Analytics is all about understanding how your users perceive and use your website or app. Analytics services allow you to track your users behavior as they navigate and click around your app, and generate reports to track progress and tweak your application to nudge users towards the behavior you want.
This article is split into two parts. The first part is a basic introduction on how to think about analytics and some commonly used terms. The second part introduces three common services that each bring incredible value to Bubble developers and are easy to implement and use.
The first section is directed at beginners, and if you know your way around the basics of analytics already, you can skip directly to the service comparison.
Why we do analytics
To understand the importance of analytics in your application, let’s look at two scenarios that illustrate different perspectives on how analytics can drastically change how well you know your own business.
The icy staircase
Imagine that in my cold and dark hometown of Oslo, there are two trendy fashion outlets right next to each other, of which you are the manager of one. You count every customer that comes in through the door, and so does your competitor, and in the evenings you compare your numbers – annoyingly she always seems to top your 100 customers by bringing in 115.
You change the window display, offer discounts and take in new brands to attract clients, and your competitor follows in your footsteps. Even if you’re the innovator, she always beats you by 15%. The battle to be the most trendy store in town is brutal indeed.
To learn more, you move out of your store and watch it from a short distance. Customers coming from both directions study the window decorations, point at the mannequins and move towards the door. Mostly, they enter at the same rate, but throughout the day you start seeing a pattern: some customers seem interested, but stop right before reaching the door. You approach one of them, and ask why they didn’t enter. Are the brands not right? Are the mannequins not wearing the coolest outfits? Are the staff not hip enough? She points at the door and gives you an embarrassed look: «I was afraid to slip and fall on the icy staircase.»
The point of the story is of course that your users will leave your front page, drop out of the signup process or stop using your app altogether for all sorts of reasons that may not be obvious. By studying what users do, and sometimes more importantly what they don’t do, you can learn to salt your staircase and remove one obstacle for visitors wanting to visit your store.
Let’s illustrate this with an online example: let’s say your website has an awesome designer look with a very visible button to sign up. Still, the sign-up rate is lower than you hoped. People are not clicking the button.
By doing some basic analytics, you discover that the majority of visitors have a screen size slightly smaller than your own: for you, the sign-up button is resting visibly at the bottom of your screen, but for a lot of visitors, they have to scroll down 100 pixels to see it: the need to scroll is their icy staircase. The store looks good, but a trivial detail is stopping 15% of users from walking through the door.
Improving your pages is not about finding one detail like that, but to use analytics to spot patterns in user behavior that allows you to keep making small improvements.
The unexpected visitor
With the learning from your last experience, you’ve salted the staircase, and you’re happy to see that you’ve reached the same level of average daily visitors as your competitor. Awesome!
But what good is a visitor if they don’t spend money? you think to yourself, and you ask your competitor to share his sales reports – to your frustration, they are also 15% higher than yours, even after you managed to raise the number of visitors. She smiles smugly, as if knowing already that her numbers were superior.
You go on a hiring spree and find the hippest fashionistas in town, train them in sales techniques and prepare them to go to war: a month later your competitor is still outselling you by 15%. Frustrated, you decide to take the same approach as last time and learn from your customers. You prepare a form for your guests where they can tell you why they didn’t buy anything today, and you get lots of answers that while interesting, don’t offer a solution. You observe one detail though: even though every customer gets a form handed out to them, the number of forms is smaller than the number of people coming through the door. Curious, you observe your customer as they fill out the forms, and after a few hours, you see one pick up a form only to put it back down without filling it out.
«What’s wrong?» you ask. «Are the staff not cool enough? Is the form too long? What can we do to improve?»
«Que?» the customer responds with a puzzled look. As it turns out, your store is near the city harbor, making a significant portion of your users tourists coming from visiting cruise ships. Your staff may be cool, but they don’t speak Spanish.
Again, analytics tells you something about your customers that gives you actionable insight. With simple analytics, you learned that a percentage of your customers shared a specific attribute that you didn’t know about and hadn’t prepared for.
