices’ may not be the best for you. Hard data and ground reality on your website may be completely poles apart than the case study you read about. Do not expect similar results from the experiments run by others in the arena. Use quantitative and qualitative research methods to devise a unique hypothesis and then launch your test. Also choose the right weapons (A/B testing or multivariate testing) and a structured CRO plan to execute it.

All businesses are different, so are their visitors’ behavior and thus their experiments. What worked for one may or may not work for others, the idea is to always keep testing until you succeed.

4. The show must go on!

CRO should go on. I will make sure that when I retire or lose a limb in the battle for conversion throne, my army is ready to fight without me.

At VWO, we come across customers who suddenly stop testing and the main reason they cite is that the person who was carrying the CRO baton has quit. Find it shocking? Even we do!

We need to understand that CRO is not just a one-person or even a one-team job. Building an organization which thrives on CRO requires not just education and training, it requires a change in the cultural fabric of the company. A CRO-friendly culture requires the HiPPOs to take a backseat and invite the soldiers from different teams (product, marketing, design and so on) to draw up a battle plan. Don’t take anyone’s word on the face of it but test everything! Celebrate successes and publicize results to get a team-wide buy in for experimentation. It is an uphill battle and hence requires you to plan ahead and properly. Remember Rome?

Find some excellent tips to build a culture of experimentation in your organization here and build a CRO army to continue the battle for conversion even if someone calls it a day.

Parting words

While you wear your shining armor of a CRO catalyst, believe in yourself and don’t let ‘Death by CRO’ scare you because with the right attitude, you are going to win it not just for yourself but for future teams within your organization. Don’t take my word for it (you might, I have seen 5000+ customers across 90 countries succeed) but test it!

PS: I hoped to save some lives with this blog. Tell me if you survived in the comments section.

The post Death by CRO: 4 Common & Deadly CRO traps to avoid (includes free survival tips) appeared first on Blog.

Truckstop.com: On Building A Culture of Experimentation

Truckstop.com has been on a formalized CRO mission for a little over a year and has seen some motivating results. The blog talks about the pillars on which Truckstop.com’s experimentation culture stands today.

The post Truckstop.com: On Building A Culture of Experimentation appeared first on Blog.

While there are a few stalwarts who stand out when it comes to reliance on experimentation to drive growth, honed over many years, some brands are formalizing this process to further root CRO into their DNA, knowing how critical it is in current times. Truckstop.com has been in business for the last 23 years and has been on a formalized CRO mission for a little over a year and has seen some motivating results. To continue reaping the benefits of an ongoing CRO process, here are some of the pillars on which Truckstop.com’s experimentation culture stands today.

About Truckstop.com

Truckstop.com was founded in 1995, when founder Scott Moscrip began offering a better way for truck drivers to find loads than by posting signs on the side of their trucks or gathering around local bulletin boards.

Now, over two decades later, Scott’s digital vision has grown and Truckstop.com’s dedicated partners work to help truck drivers and their trusted freight partners grow their businesses. His guiding idea is still the same—always work to find a better way.

Truckstop.com has 3 distinct goals:

  • Direct online sales
  • Lead generation
  • Organic search volume through content promotion

Truckstop.com is eyeing a double-digit growth in 2019, for both leads and sales, and is testing and optimizing thought leadership content and content promotion.

Truckstop.com audiences and CRO

Truckstop.com has 2 sets of audiences:

  • The freight brokers who are tech-savvy and pressed for time. Truckstop.com needs to ensure that they can provide the information in a timely manner.
  • The carriers who prefer to interact via phone due to timing and preference. Truckstop.com needs to provide a platform, which can make it easy for them to sign up or contact a representative.

In both scenarios, Truckstop.com intends to optimize the platform to understand what minor and major changes affect the business’ bottom line.

Truckstop.com places CRO culture at the core of its growth decision-making strategy with  some of the following salient aspects :

  • Having a CRO Experimentation team and onboarding new members
  • Conducting cyclical fortnightly focus
  • Organically evolving process
  • Keeping personal preferences always at the backseat
  • Maintaining subjective communication with external stakeholders

Having a CRO Experimentation team and onboarding new members

Truckstop.com has a unique CRO program. They have people from content, design, project,  management, analytics, and more that form the core group called the Experimentation team. They have regular brainstorming sessions where everybody is heard.

