3 Tools to Use for Conversion Rate Optimization

In a conversion rate optimization (CRO) or A/B testing program, the foundation on which you build determines your success. But that doesn’t mean you have to spend a ton of money and time to get up and running--far from it. There are a hundred different tools and services out there telling you that you can’t manage without them, but I’m here to tell you that 90% of your intelligence will come from 3 main sources:

1. Analytics

2. A testing program

3. Heatmaps and session recordings 

Honorable mentions: Usertesting and surveys.

I’ll go into each a little more in depth so you know what to look for and how to evaluate your options for each tool.

Analytics

The gold standard of analytics is still Google Analytics (GA). It’s free up to 10m hits per month, and offers a comprehensive toolkit. There’s just honestly no reason to go anywhere else as a small to medium business unless you have an app or hate Google products.

Caveats: 

• It’s for websites only, not apps. (You can use Firebase for apps.)

• To receive the maximum benefits of GA, you do have to commit some time and resources to configuring it and programming in custom dimensions, events and tracking. That can be a long and tedious process, and if it’s not well documented, it’s hard to keep up with the event naming convention strategy when a website changes. 

• You do get only 20 custom dimensions with the free version, so if you have a complex/deep experience and you need more than that, you will have to pay. But I sincerely doubt as an SMB you’d need more than 20 custom dimensions.

• GA is in the middle of transitioning between versions: they are sunsetting 360/Universal and transitioning to version 4. This requires entirely new custom dimensions and tagging, and the analysis is weighted towards events and not sessions. This is a slightly different way of looking at things, so if you (or your leadership team) are used to the reporting from the session perspective, it’s a bit of a mind shift to look at events instead. 

• That all being said, it’s actually far more useful to view your experience in terms of users, not sessions or events. So 🙃

As I said, for app and mobile analytics, you can use Firebase and run the analytics portion through Google Analytics. It’s a little more complex to install and configure because it works through an SDK, but it’s not too bad. (Or, as a former colleague of mine used to tell me about absolutely everything, “It’s just super simple.” I won’t do that to you, I promise.) Firebase isn’t super simple, but it’s more tedious than incomprehensible. Take some time and get your arms wrapped around a strategy on how to configure your installation, and you’ll be fine. Also, the Crashlytics is key to understanding how and why your app is crashing in the wild so you can deploy patches in a timely manner.

A Testing Program

This is The Big Question right now. With Google’s Optimize shuttering, a lot of companies are facing the bitter prospect of actually having to pay for an independent testing program. Google plans to develop a new platform that works in tandem with GA4, but it’s not here yet, and you’ve still got to test. So what do you use? You have some options:

1. VWO (Visual Website Optimizer) is a good choice and has lots of functionality, and is pretty affordable for SMBs. They charge based on your number of tracked users, and the free version is only good up to 50k users. But if you’re below that threshold or not much over, you can still get a good representative sampling in testing. The analysis interface isn’t the most intuitive, so you’ll need to invest some time to get to know the ins and outs. And you can’t use it to test apps. 

2. Convert has a lot of bells and whistles, and really is my favorite tool. But it can be really pricey for SMBs with pricing starting at $350/month. It comes in second only because of the price tag, otherwise it would be first.

3. Unbounce is really specific to building and A/B testing landing pages, not full funnels or more complicated experiences. If landing page testing is what you’re after, soo, just know going in that that’s all you’re going to get here. That being said, it’s kind of expensive for what you get: $100/mo for 1 domain, 20k visitors, though unlimited experiments. 

4. Optimizely is user friendly for experiment creation and analysis, but the price tag can be pretty steep, so I put it last on this list. For a larger business with deeper pockets, it even offers concierge testing where they just create and launch the experiments for you, but that’s out of reach for most small to medium businesses. 

5. Google Tag Manager (GTM). Yes, if you’re keen enough and willing to sink the time and resources into it, you can use GTM to create funnel redirection and experiment versions on your site. It’s a little in the weeds and you have to pull and analyze the data manually, but it’s possible. Going this route requires someone with extensive knowledge of GTM, and the experiments you can deploy would ideally be on the simpler side and not overly complex. I give this a solid middle-of-the-road rating based on the manual and limited nature of this solution. It’s also blocked by ad blockers, and does open up your site to security risks, so weigh your pros and cons carefully when considering this route.

Heatmaps and Session Recordings

Analytics and a testing program are going to tell you the WHAT. In order to make informed choices about how to choose the better option, you also need to know the WHY and the HOW of how users actually interact with your product, and it helps immensely to have heatmaps and session recordings. There are a number of these programs out there, but two really stand out:

Image attribution: ©Tschneidr / Wikimedia commons / CC-BY-SA-3.0

1. HotJar: It’s got what you need, it’s intuitive and easy to use, and has a really flexible pricing plan that starts at free. Implementation is very quick and easy with just a code injection, and away you go. Use the heatmaps to see where people are clicking and engaging, and spot check session recordings to see how actual users engage.

2. Microsoft Clarity works really well if your company already uses Microsoft products. It comes free with your MS suite of tools, has a very similar data set to HotJar, and is also a simple code injection to implement.

Honorable Mentions: Usertesting and Surveys

If you have a really complicated funnel and you know you should be getting somewhere but just can’t seem to get traction, check out Usertesting.com. It’s pricey. Very. But if nothing else is giving you answers, this definitely will. If Usertesting is out of the question for financial reasons, your best bet is to put up some quick surveys on your site. GetSiteControl is a good tool for that, and it’s cheap and easy to use.

And that’s it: all the tools you really need to get your own conversion testing operation up and running. Shoot us a message if you have questions—we’re always happy to chat.

Happy testing!