Google’s Page Experience Update 2021: Prioritize UX Optimization for Success

Your website represents the character of your brand in the digital world. It is the catalytic force behind converting visitors into customers who first get hooked to your brand and then go on to become staunch ambassadors. Therefore, every element on your webpage needs to be present for a reason; which is to serve a…

Your website represents the character of your brand in the digital world. It is the catalytic force behind converting visitors into customers who first get hooked to your brand and then go on to become staunch ambassadors. Therefore, every element on your webpage needs to be present for a reason; which is to serve a memorable experience to the visitor.

What is a good page experience?

You deliver a good page experience to visitors when your page is built keeping in mind their expectations. It can be tempting to talk about what you want visitors to know about your product and build a page that persuades them to take the action you’d like them to take. But a website with great UX enables visitors to perform the tasks they wish to perform. 

Page experience is also about ensuring that your web pages load quickly, serve relevant information as clearly as possible without distractions, and are a safe browsing environment for visitors. 

The importance of UX here cannot be stressed enough. 

A PWC survey reveals that 32% users walk away from a brand they love after a single bad experience. 

In the same survey, 54% of US consumers say that customer experience on digital mediums of most companies needs improvement. 

This means a high number of websites are not optimized for page experience. This is likely to change with Google’s new algorithm – Page Experience, which is set to launch in June 2021. With this update, Google is aiming to prioritize websites that serve good page experience to their users over others. 

Google's Page Experience Update 2021

A win-win situation for brands and customers alike

Brand queries have always impacted rankings, but most brands that rank are not as big as the Nikes or American Expresses of the world. With this update, there’s even more opportunity for smaller brands to rank high if they have the right kind of content and deliver a delightful browsing experience. 

So if it was previously tough for you to fetch website visitors organically, this is your chance to rank higher. Even if you’re not well-known so far, you can perform better on the search engine results page if you show content that is relevant and useful to users. In the process of focusing on UX, you’re also serving your visitors better and paving the way for higher conversions on your website. 

Build Experiences That Your Users Love With VWO

There’s quite a bit of information available on Google’s algorithm, and you’re sure to have gone through the factors that constitute page experience: 

  1. Core Web Vitals – The loading performance, interactivity, and visual stability of your website.
  2. Mobile Friendliness – The experience delivered to users who’re accessing your website through mobile devices.
  3. Safe Browsing – To check whether your website contains malicious or deceptive content.
  4. HTTPS – To check if your website’s connection is secure and the pages are served over HTTPS.
  5. No Intrusive Interstitial  – The accessibility of the content on your website. 

We won’t repeat what you already know. Things like AMP, load speed optimization, security, malware control, are straightforward optimizations to help you get ready for this update. 

According to us, the bigger challenge is identifying the user experience that wins you the delight of users and improves conversions. This is where brands need to put in work. There is some input provided by Google to define the core metrics within the web vitals – Largest Contentful Paint (LCP), First Input Delay (FID) and Cumulative Layout Shift (CLS).

But in order to get these metrics right, you need to improve upon what successful brands have been doing for a long time now – 

  1. Listen to your customers
  2. Understand what your customers want 
  3. Give customers what they’re looking for
  4. Constantly experiment and evolve
Identify The Ux That Wins You User Delight

If you’ve incorporated the above steps in your optimization strategy, and have ensured that your online presence is technically viable in terms of speed and security, you’re potentially ready for this update. 

Vwo Insights Free Trial

Framework for improving website UX and performance

If you’re a marketer in today’s impact era, it is necessary for you to see the value in having a single customer view and analyzing entire user journeys. You need to – 

  • Uncover what your audience really wants
  • Plan your growth roadmap and prioritize based on your goals 
  • Learn what drives growth by testing various experiences 
  • Delight your audience by implementing the learnings at various touchpoints
  • Build a process where the above activities run in a continuous loop 

At a micro level, this framework incorporates the below process of experimentation – 

Framework For Improving Website Ux
  • Research and identify which parts of your conversion funnel need fixing. Use a combination of web analytics, heatmaps, session recordings, funnel analysis, on-page surveys, customer interviews, and other methods for valuable insights into visitor behavior. 
  • Build a structured hypothesis. This should be based on observations that reveal a problem that needs fixing, include a solution for a defined audience, and the expected impact/uplift on the goal. 
  • Prioritize your test ideas using the ICE model. 
    • Importance: How valuable is the audience for whom you’re making this change?
    • Confidence: How confident are you of achieving a better experience for users from this change?
    • Ease: How easy is it to implement this change? 
  • Choose the right kind of test based on your hypothesis – A/B test, Split URL test, Multivariate test
  • Measure to learn whether your UX improvement efforts are working. 

