Today is Monday. On Wednesday I have an long-haul flight to São Paulo and I just realized my traveling sneakers are done. I need my sneakers to be comfortable, because with my luck, the gates for connecting flights are always on the other side of the airport.
I need to find a new pair ASAP.
With everything I have to do to prepare, there isn’t enough time for a trip to the mall. In-between thoughts of the presentation I'm preparing, I browse different online stores and I find what I’m looking for.
But there’s a problem - I need my sneakers delivered overnight or I risk not having them in time for my trip. That’s not a risk I’m willing to take.
I look around the product page of the first store - there’s no information on shipping on that page, none.
I end up buying from my second choice simply because they clearly state next-day shipping is available.
Later I find out that first store also provided free next-day shipping, but that information was provided on the last page of the check-out flow. What a waste.
Have you noticed something that is extremely important in all ecommerce, but that I didn’t care about at all? Something that is the cornerstone of every site that is selling online, but in this instance, didn’t play any role whatsoever?
That all-so-important thing is the “Add To Cart” button.
It didn’t play any role in me passing on my first choice and buying from somewhere else. That button is extremely important, we have written about it before here and I personally have written a whole mega article just about that button here.
While the “Add to Cart” button is important and it’s one of the defining features of an ecommerce site, it does not carry the entire burden of of getting itself clicked.
You can have the most “optimized” button in the world but if it doesn’t work with the rest of your site or you’re just plain sloppy and forget to mention next-day shipping (for example) on your product pages... you are still going to lose business.
This article will try to give you a more holistic picture on what else you need to pay attention to when selling stuff online besides that damn button.
So let’s get started.
How Do We Discover Why People Aren't Clicking "Add To Cart"?
Image via Hello Matcha
So imagine that we’re online tea sellers, specifically we’re selling Matcha green tea online. The above picture is of our product pages.
We have high-quality product photography, psychological pricing, prominent “Add To Cart” button, some shipping info, instructions about making the tea, brief intro into Matcha teas and so on. Everything seems to be in place.
Additionally our Google Analytics data which tells us how many people land on this page (and how they got here) and we can see how many people clicked our buy button.
This is pretty good, but we also see that only .56% of our traffic is clicking "Add To Cart". That's not enough.
What we really want to know is what we should change about this product page to make even more people take steps towards buying.
User testing is one of the tools that can help us with accomplishing this goal.
User Testing Helps You See The Page Through Your Customer's Eyes
Your shop is online, you have your customers and because you are a smart marketer you have set up email list segmentation which allows you to email different segments (repeat, biggest spending etc) of your customers.
Before we go forward with changing things around based on “best practices” we must first get an understanding on what the buying experience is like for our customers.
You’ve looked at your product pages hundreds of times and you know by heart where everything is situated and it all makes sense... to you - very often, this is the exact opposite scenario for the majority of your visitors.
To solve this, use remote user testing to see first hand how visitors are using your site.
Remote testing works by recording audio & video of what’s happening on your visitor’s screen via screen capture software.
The facilitation is done either through pre-loaded prompts that appear over the top of the website being tested (when using professional user testing software) or simply in the form of activities do be done sent to the testers email.
In the end you will get a video recording where you can see and hear exactly what was going on.
Image via GotGroove
What Do You Ask During a User Testing Session?
As questions go, our goal is to understand what information is lacking from the buyer's perspective that makes them hesitate clicking "Add to Cart" and if the information they need to feel comfortable purchasing is easily accessible.
If I’m the owner of an online shoe shop, a customer would most likely want to know if they have my size, shipping info, materials used etc.
Possible tasks for user testers would include:
- Browse around the site, what kind of shoes is X selling?
- Find a pair of men’s loafers that you like.
- Do they have your size in stock?
- Add a pair of loafers to the shopping cart, then open the shopping cart and change the size of the shoe one size down.
- Find another pair of shoes you like and add that to the cart.
- Are the shipping options clear? How much are they, and what is estimated delivery time?
You get the picture. Try to imagine yourself in your customers shoes and what they would need to know to buy from you. Then derive tasks from that and use them.
Once you have gone through the process once with your current setup, it’s time to analyze and improve based on that feedback and then go again for another round.
Image via Dale Ahn Design
What Customers Do You Ask To Participate?
The question of who should to the testing is an easy one - your own customers!
Like I said earlier, you are a smart marketer and you have set up your list segmentation already. The only thing left is contacting those segments and asking for help with improving your site - you can offer discounts, free shipping etc to get customers to be more willing to join you.