Last year, they also got on board an email specialist. They have since started experimenting through emails. There is no formal training as such for people who join the CRO team; it is an organic process of learning and development. The core team guides them through and does a lot of brainstorming to get clarity on what the experiment is about, why they are doing it, what results does it aim to achieve, what the success metrics are, and other possible ideas? They work together to create the mindset to run an experiment.

Conducting cyclical fortnightly focus

The Experimentation team run experiments based on the current focus area which are broken down into segments like desktop website, mobile website, paid traffic, and more; and they meet weekly and talk about likely experiments based their observation repository. They focus on each segment for 2 weeks.

This way, they are able to run the experiments targeted within the stipulated 2 weeks. A lot depends on the size of the test as well. Implementation for smaller tests is easier so they sometimes manage multiple tests. However, for larger tests that need page design and content, it may take extra time. In most cases, the team is able to accomplish its fortnightly goals.

Organically Evolving Process

At Truckstop.com, it’s not regimental on when and how to conduct the tests. A general decision is based on the team’s understanding from the observation repository and analytics, which serve as the basis for the next action steps. Their test pipeline is also based on what their upcoming fortnightly focus is going to be. So their experimentation plan is quite fluid, though the fortnightly focus is largely known to them.

Kevin Gamache, Senior Web Analyst, Truckstop.com, shares,

“CRO is a continuous process for us,  we keep at it regularly. We don’t plan out that far in advance as we also have a backlog available of observations throughout 2018 which serve as the foundation for 2019. But, really it is much more organic. We test and see what are the results and then we think ‘Wow! Now let’s do something else and see how that responds’. So, it becomes an organic process that way rather than a methodical schedule.”

Keeping personal preferences always on the backseat  

An experiment that the team at Truckstop.com ran was for a pop-up lead form; everyone on their team hated pop-ups. Although most of the Experimentation team did not fancy the idea, they went ahead and conducted the test. The employees didn’t think it would work, rather hoped it wouldn’t, as the idea was too conventional and tactical for the modern digital age. However, contrary to everyone’s expectation, it worked and the results went through the roof. The pop-up form was then integrated as a part of the website, based on the test’s success.

Another interesting test worthy of mention here was when Truckstop.com aimed at decreasing the number of steps in the sales funnel. While the number of steps were reduced in the test, the end conversion rate, in fact, ended up being lower. The failure of this experiment proved to the team that pushing visitors early into the sales funnel wasn’t a good idea, which in turn gave them confidence on the status quo.  

These two were pivotal experiments as they helped the Experimentation team become more receptive towards ideas irrespective of the members’ personal preferences. They now let test results do the decision making.

Maintaining subjective communication with external stakeholders

Everyone loves and prefers the taste of success. Accepting a successful experiment is easy. However, when testing consistently, most tests either fail – that is, results witnessed are usually not what was hoped – or provide small incremental uplifts. With time, Truckstop.com has opened up about running multiple tests irrespective of the end outcome. While the Experimentation team is aware of all the tests conducted and their results, the larger organization is informed only of the big wins or when there are interesting learnings or validation. This focuses the communication on large changes and results rather than inundating them and having those communications be ignored or lost in the shuffle.

Tara Rowe, Program Manager, Truckstop.com, believes,

“Our experimentation process is about learning how we can make things better for the next time we repeat a similar process. Most of our communication is at the tactical level with our broader marketing team. However, there are those wins that are so extraordinary or that effect the entire business, those specific results are then shared to our executives.”

Truckstop.com now maintains the ideology that no test is a failed test – you either find a winner or validate the current website version.   

Closing Thoughts

Every organization has its own set of challenges. If you have just started out or have been there for a while and haven’t tried ingraining CRO in your growth, then you can take inspiration from Truckstop.com’s experimentation pillars to build and maintain a robust CRO foundation to drive growth. The single solution isn’t to have a CRO team – it is about accepting the fact that both failure and success go hand in hand in an experimentation process and finding what approach works for your organization.