Other high impact changes that you can make

Content optimization

  • Remember that content will remain king and play the most critical role in determining page rankings. Audit your content to ensure it is easy to understand, solves a problem for visitors or addresses their needs, and is unique. Figure out which kind of content fetches the attention of visitors, and what persuades them to take action by A/B testing
  • Look at the content quality on your competitions’ websites, and find out how their content is different from yours – then aim to make yours better.
Ab Test Copy Changes With Vwo

Speed optimization

  • Optimize your website for speed and reduce 4xx errors (minimize HTTP requests, use asynchronous loading files, check JS loading and server response times, use compression and caching, and check image file sizes). 
  • Ensure that you have lazy loading correctly set up. 
  • Get rid of any broken links.  

UX optimization

  • Revisit your popular pages regularly and study the heatmaps of these pages. You will be able to identify the friction points and fix them to deliver a better UX.
  • Optimize for mobile search. Get your website mobile-ready by leveraging browser caching, reducing code, and reducing redirects. Also ensure that the design is responsive to smaller screens and the site structure is optimized for mobile.
  • Make sure that your CTAs are contextual based on specific points in the user journey on your website. Ideally they should convey some benefit to the user, apart from being well-designed and correctly positioned.

SEO optimization

Use alt text descriptions that are short, contextual, and ideally have a keyword because they are used by search engine crawlers for indexing.

Security optimization

  • Check your pages for malware or any deceptive content. Make sure it is safe for browsing and any personal information that the user enters is not at risk.
  • Use a secure HTTPS connection for your web pages.

Conclusion

Once the update is rolled out in mid-June 2021, user experience will become a direct ranking factor on Google. So if you haven’t already assessed your optimizing strategy and prioritized user experience, you’re pushing it to the wire. 

Start making use of heatmaps and A/B tests to analyze your website’s interactivity and its responsiveness to visitors. Run more experiments and figure out the design, functionality, and interactivity, to which your audience responds positively. Then make sure that you adapt to their preferences as swiftly as possible. This is the most logical way to achieve customer delight, improve your rankings, and achieve higher conversions.

When Should You Build vs. Buy Martech?

When marketers and IT folks think about whether to build or buy a customer experience (CX) platform or capability, there are usually obvious reasons to do one or the other.

As an economist by training, I tend to think of this choice in terms of what k…

When marketers and IT folks think about whether to build or buy a customer experience (CX) platform or capability, there are usually obvious reasons to do one or the other.

As an economist by training, I tend to think of this choice in terms of what kind of market exists for a capability.

When to Buy:

If the market is mature and there is a clear market leader

OR      

  • The market is highly saturated with many vendors

When to Build:

  • If the technology on the market cannot meet the needs of the organization, because it is highly specialized to the organization

OR

  • The tech simply does not exist

But, when there is not a clear choice, what factors should marketers consider?

First, it helps to think of homegrown versus off-the-shelf platforms and capabilities as less of a trade-off between internal resource costs and software licensing costs and more as a spectrum of customization. Ask yourself the following questions:

  • How much customization is possible using the current vendors in the market?
  • Which vendors are highly customizable (Drupal for example) and which function best out of the box?
  • Is the capability you need not something that could truly function well on an existing vendor platform?

If the function can be performed by customizing an existing technology, buying that technology will likely be easier and the most cost effective in the long run. Hiring specialists to customize the product, either as full-time employees or as a professional service will also be necessary. If the function cannot be performed adequately by an existing platform, building should be considered. However, there are more topics to consider before choosing to build.

Almost every platform in the martech stack will require some customization, including setup, connections to other systems, schema, and adjusting for the data itself. Therefore, customization should always be measured in relative terms (i.e., this content management system requires more customization than that one, etc.).

If that previous line of thinking led to “build”, then there is another important factor to consider: will this new capability or platform form a key part of your organization’s competitive advantage? For example, if you are a direct-to-consumer retailer, an ecommerce engine may be your bread and butter, but your competitive advantage is the product you sell and/or the customer service you provide. For an online travel agency (OTA), however, the booking engine is the competitive advantage, and thus, should be a custom-built solution. Additionally, if you are also thinking about selling or spinning off this new technology as a revenue driver, it makes sense to build.