As segments go start with your best customers, since they have bought from you numerous times already they trust you and are more willing to help you out. Be aware though, that you shouldn’t only rely on them - they have used your site extensively so try to include customers from all segments to get a fuller picture.
Additional segments you can test might be:
- New users
- Email visitors
- Social Media visitors
- Referral visitors
- High spenders
Displaying different messaging to these kinds of visitors can be configured with an optimization tool such as Optimizely.
If you need to make a case before investing in these tools, you can use any screen capture software, and instructing users to follow the think aloud protocol as they perform a set of tasks given to them in an email. Windows 10, for example, has screen capture build right into it, so your users don’t need to install any additional software at all.
Using Session Replays And Heatmaps To Watch How Visitors Interact
User testing on it's own can lead to some serious bias however. Because people know they're being watched, it can influence the answers they give.
Session replays are similar to user testing in that it enables you to see customers interacting with your site - where they click, what drives attention, which form fields they hesitate on etc. - without having a test administrator standing over their shoulder.
With the session replay, you're viewing videos of users screens as they use your site - there is no possibility to make them complete tasks or ask questions afterwards.
It’s biggest strength and weakness is that it records everything that everyone does - so depending on what you're trying to find, you may need to watch a lot of replays to get what you need.
Let's say you're wondering why more people aren't clicking "Add to Cart" and you watch through some session replays. In the replays for one of your more popular items you see that a significant portion of people are clicking the "size" dropdown, but one of the sizes is out of stock.
After which they either 1.) Leave or 2.) try to find another item in a similar size. If they leave, it's an easy assumption that they had no reason to click "Add to Cart" because you just didn't have what they wanted.
Using Heatmaps to Find Where Your Visitor's Attention Does And Doesn't Go
Moving on from session replays, heatmaps are the next step to find out how to get more people clicking "Add to Cart".
Strictly speaking heatmaps are divided into 2 categories - click-tracking heatmaps (see where the user is clicking and moving with his/her mouse) and eyetracking heatmaps (tracking users eye movement). By far the more popular is the clicktracking variety simply because it’s way more cost effective.
Another big difference between them is that click-tracking is used in real-time while eye-tracking requires specialized equipment and is usually done in focus groups and in a lab environment.
Then there is the debate if mouse-tracking has a high or low correlation with eye movement.
Meaning that, depending on who you listen, my mouse is either always around the area I’m reading or my mouse is in a random place while I’m reading. There really is no right or wrong answer here, so just use it with caution.
As a word of advice, use heatmaps that show actual clicks and disregard everything else.
Image via HowerOwl
With either tool however, the point is to see where the "hotspots" are on the page, or if the right elements are getting attention at all.
It's not uncommon to discover when viewing heatmaps for the first time, that important areas of the page aren't grabbing the visitor's attention, and therefore might as well be invisible.
Click-tracking heatmaps enables you to get progressively more data. With user testing realistically you’re getting data from 20 - 25 people, with session replays around 150 - 200 (you don’t really want or need to look through all the thousands of sessions).
And finally with heatmaps it combines clicking data from all your store's visitors and overlays them graphically over your website with easy to understand “hot” and “cold” areas - with the hottest ones getting the most clicks and vice versa.
Popular Click-Tracking Tools
As far as tools go, most of the popular ones include both session replay and heatmapping capability along with a host of other features like form analytics and scroll maps.
Tools also different on the definition of what a heatmap even is - some include mouse movement (hover) data while others count only clicks.
CrazyEgg, for example, has click-maps that show different traffic sources in different colors - great for visually seeing how behaviour differs between traffic sources.
On the other hand, it doesn’t include session replays which more advanced tools have. So your best bet is to read up on a couple of tools and their use cases, what they can and can’t do and decide then depending on your needs and budget.
When it comes to analyzing, go and read “Analyzing Survey Responses” section in this article to get started. It talks about user surveys, but same principles hold true here.
Image via UXPin
Example - Heatmap Of Category Pages
Image via Nielsen Norman Group
In the case of Pottery Barn (left picture) thumbnails of bookcases were studied intensively, while descriptions were mostly left alone. The opposite was true for an Amazon page which featured TVs (right picture) - only 18% was spent on photos, while over 80% was spent on the text.
In the case of the TVs, pictures were of no help whatsoever when deciding. Are you really going to choose your TV based on what’s on the display? So because you like football and there’s a picture of a football player on one of the TVs I’m going to buy that one? Of course not.
Still, be very careful when making these sweeping generalizations. Just because it works for Amazon doesn’t mean that it will work for you and vice versa.