Moving again to the online world, analytics will let you to learn a lot of similar things about your users: where they come from geographically, what kind of device they are using, the time of day they’re most likely to visit, how they found your site, how quickly it loads and hundreds of other details.
The world’s best desktop app is not good enough if a majority of your users are trying to access it from mobile devices.
Working with data
To make sensible business decisions, you’re gonna need data. With Bubble, you can not only build tools to gather data yourself, but you can use external services that are designed to collect, aggregate and visualize the data for you in a digestible format. The world of analytics may seem overwhelming at first, but it’s not any different than any other project: it just needs to be broken down into smaller chunks done one by one.
For an easy glossary of different terms used in Analytics, have a look at this page.
As an entrepreneur or project manager, a keyword in your work should be focus. What does that mean exactly? It means that what you chose not to do can be just as important as what you choose to do. Spending your time and energy in the wrong place is counterproductive, costly and energy-draining, so pick your areas of focus carefully.
Analytics is no different. What you’ll find as you start exploring the different tools is that the amount of data they can provide is baffling. Depending on your project, some things can make or break your business, while others are distractions. Don’t make it into astrophysics – simply split the data into two categories:
- nice to know
That’s it. Most data points are interesting, but not all have actual value in decision making. Focus on those that do, and try to avoid being distracted by everything else. You may have heard the term paralysis by analysis – it was invented for a reason. Split your work into projects: one week you may focus on the front page, while the next you focus on your sign-up funnel. It’s easy to drown yourself in data, but having a disciplined approach helps you make those small changes that over time increase your traffic and conversions.
Most likely, your app will consist of different sections. Let’s say you have created an accounting app for SME’s, then your app and the front page of your website will be two different sections, and they’ll have different analytical needs.
For the front page, you may be interested in things like where your users are coming from, how long they stay on your site and whether they sign up for your service or not and where in that funnel they drop off.
In the logged-in section of your app it can be useful to learn more about how the app is used: actions that the users take and don’t take, how frequently they access it, how long they stay paying customers, how many users they refer – the list can be long, all depending on what your app does.
Think about how data can help you improve the total quality of your product, not just increase your traffic or revenue. On your sign-up page, conversion is of course paramount, but if you have a documentation section, the total amount of traffic may not actually be that interesting. What you’ll want to know is if users actually find the answer they were looking for and if it relieves the pressure on your customer support. Different kinds of data requires different kinds of measurement, and the different tools below all have core strengths that are useful in specific scenarios.
Tools of the trade
We’ll look into four different tools in this article.The great thing about this constellation of services is that they don’t only have different strengths but clearly identifiable differences in what they offer. In other words, you may find that one tools can be immensely useful to your particular business while another can safely be ignored for now. Let’s have a look.
You may have heard of this one. Jokes aside, one of the easiest ways to gather data about your users is to use Bubble for what it’s good at: flexibility. If you need to gather data on any kind of user event, Bubble gives you the ultimate freedom. Any workflow can create and update data in your database that ultimately can provide valuable data down the road, and it will all be identifiable down on to a user level. If you want to gather data on how often Users logs in, you can simply add a workflow that saves that data every time.
There are several upsides to using Bubble:
- It’s all saved in the same database
- you’ll get a 100% success rate in saving data
- Ad blockers are not an issue
- You can save any kind of conceivable data on any kind of conceivable trigger
While Bubble is great for collecting data at very specific triggers, you have no data on your users apart from what you’re actually collecting. Bubble can’t tell how they reached your site, how old they are, what language they are speaking, what country they are in or any other kind of information that you haven’t specifically made an attempt to collect. Aggregating the data and visualizing patterns ways that are easily digestible may prove difficult. Bubble is great for collecting data about highly specific events that can flexibly be categorized or traced down to a user level, but for big data you’ll want to look more broadly.
It’s difficult to talk about any kinds of web analytics without mentioning Google’s behemoth. With its incredibly generous freemium model, GA has become the go-to tool for every website becoming serious about collecting and aggregating data. GA being free is a win-win for both you and Google as most of their products: you get a free product, while they get to collect astronomical amounts of user behavior data on practically every website out there. This in turn helps their Google search engine become ever more precise.