It is never a good idea to reinvent the wheel, no matter what your IT department or developers say. In the long run, the focus of every martech vendor is on making money (and, thus, improving and supporting a good product) and will provide more value than any resource cost savings that you may gain in the short-term.

If, after all these considerations have been weighed, you do decide to go down the path of building your own tech, it is important to regularly reconsider the vendors in the marketplace or any available services that could replace your built tech. This can help to avoid the sunk cost fallacy, which companies with built tech are especially susceptible to.

The sunk cost fallacy is a behavioral bias towards valuing something you have already invested in more than something else that is better and will cost just as much in the future. This behavior occurs especially with legacy technology that may have at one time been a competitive advantage to the business and/or market leading. But, the sunk cost fallacy can also happen with newer tech (built or bought) if it is not compared regularly to competitors.

The sunk cost fallacy can lead a business to continue to customize existing platforms even when better platforms can be bought in the marketplace. The upfront cost of switching to a new platform is perceived as greater than the incremental costs of maintaining the existing platform. In the long run, the existing platform maintenance costs, as well as lost opportunities from working off an outdated platform, will likely outweigh the cost of implementing a new solution.

Aside from being aware of this bias, you can prevent your company from falling prey to it by making sure to measure costs and benefits in the long-term, and only considering future costs and benefits (rather than already-spent budget, aka sunk costs) when comparing existing tech to competitors.

Another important measurement needs to be the effectiveness of this new platform. Typically, businesses perform a cost-benefit analysis before undertaking the implementation of new technology, but these measurements can be important to gather both during and after implementation as well. They not only tell you how accurate your cost-benefit analysis was, but whether you are still on the right track to achieve what you set out to do. Knowing what margin of error is valid for future projects can help improve future cost-benefit estimates and there are certain qualitative “lessons learned” that may speed up future implementations.

Finally, I think it’s important to talk about emerging martech – it is a risk vs. reward trade-off that you must judge for your own organization. Knowing what risks exist before implementation is not unique to emerging tech but it is much harder to gauge. So, unless you are willing to take that risk, my advice is to neither build nor buy, but wait until you’ve reached the part of the hype cycle with which your business is comfortable. 

The Hype Cycle

Source: https://www.bmc.com/blogs/gartner-hype-cycle/

If you’re still not sure whether to build or buy, the Merkle Marketing Technology Consulting team can help. Reach out to us at marketing@merkleinc.com to chat with our experts!

The 4 Contextual States That Govern All Interfaces

Did you know that interface states are a lot like human states? Humans cycle through emotional states as they go through life. In the same way, interfaces cycle through contextual states as users interact with them. Problems can arise when these states…

Did you know that interface states are a lot like human states? Humans cycle through emotional states as they go through life. In the same way, interfaces cycle through contextual states as users interact with them. Problems can arise when these states aren’t visible. Imagine if humans never showed their states through body language, facial […]

The post The 4 Contextual States That Govern All Interfaces first appeared on UX Movement.

Delivering an Amazon Redshift Cloud Database with Minimal Disruption and Expense for a Leading Retailer

Since 2010, a leading apparel company used a CRM database solution hosted in an on premises Netezza environment at Merkle. In 2017, the hardware was reaching its end of life, and the company was at a crossroads when considering the investment to update…

Since 2010, a leading apparel company used a CRM database solution hosted in an on premises Netezza environment at Merkle. In 2017, the hardware was reaching its end of life, and the company was at a crossroads when considering the investment to update. The Merkle team engaged with the retailer to better understand their current challenges across both IT and marketing to define the best plan forward.

The Power of Exponential Growth | Data Viz to Simplify Complexity

There has been a lot of heartbreak around the world with the CV-19 pandemic. This chart, from NPR, illustrates some cause for optimism. It shows the 7-day average new cases per day across the world. It is crucial to acknowledge what’s hidden in the aggregated trend above: The impact on individual countries is variable. A […]

The post The Power of Exponential Growth | Data Viz to Simplify Complexity appeared first on Occam’s Razor by Avinash Kaushik.

There has been a lot of heartbreak around the world with the CV-19 pandemic.

This chart, from NPR, illustrates some cause for optimism. It shows the 7-day average new cases per day across the world.

Covid 19 Global cases over time

It is crucial to acknowledge what’s hidden in the aggregated trend above: The impact on individual countries is variable.

A large percentage of humans on the planet remain under threat. We don’t nearly have enough vaccines finding arms. We have to remain vigilant, and commit to getting the entire planet vaccinated.