There is no substitute for running tests yourself on your products with your audience. This is just to exemplify that different things work in different product categories and surprising results can be found.
Using Exit Surveys And Live Chat To Discover Why Visitors Leave Without Buying
Remember the story at the beginning of the article when I chose another vendor because I wasn’t sure that they provided overnight shipping?
If the store owner looked at their data, all they would have seen was that I was looking for sneakers, landed on a product page and then left without clicking "Add to Cart".
Fortunately there are ways for marketers to get more context on why I left and thus make changes to make me stay longer on my next visit and possibly buy. I’m talking about exit surveys and live chat.
Exit surveys pop up usually on the right side of the screen just as you’re about to leave the site. They can either provide text based answers where you choose the correct one or a free form field. In the shipping example the exit survey could have looked like something like this:
Image via Qualaroo
While this wouldn't have persuaded me to buy, it would give me the opportunity to let them know why. But for me as an business owner it gives me an answer to why people are leaving which I can then use to improve my store.
Another way to achieve the same goal is to use live chat where customers and potential customers can get answers to their questions and continue shopping with the piece of mind that what they were searching for is either available or not.
And that is exactly what your customers are doing - already back in 2012 customers said that they liked live chat because they get their questions answered quickly (79% of responders) and 46% agreed it was the most efficient communication method.
Angela Sokolovska digs deep into Live Chat in this article, which I highly recommend you read for more context on how useful it can be (and the massive opportunity it affords you.)
Image via Shopify
Now I understand that all this feedback and mountains of data can be overwhelming.
User tests, session replays, heatmaps, live chat etc all produces a lot of data and it takes a lot of time to go through it all. But here’s the thing - you don’t need to go through every single piece of data and analyze it like a mad person.
The idea is to “triangulate the truth” by watching and learning, and to develop a well-informed hypothesis, which will be the basis of a testing plan focused on getting more people to click the “Add to Cart” button.
Use Personalized Content To Tailor Messaging To Specific Traffic Sources
Image via Vizion Interactive
It’s no secret that traffic from different sources converts at different levels.
You can’t reasonably expect that traffic coming from your newsletter and traffic coming from organic Google search will convert at the same level. That’s never going to happen.
Different traffic sources want different things and their behavior and clicking patterns are different. People that know you (like your newsletter subs) have bought from you and they trust you already, while people coming from other sources may have never heard of you and so you must work harder on earning their trust.
One of the easiest things you can to do earn great trust right away is to make sure the messages you're promoting outwards on your social channels, ads and the like match what is happening on your actual pages.
Take a look here at this Facebook ad:
Image via Unbounce
And now compare this with the messages on their landing page:
Image via UnBounce
It’s a perfect match!
They are using the same words and call to actions (not to mention graphics) on both the ad and the actual landing page. That’s the way all clickable ads should be done.
Getting messages to match on ads and their respective landing pages is relatively easy but what if we could personalize pages for all the different traffic sources you are likely to encounter?
Now that would be awesome, and luckily for you it’s doable.
We already have all this data that we gathered through Google Analytics, user testing, watching session replays and exit surveys and live chat and heatmaps and more.
So why not put this into good use and instead of simply visually seeing how different sources interact differently with our pages but also change the pages depending on their browsing behaviour?
This is where personalization and behavioural targeting software comes into play. There are a multitude of different companies that offers these services starting with your favourite A/B testing software like VWO or Optimizely and a bunch of companies like Commerce Sciences, Personyze, Monoloop, Monetate and more that offer these services.
Essentially what they offer is an interface that allows you to choose your variables and then what kind of content they should be greeted with.
Image via VWO
You can segment by traffic source, previous purchases, their location and more. This is where all the research that you have done before comes in handy - you are creating different experiences for different customers based on what works best for them.
This is the holy grail of ecommerce - customer sees only the things that interest them in a way that is personally unique to them.
"Add to Cart" - these 3 simple words are the most important words in ecommerce, although you might be mistaken seeing as I have only mention the actual button once or twice throughout this article.
It is unquestionably a very important piece of the ecommerce puzzle.
All these small things play a huge and important role in your success. Just remember,it doesn’t live in a vacuum and to be truly successful you need to look at the whole picture and work on all the different parts of the puzzle.
It’s not going to be quick, it will take time and hard work but I promise you it will pay off in the end.
About The Author
Ott Niggulis is a chef/paramedic/freelance writer who focuses on marketing and CRO. Marketing is a numbers game and he loves numbers. Follow him on Twitter.