The post Truckstop.com: On Building A Culture of Experimentation appeared first on Blog.

How Heatmaps Helped Increase Prospective Student Inquiries with Hotjar

Heatmaps are just the first step to obtaining useful insights on your website visitors. Today we’ll find out how heatmaps helped increase prospective student inquiries by 20% for a University and have a chat with Andrew Michael of Hotjar. Find ou…

Heatmaps are just the first step to obtaining useful insights on your website visitors. Today we’ll find out how heatmaps helped increase prospective student inquiries by 20% for a University and have a chat with Andrew Michael of Hotjar. Find out what he has to say. How Heatmaps Helped Increase Prospective Student Inquiries by 20% […]

The post How Heatmaps Helped Increase Prospective Student Inquiries with Hotjar appeared first on Conversion Sciences.

A/B Testing with a Small Sample Size

The question “How to test if my website has a small number of users” comes up frequently when I chat to people about statistics in A/B testing, online and offline alike. There are different opinions on the topic ranging from altering the si…

The question “How to test if my website has a small number of users” comes up frequently when I chat to people about statistics in A/B testing, online and offline alike. There are different opinions on the topic ranging from altering the significance threshold, statistical power or the minimum effect of interest all the way […] Read More...

A/B Testing Insights Through Ecommerce Customer Journey

A/B testing has evolved from a simple comparative study to a process backed up by a data-driven approach. Here are some some inside tips and tricks which can help you get huge uplifts in your business metrics.

The post A/B Testing Insights Through Ecommerce Customer Journey appeared first on Blog.

A seamless eCommerce customer journey is about transitioning from just “buying” to “experiencing.” The key to growing your online business is to let your customers explore, decide, and share, and then learn from their experience.

Over the years, A/B testing has evolved from a simple comparative study to a process backed up by a data-driven approach. Businesses and website owners can utilize this approach at every stage of their customer journey.

In this blog post, we take you through some inside tips and tricks which helped VWO customers get huge uplifts in business metrics such as conversions and revenues. Take notes as you read this blog, as the next big testing idea for your own ecommerce store could come in from here.

1) Easy Visibility and Access To Relevant Information for Visitors

Customer Interaction: Splash pages

A splash page is a screen that pops up when you first enter a site. As splash pages are the pages visitors see first when visiting a website, these are also an important source of revenue.

Let’s first look at some of the industry best practices regarding splash pages.

  • Make the key message and the exit button both easily visible.
    While interested visitors should be able to move through next steps, the disinterested ones should also be able to move through, that is, exit the splash page easily.
  • Make sure your users interact with your splash page first and then the page they are looking for.
  • Design the user/visitor flow such that those who have already visited your splash page don’t get to see it again.
    This should be spaced out based on factors such as time span, say on a revisit after 7 days.

Case Study

Warner Music Group, an American multinational entertainment and record label conglomerate headquartered in New York City, noticed that conversion rates were lower than the industry average.

They redesigned their splash pages, and ran 3 A/B tests before making these live.

In the variation, the image on the left was resized and moved below the content. The purpose was to provide a quick snapshot to the visitor in one go.

The redesigned splash pages showed a 4% improvement in the conversion rate. Look at the comparison of the control (left) and variation (right) below:

Optimizing The Customer Journey: WMG A/B Test

To learn more on how WMG went through with their other tests and execution, you can read the complete case study.

2) Best Possible Customer Experience on the Home Page

Visitor interaction: Home page

Your home page is the entrance to your store. Jazz it up with the very best of what you have got.

Some of the home page best practices include:

  • Convey value proposition straight up. What a visitor sees in the first few seconds on your home page is crucial. It’s important that you communicate your value proposition instantly, and don’t come across as just another mom & pop store.
  • Have an easily identifiable Sale section. Fish all the discounted products on your home page, and line those up in a separate section that’s dedicated to them.
  • Organize the navigation bar. Make sure your visitors easily find their way around your website. Place the navigation bar at the standard positions where visitors expect to find them. You can also make your navigation bar persistent.

Case Study

MedienReich ComputerTrainings, a German company which provides various software training courses, also applied the same learning and increased website engagement by 40%.