Recent worries about Covid were increased by the proliferation of virus variants around the world. Variant B.1.1.7 was first identified in the UK. Variant B.1.351 was first identified in South Africa. Variant P.1 in Brazil has 17 unique mutations. The variant identified in India, B.1.617.2, had a particularly devastating impact (see the blue spike above). There are multiple "variants of interest" in the United States, Philippines, Vietnam, and other countries.

A particularly dangerous thing about variants is that they are highly transmissible (evolution, sadly, in action).

Some journalists rush to point out, hey, the death rate remains the same.

I believe this is a mistake. It imprecisely minimizes the danger, and results in some of our fellow humans feeling a false sense of hope. This is possibly due to a lack of mathematical savvy.

As Analysts, you can appreciate that a lay individual might not quite understand the complexity behind infection rates, and the impact on death rates. At the same time all of us, journalists and Analysts have to figure out how to communicate this type of insight in a way that everyone can understand.

This reality is similar to what we face in our business environment every single day. We have too much data. It is complicated. There are a lot of things happening below the surface. We somehow have to figure out how to preserve the complexity, but be able to communicate it simply.

Inspired by the work done by Adam Kucharski, mathematician, epidemiologist, and Mona Chalabi, Data Editor at Guardian US, I want to showcase how we can simplify complexity – in this case using Covid data, but the lessons apply across multiple use cases for Analysts.


The Problem.

Let’s simplify the challenge of explaining the problem we face with these variants down to these two scenarios:

Which of these is more dangerous:

1. A variant that’s 50% more deadly?

Or

2. A variant with 50% increase in transmission?

Take a pause with everything you know about Covid and math.

Ponder the problem, and what do you think the answer is?

Get a Post-It. Do some rough computations. Note your assumptions.

Did you choose #1 or #2?

Ready?

The Answer.

Have you heard someone say humans are not wired to understand the impact of compounding interest?

Applies here as well.

As Adam explained:

An increase in something that grows exponentially (transmission, in this case) can have far more effect than the same proportional increase in something that just scales an outcome.

Or, in English, as I suspect you all already noted above: #2 is the worse scenario.

It is far worse if the new virus variant is 50% more transmissible.


The Answer in Equations.

R represents the reproduction number. Let’s assume R to be 1.1.

This means that every 10 people who are infected will infect another 11 people.

[You can assume whatever R, math still works.]

F represents the fatality rate. Let’s assume F to be 0.8%.

This means that 8 out of every 1,000 people who get the virus will die.

G represents generation time. Let’s assume G to be 6 days.

This means that from the time that someone is exposed, it takes around 6 days for them to infect the next person. So each month, the virus can generate about 5 times. (30/6=5)

One final assumption, let’s assume 1,000 people were infected.

Scenario Normal

1,000 x (1.1^5) x (0.8%) = 12.9 fatalities after 1 month.

[For extreme clarity: 1.1^5 denotes 1.1 to the power of 5.]

Scenario 50% More Deadly

1,000 x (1.1^5) x  (0.8% x 1.5) = 19.3 fatalities after 1 month.

A sad increase for sure. But. Wait.

Scenario 50% More Transmissible

1,000 x ((1.1 x 1.5)^5) x 0.8% = 97.8 fatalities after 1 month.

OMG.

12.9 to 97.8.

This is why all the scientists, and ultimately Boris J as well, got so freaked out about a variant that was 50% to 75% more transmissible.

The math is scary at the higher end of that range.

If folks in your circle are less well versed in exponential growth (in their bank account, in loss in retail stores, or pandemics), do please take a moment to illustrate that for them.

You’ll be helping them think smarter.

Analysts typically feel that they are done at this stage. They got the data. They got the formula. they did the math. It is all so clear.

And, they are right. It is clear. But. It is not yet as accessible as it could be.

I urge you to think about accessibility of your work.


The Answer Visualized.

The “problem”, if I may use that ugly word, with the formulas above, is that they are a little bit dry.

Well. Maybe, it is more accurate to say: They are not quite as accessible.

A lot of people have a natural aversion to math. Even simple formulas like the one above can seem intimidating.

In life, and at work, perhaps the #1 job we have as Analysts is to be able explain data in a way that’ll be understood.

This is where Mona stepped in to help. She built on Adam’s excellent insights and explanation, and sketched some pictures that made the analysis potentially accessible to everyone on the planet.

Let’s do the exercise again.

Scenario Normal.

You’ll recall from above, the normal Covid scenario was:

1,000 x (1.1^5) x (0.8%) = 12.9 fatalities after 1 month.

