This post is part of a series in which we look at customer research projects that can make a difference in the way you understand your customers. In this opportunity we will explore how to go about identifying why qualified leads may not be converting during their free trial.
This type of research will help your team build a common understanding of your users and generate actionable insights that can drive optimisations for your on-boarding process and ultimately impact your conversion rate.
Although a bit skewed towards SaaS companies, this project can be adapted to other business models and serve as inspiration for your team. This post will cover what datasets can be used for qualitative analysis, what type of customer segmentation you may need and the outcomes for each project.
I hope you find them helpful. Let’s get started!
Understanding why people do not convert can feed into your messaging strategy, help you identify what your product is being compared to and point to a very specific set of features your product may be missing. This project can benefit both the product team and the acquisition team.
There are different data sets you can use to understand why people do not convert. For example, a common practice would be to create an automated email after the trial period is over. This type of email can be sent 2 or 3 days after the trial has expired.
You want to nudge people who still have the experience fresh in the minds. Write this email from your heart, remember you’re asking people to go out of their way to help you. Here’s an example of the type of email we send at the end of the trial. Specifically 2 days after the trial has expired.
Since these type of users tend to be disengaged by default, you shouldn’t count on many responses but it is worth doing anyway. Even a few email responses can be extremely helpful. Some people will come back to you and even offer to provide more details. Arrange a call with them right away and let them speak.
The purpose of this call is not to convince them to buy; instead you want to deeply understand the context behind their initial feedback. Another juicy dataset for this research project is sales data, demo or lost opportunity notes.
This is where having access to your sales teams notes will come in handy. If some of your potential customers also engaged with demos or were part of a sales process, any information you can gather around why the sales team lost a particular deal would be very helpful. Support tickets or live chat conversations can be handy here, too. If people who did not convert asked questions or opened tickets during their trial this can potentially help you understand any frustration or task they wanted to get done but couldn’t.
Now that you have gathered as much feedback from people who did not convert it is time segment your data. Not all feedback is equal.
Customer segmentation is critical for this research project. Here you want to listen to leads who were qualified one way or another, even it is is just a basic qualification, otherwise you will end up making decisions based on feedback from people who would have not converted even if you had the best product in the world. As Lincoln Murphy of Sixteen Ventures puts it:
Figure out what a bad-fit customer looks like, and what a customer with success potential looks like. If you don’t know that, nothing else matters, because you’re bringing in the wrong customer, and they’re never going to succeed. Then look at your existing customer base and see how many of those are bad-fits. This will give you a good idea of the eventual churn for your company. You’re bringing in the wrong customer, and they’re never going to succeed. This means getting deliberate. Go after the right customers and don’t sign the ones that have no potential. Take the ones who do have potential through a process that onboards them and ensures success. Use logical segmentation to give them the appropriate experience for that segment
Since NomNom imports all the user properties and segmentation data you need straight from your integrations, you can use those properties to group users based on the specific qualification criteria and filter feedback by those segments. Once you have filtered feedback from qualified users that did not convert, you are ready to send that data set to your research project.
Now that you have all the qualitative data you need, from the right segment, you can use your product analytics to dig deeper. We highly recommend checking Amplitude’s 10 Steps To Get You Started with Behavioral Analytics. This is where having a healthy tracking of properties and events can help you gather better context around what people do and what people say. For example, was this segment of users completing specific actions during the trial?
If you have already established a series of critical steps your trialists should take in order to experience an aha! moment with your product, you can look at this specific segment of users and compare their behaviour vs. users who were a good fit and converted.Did they invite other users? Did they experience any specific issues that were not reported?Combining inputs from tools like Amplitude and Fullstory can offer very useful context as part of the research process. At this point, you have enough data and you can start formulating a hypothesis about why qualified leads don’t convert. With your hypothesis at hand and all the evidence behind them it is time to experiment.
Building your hypothesis
We highly recommend checking out The 5 Components of a Good Hypothesis by Teresa Torres. Teresa suggests that the following elements need to be included in your hypothesis for it to be supported or refuted by an experiment:
- The change that you are testing
- What impact you expect the change to have
- Who you expect it to impact
- By how much
- After how long
A good example of a hypothesis that contains these elements is:Given evidence XYZ, we believe that adding X functionality to the on boarding process will increase user engagement during the first session by 50% Hypotheses are informed guesses and should include as much detail as possible. This will help you understand who to talk to and what kind of test you should execute on to prove or disprove your hypothesis.
Sharing insights with your team
Collecting the right evidence and formulating strong hypothesis can help you and your team have better conversations about what to build next. Make sure you have the right tools in place and that your are constantly gathering data across your customer’s journey. The more you optimize your research process the easier, faster and more effective your research projects will be.