They replaced the 3 broad category types with 8 of their best-selling courses, such as Photoshop, InDesign, and AutoCAD.

While doing so, they also adhered to another best practice, that is:

Provide right information in right order.

You can refer to the comparison of the control and the variation below:

You can learn more about the MedienReich success story.

3) Connecting Prospective Customers with Products

Visitor interaction point: Category page

When it comes to category pages, every action requires equal caution. For such pages, you need to plan per your product type and page layout and design than following an industry best practice blindly. Be it how your products are displayed, the position of the search box, or the number of filters, every field or option should be used with discretion or should be tested and used accordingly.

For example, Buyakilt.com added filters to increase their revenue by 76%, but UKToolCentre considered filters as a distractor for a particular category and had these removed to increase their engagement by 27%.

When the focus of a website is to have customers sign in and then buy, it’s important to maintain customer focus and continuity.

Case Study

Muc-Off offers a broad range of cleaning products for bicycles, motorcycles, and other motor vehicles, and electronic goods. Its products are sold through major retailers around the world as well as through its website.

When Spot Studio, an agency hired by Muc-Off, also tried applying this learning, they observed that their design caused incongruity in user experience. Visitors came there to purchase products; but after landing on the page, they were finding information resources instead of a shop front.

The test they did after rearranging their departments page (or the category page) gave them 43.78% more product views.

Here’s the control-variation comparison, as noted on VWO Heatmaps.

To find more about what Muc-Off did to improve the customer experience on their category page, here’s the complete case study.

4) Judicious Use of Reviews and Testimonials on Product Pages to Gain Visitor Trust

Visitor interaction point: Product page

According to Local Consumer Review Survey 2018:

  • Consumers read an average of 10 online reviews before feeling able to trust a local business.
  • 57% of consumers will only use a business if it has 4 or more stars.
  • 89% of consumers read businesses’ responses to reviews.

Case Study

Express Watches, a client of VWO’s Certified Partners, High Position, is an authorized Seiko watch dealer that ships to over 23 countries. They have been in the watch retail industry for more than 21 years and believe in getting the basics right to satisfy their customers.

Considering the challenges related to replicas plaguing their industry, Express Watches also decided to use the above learning to reinforce their brand positioning.

However, to reduce visitor anxiety and achieve the desired results, they included a TrustPilot widget with some great customer reviews. These were tested against the erstwhile section Why Buy From Express Watches?

The variation was declared the winner with a percentage improvement of 58.39% and 99% chance to beat original, as shown below.

To read the Express Watches case study in detail, click here.

5) Resolution of Customer Friction Points at the Checkout Page for Significant Website Conversions

According to 40 Cart Abandonment Rate Statistics:

  • 26% of US online shoppers have abandoned an order in a particular quarter solely due to a “too long / complicated checkout process.”
  • $260 billion are recoverable through checkout optimizations of just US and EU e-commerce sales.
  • 34% of US online shoppers have abandoned an order in a particular quarter solely due to the reason “the site wanted me to create an account.”

As an eCommerce business owner, it’s important that you regularly interact with your customers and new visitors to get ongoing feedback about their experience, as you wouldn’t want to lose them at the final gateway, that is, the checkout stage.

Case Study

PayU India is the flagship company of Naspers group, a $93 billion Internet and media conglomerate. Through its proprietary technology, PayU provides state-of-the-art payment gateway solutions to online businesses.

The concern for PayU was checkout page drop-offs. Through VWO’s Visitor Recordings and Heatmaps features, the PayU team found that providing communication information was a key friction point. They hypothesized that dropping the less important of the two fields—the telephone and the email—would help them increase conversions.  

They tested the variation without the email field against the control. The results demonstrated that dropping the email field showed a statistically significant improvement of 5.8% compared to the control.

Look at the comparison of the control and variation below:



While the above A/B testing ideas worked for these companies, these may not necessarily work for you as well.

These case studies should provide you insights and ideas on how you can create strong hypotheses and test these further.

These examples should give you a good idea as to how A/B testing can be a good starting point for conversion optimization and the extent of results you can achieve through testing.

Do share your A/B testing experiences in the Comments section below.