Here’s Mona’s simplified visualization to make the formula a ton more accessible to all humans of the planet:

Covid 19 transmission - normal scenario

Isn't it more accessible compared to the formula?

Of course it is.

We all have this opprotunity in our day to day business work (in addition to an opportunity for all journalists who have to make this, literally, life and death data more accessible).

Let's keep going.

Scenario 50% More Deadly.

The challenge is to simply visualize this formula:

1,000 x (1.1^5) x  (0.8% x 1.5) = 19.3 fatalities after 1 month.

Here’s the visual:

Covid 19 transmission - more deadly scenario

Simple. Effective.

And now to our final scenario.

Scenario 50% More Transmissible

Our formula for the impact of 50% more transmissible:

1,000 x ((1.1 x 1.5)^5) x 0.8% = 97.8 fatalities after 1 month.

Mona’s visual to illustrate the impact:

Covid 19 transmission - more transmissible scenario

The simple visualizations, (possibly hand drawn?), make the data exponentially (there’s that word again) accessible.

While the data is heartbreaking, I had a momentary smile from appreciation for Mona and Adam for this timely and accessible lesson for all of us.


The Answer Visualized, Try 2.

There are many other ways to illustrate this data.

One thought I had was, I wonder if we should have three boxes?.

One for 1,000 infected. A second one for the increase in infections (huge in scenario three).

A third one for the unfortunate increase in deaths.

More information to sketch, I’m so wary of clutter in these cases. Something to sketch and see how it comes out.

My friend Kaiser Fung illustrated an alternative approach in a recent post on his excellent, and I really mean excellent, blog Junk Charts.

For a (poor) visualization used in a video published in Germany, showing the danger posed by new variants, Kaiser whipped up R = 1.0 and R = 1.4.

[Note: Being 50% to 70% more transmissible means the reproduction rate goes from 1 to 1.4.]

Here’s R = 1.0.

Kaiser Fung - R 1.1 visual

A very different approach to simplifying the complexity in the data, and a very different approach from the ones earlier in this post.

All in all, such an interesting visual.

I like that in a way it captures the haphazardness / randomness of the actual spread.

And here’s what happens when R = 1.4.

Kaiser Fung - R 1.4 visual

It depicts something truly heartbreaking, but does so in a mesmerizing way.

[I grew up in India, R1.4 reminds me of a mandala.]

More infections obviously mean more deaths (holding deaths constant as above).

The super nerd in me loves Kaiser’s version. There is an organic chemistry virusy nature of the visual that holds a certain appeal.

My experience would suggest that Mona’s is unquestionably more accessible. I would use something like Kaiser’s for certain audiences.

I wanted to share Adam's formulas, Mona's visuals and Kaiser's mandalas to highlight the diversity in the paths we can take on the quest for data accessibility.


Bottom line.

In the quest to communicate your insights more clearly, you can pick the path that works optimally for your audience knowing that there are multiple paths for simplifying complexity.

However you choose to do it, I urge you to figure out how to convert your numbers, assumptions, and formulas into a visual story that’ll make your insights more accessible.

Good for you. Good for the audience. Good for your company/planet.

Carpe diem!

As always, it is your turn now.

Please share your critique, reflections, and your lessons from the approaches you have taken in your quest to simplify data’s complesity. Thank you.

The post The Power of Exponential Growth | Data Viz to Simplify Complexity appeared first on Occam's Razor by Avinash Kaushik.

Now is the Time to Prep for Amazon Prime Day 2021

Prime Day is one of the most important events of the year for Amazon advertisers. Sales over Prime Day have seen major growth over the past several years, and advertisers can expect that to continue in 2021. The event is anticipated to be in June this …

Prime Day is one of the most important events of the year for Amazon advertisers. Sales over Prime Day have seen major growth over the past several years, and advertisers can expect that to continue in 2021. The event is anticipated to be in June this year and preparation is already underway. If you are not yet thinking about how to prepare for Prime Day, now is the time!

Prime day sales

How to Fit Many Menu Items in a Navigation Bar

It’s hard to fit a lot of menu items in a navigation bar with limited horizontal space. You can try squeezing them together, but you’ll end up overwhelming users with too many items at once. Many interfaces make the mistake of trying to fit too many it…

It’s hard to fit a lot of menu items in a navigation bar with limited horizontal space. You can try squeezing them together, but you’ll end up overwhelming users with too many items at once. Many interfaces make the mistake of trying to fit too many items in their navigation bar. As a result, users […]

The post How to Fit Many Menu Items in a Navigation Bar first appeared on UX Movement.