One of my favorite parts of any presentation is the Q&A portion. Sometimes I even get to answer questions for 20 – 30 minutes. My second favorite thing would be to hold a panda in my arms like this guy (if you can make this happen, let me know).
Going back to the Q&A stuff. This is the best way for me to give the audience relevant advice to their situation and for me to figure out what I should be writing about.
Here’s a question I got recently (and which I almost weekly):
“Our product (available on web, iOS, and Android) has a multi-step signup funnel that users have to go through before using the actual product and we aren’t happy with our customer onboarding metrics. We think we are losing a significant amount of users but we aren’t sure how to start analyzing it and improving it. This is made worse since we are actively spending $10,000+/month on paid user acquisition campaigns.”
This is a great question because all products have some kind of funnel that users must complete before they can use the product. Even if your signup funnel is only one step, you’ll likely have a series of actions that users should complete during their onboarding.
As the reader mentioned, this becomes even more important once you start running paid user acquisition campaigns to get more users.
Before we can start optimizing anything, we need the right customer onboarding metrics.
Onboarding funnels don’t have to be complex. Here are the general metrics that I recommend to my clients to start tracking:
- Overall Conversion Rate: obvious but a lot of companies don’t have this. This is the overall performance for the onboarding and how it is trending over time.
- Drop Rate in Between Steps: now we can go deeper into where users are dropping off the most and where the biggest opportunities will be.
- Time to Complete Funnel & Steps: this is the average time it takes to complete the funnel any given step. We’ll look for outliers here e.g. users who do this quickly vs users who take a long time.
- Valuable Segments: what are the most important segments that you should be tracking? Examples include authentication type (facebook, google, email), location, marketing source, etc.
Storytime. I have a client who has a mobile consumer app and averaging $5M/year in MRR. Their top of funnel metrics was solid and could be scaled. Their onboarding, on the other hand, was pretty poor.
Users were being thrown into a random flow to get them set up but it was just taking too long. They started running A/B tests to remove steps and get users closer to the actual goal (using the app).
If you’re starting, here are the 4 steps that you can go through to start finding optimization points.
Step 1: Create a Conversion Funnel Report Using Accurate Data
The main report that you need here is called a funnel report and you can get it from tools like Mixpanel, Kissmetrics, Amplitude, Heap Analytics and much more. This is one of the many reports that you should consider creating especially as your company grows.
This report looks like this:
The above report comes from Mixpanel and it shows us the overall conversion rate for the funnel at the top and the conversion rate from step to step in between the columns. The inverse of the conversion rate is our drop off.
You want to be able to get this level of granularity inside your funnel to understand what is going on. You’ll be looking to improve the areas with the lowest conversion rate in step 3 but for now, let’s make sure that you have accurate data going into this report.
Once you set up a tool like Mixpanel, you need to make sure that the numbers you’re getting are accurate. This means your development or engineering team will need to double check your funnel report data against another data source like your product’s database.
Keep in mind the following:
- When checking different sources of data, ensure that they are both in the same timezone.
- Look at multiple time periods e.g. multiple weeks or days
- A 5%-10% discrepancy is ok. Analytics tools are sometimes blocked by adblockers and it is common to “miss” some of these users.
Finally, you should be able to know exactly what is the criteria for how each step is counted. For example, if a user has to submit a form in step 2, what happens if the user submits the form twice? Or what if they submit the form but it has incorrect information and the user has to submit the form again? Are you going to get double the user numbers?
Once you have a funnel report like the one above set up, you can move on to step 2.
Step 2: Find Valuable Segments
The overall performance of the funnel is interesting and for a lot of companies, it will make a big difference to have.
However, you want to go a step further and earn bonus points. To do that, you should look into segmenting your funnel.
In the above image, we are segmenting our funnel by country to see if there are any specific countries that convert at a higher rate than the average.
Segmenting will let you look at specific groups of users and how they performed at any given step. You could use this to find user groups that are converting at a higher rate e.g. users from a certain marketing campaign.
There are basic segments that your analytics tool will give you out of the box such as city, country, browser information, mobile device information and marketing campaigns through UTM parameters.
There are also advanced segments that you will have to manually implement to get access to. These segments will depend on your product but here are a few examples that I have seen from past clients and readers:
- Authentication type e.g. Facebook vs Twitter vs Google
- User type e.g. admin vs end-user
- Billing plan type e.g. free vs premium vs super premium
To figure out what advanced segments you should be tracking, think about what actions or choices your users can make throughout your signup funnel. You can then work with your development team to add them as user attributes to your analytics tool.
Step 3: Understanding Drop-offs and User Expectations
Now that you have accurate data and advanced segments, you can start to analyze your funnel report to see what is going on. The areas with the lowest conversion rate or highest drop off rate should get your attention right away.
However, your data will only tell you a limited amount of information about your drop-offs so you’ll need to go through the signup funnel just like a normal user would.
A lot of drop off issues come down to expectations, that is a user was asked to do something they didn’t expect or vice versa.
I had a client once who was asking users to “Request a Trial” using a sales form where they asked for name, email, company size, etc. This is normal.
However, if the user’s company was under a certain size, they would get thrown into a self-serve signup process. This didn’t really match up the expectation of requesting a trial which usually means that a sales team will contact you.
They decided to change the language to “Start a Trial” which made more sense to their users. They were able to see a higher conversion rate from step 1 to step 2 just based on this small copy change.
Step 4: Develop Tests
Our final step is to develop a test where you try to improve the conversion rate of a specific step and the overall funnel performance. You can remove/add steps, change the copy in a page, send targeted messages to guide users through critical steps and much more.
You’ll now get into the world of A/B testing and your funnel report will be a critical tool in testing what works and what doesn’t. Consider tracking a specific A/B test variation as a segment which will let you track how those users (who saw the variation) performed throughout the funnel.
This is a good starting point. Onboarding flows are the glue between acquisition and retention. I wrote another article where I talk about some best practices when designing (or redesigning) flows based on products like Slack and Spotify. Check that out and as always, let me know if you have any questions.
I find most companies are stuck with high-level metrics and they aren't able to properly understand what actually drives user growth for their web and mobile products. To do that, you need the right data and the right tools.
If this sounds like your situation, then you should download our free tracking plan (and tutorial video). This is the document that you should create before you ever implement tools like Mixpanel, Amplitude, Segment, and Intercom. Click the image below to download your own free tracking plan (and tutorial video).