The post A/B Testing Insights Through Ecommerce Customer Journey appeared first on Blog.

We Used Ahrefs to Analyze 25 Highly Competitive eCommerce Search Terms. Here’s What We Found

How can you rank your eCommerce product pages for competitive terms? What page elements help product pages rank well for trending and profitable keywords? Is building as many backlinks as possible really the key to ranking well? (Hint: nope) To help answer these questions, we identified 25 competitive eCommerce terms that have been trending upward

How can you rank your eCommerce product pages for competitive terms? What page elements help product pages rank well for trending and profitable keywords?

Is building as many backlinks as possible really the key to ranking well? (Hint: nope)

To help answer these questions, we identified 25 competitive eCommerce terms that have been trending upward in search volume in the past year (according to data from Google Trends and Shopify).

We used Ahrefs to analyze the top ten results for each keyword, then looked for trends among the results.

Some of our key findings:

  • Site authority and page authority helps product pages rank in the top 10—but they are not enough to get the top spots on their own.
  • For high authority sites, a few quality backlinks gave a strong boost to rankings.
  • Ranking in the first organic position was not a guarantee that a site would receive the most traffic.

Note: We’ve worked with dozens of eCommerce companies to increase conversions based on organic traffic. We can create a custom SEO strategy for your business. Contact us here.

It’s Difficult to Rank in the Top 10 with a Low Domain Rating

Ahrefs Domain Rating (DR) is a measure of the strength of a site’s backlink profile and overall authority.

DR is measured on a scale from 0-100. Sites with a DR over 90 tend to be household names like Amazon and eBay. Sites with a DR over 60 are often more niche sites—but ones that have a strong backlink profile.

Here’s the distribution of DR for every page that ranked in the top 10 organic results on Google from the results of our 25 keywords:

eCommerce product pages: Ahrefs Domain Rating of Sites Ranking in First 10 Positions on Google

Most sites that rank for these competitive keywords have thousands of backlinks.

If you don’t have that many backlinks, don’t worry. We’ll talk more about ranking product pages on sites with lower domain authority scores.

Site Authority Helps You Get into the Top 10—but Not to Advance into the Top Spots

How important is your site authority to ranking for the first position of Google? We looked at the median DR of sites by ranking to find out more:

eCommerce product pages: Median Ahrefs Domain Rating by Google Ranking

From the previous graph, we know that having site authority is important to ranking in the top 10. This graph shows that once you’re in the top ten, more authority won’t necessarily push you to the top.

If there was a strong correlation between ranking in the top of the first ten results and overall authority, you would expect the median Domain Rating of the top spots to be better than the lower spots.

Page Authority Matters to a Certain Extent

Ahrefs has another metric to estimate the ranking power of a single page called URL Rating (UR). You can think of it as an estimation of how Google rates the page authority of a single URL.

Like Domain Rating, URL Rating is also based on a 0-100 scale. Here’s the median UR at each position for our eCommerce keywords list:

eCommerce product pages: Median Ahrefs URL Rating by Google Ranking

Like Domain Rating, increasing the authority of an individual page seems to have a point of diminishing returns.

That doesn’t mean that aiming for a higher UR is a bad choice. It may very well contribute to you outranking some of the bigger sites like Amazon. Consider the current rankings for “yoga mats”:

A sample of what's displayed in a google search for "yoga mats". Proving that some outrank bigger sites like Amazon when they have a higher UR.

The top three results are product pages from smaller eCommerce sites that are outranking the product page from Amazon.

The smaller sites’ product pages all have a better backlink profile and UR than the product page. That doesn’t mean that page authority is the only reason they outrank Amazon, but it definitely helps.

One of your advantages against a big company like Amazon is that you’re willing to spend more time ranking a particular product page. Amazon isn’t going to do as much link building or optimizing for a single product page.

While site authority and page authority do seem to be necessary to rank for these competitive terms, Ahrefs’ DR and UR are not perfect estimates of how Google decides how to rank pages.

Consider this screenshot of the current rankings for “tank tops.” The second-to-last result has a lower DR and UR than the last result:

A sample of what's displayed in a google search for "tank tops". Proving that the second-to-last result has a lower DR and UR than the last result.