GA is quick to set up and easy to use. A majority of the data and reports is available from the moment you’ve set it up and it’s had time to collect it. The dashboard may seem overwhelming at first, but you’ll quickly learn to navigate the different categories and Google offers thorough, helpful videos to get you started.
So why would you choose any other tool? Well, Analytics comes with pros and cons like everything else.
Google Analytics is great for:
Google Analytics is all about traffic. It collects massive amounts of anonymous data on who your users are, where they come from, how they found your site, how they navigate it, what your popular pages are, how long they’re staying and a range of other data points. It’s commonly used in tandem with marketing channels such as Google Ads to analyze the return you get on your marketing spend.
How difficult is Google Analytics to use?
GA can be easy or complex, depending on your needs. Implementing and reading the standard reports is easy, while setting up more advanced reporting like the Enhanced eCommerce module (which lets you track product sales) is more complicated
Mixpanel has become another huge player in the analytics community, and for good reason. While Google Analytics provides incredible amounts of aggregated data and metadata about your users out of the box, Mixpanel is more about identifying and analyzing actions that your users take. While Google Analytics does have event tracking, Mixpanel is easier to set up and with its support for a big number of event properties (information that you can add to an event, such as the amount on a cart checkout, the color of a t-shirt sold, etc), it’s more product-focused than GA.
Mixpanel is great for:
Mixpanel is all about events. It lets you easily track specific user actions in your app (such as a button/link being clicked, transaction being completed or a form being sent), attach information to that event (amount, category, etc) and flexibly set up reports that combine events into funnels. For example, you could follow a customer all the way from first viewing a product to actually having paid for it and analyze where in that journey customers drop off and try to remove obstacles on that path to increase the conversion rate.
How difficult is Mixpanel to use?
Mixpanel requires a bit more work to set up, but is not particularly difficult to use. Bubble’s own plugin will add the needed snippet to your headers and comes with documentation. The strength of Mixpanel and Bubble together is that you can record events simply by adding it as a step in any workflow, making it extremely flexible and easy to use.
Hotjar let’s you collect and analyze data in a more qualitative way: they specialize in visualizing how your users see and interact with your site and even lets you record single users sessions. How far down do users scroll? How do they move their mouse? Where do they click? You’ll see the results not as graphs and data tables, but as overlays on your actual site as heat maps, showing your most popular clickable elements as bright red.
Hotjar is great for:
Hotjar is all about qualitative data and user feedback. It lets you actually follow a single user’s interaction on a page, down to scrolling and mouse movement. Heatmap overlays quickly identifies which parts of your page users find most interesting, and by watching recordings of real interactions, you can test hypotheses and see exactly how users spend their time on your page. It also lets you set up survey forms to collect user feedback. Hotjar is easy to set up and provides data more or less out of the box.
How difficult is Hotjar to use?
Hotjar is set up to give answers quickly and is easy to use. It doesn’t overload you with data like Analytics can do, but excels in the fewer tasks that it performs.
Using Google Tag Manager to manage implementations
If you’re new to analytics, introducing one more tool that doesn’t actually provide any data on its own may seem like the last thing you want – but hear me out. Google Tag Manager serves one simple purpose: it serves as a container for snippets of code (Tags) that you ma want to add to your site. Most analytics tools work by adding some code to your page header, and with GTM you only need to add one. Every service you want to add after that can simply be loaded into the Google Tag Manager interface, and it will be live instantly. There’s no need to re-deploy your app.
Take my word for it – the few extra steps and little bit of learning to implement Tag Manager is worth it. It’s out of the scope of this article to provide a tutorial, but MeasureSchool provides a great introductory look at how to set it up for the first time (and lots of useful lessons on the tools discussed in this article).\
Thanks for this great writeup.
But what about Segment? It appears that the Bubble plugin for Segment is the only plugin option for server-side analytics, as opposed to Google Analytics, Mixpanel and Hotjar is that they only offer client-side tracking.
The problem with client-side tracking is that ad blockers and Do Not Track features in browsers can block client-side tracking…right?