DR and UR are useful concepts, but take them with a grain of salt. Better scores in a tool like Ahrefs does not always mean you’ll get better rankings in Google.

The Quality of Backlinks Seems to Matter as Much (Or More) as the Overall Number of Backlinks

If you want to rank a product page, how many backlinks do you need to build? We analyzed the median number of referring domains by position to find out.

eCommerce product pages: Median Number of Referring Domains by Google Ranking

We can see a slight correlation between the number of referring domains to a page and its Google ranking.

At first glance, this seems to contradict the previous graph in which UR stopped mattering after a certain point. After all, UR is a measure of the backlink profile of a single URL.

How we reconcile this difference is to remember that Ahrefs UR takes into account the quality of a backlink. Our guess is that sites that rank in the higher position are more likely to have some low-quality backlinks. That means more referring domains but no meaningful change in UR.

All of this points to the importance of just building high-quality backlinks. We saw in our data clear examples of product pages ranking without a huge number of backlinks.

Here’s a screenshot for what comes up first in Google for “yoga leggings.”

A sample of what's displayed in a google search for "yoga leggings". Proving that product pages rank without a huge number of backlinks.

Yoga leggings a relatively competitive term with clear buyer intent. Yet, the first domain ranking only has one backlink (we checked, and it is from an authoritative site). The next two results have many more backlinks, but that hasn’t earned them a higher ranking. Take note that all have a DR above 50.

The rest of the pages ranking for “yoga leggings” have a UR between 11 and 20 and a DR above 40. This reinforces our earlier conclusions that building backlinks to a site and a page only work to a certain point.

Ranking First Doesn’t Guarantee the Most Search Traffic

Conventional wisdom says that the higher you rank, the more traffic you will get. This is almost certainly true for a single search as people mostly click the top results. But, pages that rank for competitive keywords also tend to rank for other long-tail phrases.

This means that it’s possible that a page that doesn’t rank at the top for a competitive keyword still gets more overall search traffic than the top ranking site.

For example, look at the top 3 results for “heat vests”:

A sample of what's displayed in a google search for "heat vest". Proving that it's possible that even if you aren't ranked at the top, you can still get quality search traffic.

The third result receives more overall search traffic than the first result. To be clear, this is not a measurement of traffic from this search alone. Instead, it’s a measurement of all search traffic when looking at all the keywords this page ranks for.

To see how often a page that doesn’t rank first for a head term gets more overall search traffic than the first ranking page, we looked at the position with the most traffic for each of our 25 eCommerce keywords:

eCommerce product pages: Position with the Most Traffic by Keyword

In only 10 out of 25 keywords did the top ranking site have the most SEO traffic overall. In the other 15 cases, a site that didn’t rank in the first position received the most SEO traffic.

So how important is it to rank at the top of a competitive keyword? We looked at how overall traffic is distributed amongst the sites ranking in the top 10:

eCommerce product pages: Median Overall Search Traffic by Google Ranking

The product pages we analyzed that ranked first did receive the most overall search traffic, but it’s not as disproportionate as is often portrayed.

You may have read that long-tail keywords are responsible for 70% of all SEO traffic. This graph shows why focusing on the long-tail is a much easier path to getting SEO traffic.

Instead of relying on just one search which fluctuates based on the ever-changing Google algorithms, ranking for hundreds or thousands of long-tail keywords can make your product page more resilient against Google’s changes.

Small Relationship Between Ranking and Total Keywords

We also investigated whether a page that ranks higher will rank for have more keywords:

eCommerce product pages: Median Number of Total Keywords by Google Ranking

The minor correlation between position and keywords reinforces how nebulous the long tail is. Still, there are general principles you can follow to maximize long tail rankings (as we’ve described here).

In Closing…

  • Increasing your site and page authority with a robust backlink profile can give you the chance to rank for keywords that you otherwise would have almost no chance of ranking for.
  • A handful of quality backlinks can help your product page break into the top 10 and even outrank some of the biggest eCommerce sites.
  • Instead of focusing on ranking for a singular term, try to rank for many long tail keywords.

Note: Want a custom in-depth assessment of your search ranking performance? Contact us to get started.