3
3 Comments

Optimising the Activation Flow and why it's so important to focus on it.. PT.1

Now that the new landing page is converting so much better (up from ~0.5% -> 3% still improving though) I've started to focus on optimising my onboarding and activation pipeline to start converting more users from free -> paid.

Thought I might as well document my thinking behind the changes.

What is activation?

The only goal of an onboarding flow should be to create the shortest path to the AHA moment for your users, which will lead to them to becoming an activated user and thus more likely to convert into that sweet sweet USD.

The AHA moment is that moment when your users literally says, AHA. They get the value of your product. The faster you can deliver them to that value the better. That's why one of the most important metrics I like to focus on is the TTV or (time to value).

Your goal should be to have the shortest time to value as possible. Do that and you'll have a much higher conversion rate guaranteed.

You might have a few different value props and that's ok, but your AHA moment is your most important one.

The difficult part is succinctly defining that value into a set of minimal actions. This is something you will likely get wrong at first and take some time to actually figure out, as you get closer and closer (or in that case of some of my products, further away) from PMF.

I definitely got it wrong at first.

My current definition of the AHA moment or moment of value for Rella is the moment that the user see's a REAL event, GENERATED by THEIR PRODUCT, ideally deployed and is able to use that event to power a DECISION.

This is still probably wrong and will definitely change in the future but for now based on the information I have, that should be the first aha moment for them.

For me this won't count as an activated user, as due to the nature of the product, it takes a while for a user to become activated as they need to generate a significant number of events that can power their decision making, at which point in time they'd be considered activated (this I'm still figuring out).

This is also not something I can achieve in my onboarding flow entirely again due to the above requiring time, but I can definitely push them towards it.

I wrote a whole article on this topic a few years ago so give that a read if you want: https://www.indiehackers.com/post/one-of-the-most-important-saas-metrics-thats-often-ignored-03c0c07e8f

So with that in mind...my thinking was how can I reduce the time to that moment as much as possible?

Why is this so important to focus on?

Well for two reasons:

  1. You're going to see more conversions -> more paying users -> reduced churn -> compounding learning effects
  2. The most important reason -> It forces you to constantly focus and ask yourself 'What value do we offer'.

Quick pitstop - Commitment

Another important feature of an onboarding flow specifically is building commitment. Getting the user to mentally 'commit' to going through the onboarding, usually through the sunk cost fallacy. You'll this quite common on flows like those myer briggs behaviour tests where they make you answer 25 minutes + of questions, only to then put your results behind a paywall. Since you've spent so much time on doing the questions, you feel like it would've been a waste to not pay the money now. In some cases like the above, building commitment CAN be deceptive, in other cases it's much more subtle.

One good example (can't remember it for the life of me) was a SaaS that had at the very start 'I'm ready to make my XYZ better' which was a button they had to click to continue. Adding this change, led to an XX% increase in onboarding completions, purely because it activates some sort of 'commitment' mechanism in the brain.

The old way:

Firstly we need to look at how we used to do it. The previous onboarding flow was quite bad for a number of reasons which just goes to show how easy it is to get it wrong at the start based on a number of wrong assumptions. I don't actually have any videos or screenshots of it (and god forbid I take the 5 minutes to git checkout and have a look) so text will have to suffice.

So the old flow used to have the following structure:

  1. Details:
  • First name
  • last name
  • Company name
  • Company size
  1. Integrate with Slack
  2. Send the first event:
  • Used to allow them to trigger their first event with a button press. People would just press this and then move onto the rest of the product and never come back. It was a pseudo activation.
  1. Complete

This was bad for multiple reasons. Firstly and most glaringly obvious issue is that the AHA moment is on the third step AND it's essentially fake-able. Users could choose a template button, then hit 'send event' which would allow them to sign in and NOT activate them.

Tsk tsk tsk.

Next I had the slack integration flow second. The reasoning for this was at the time I thought this would build commitment and if the integrated, they would be more likely to then send their first real event. Well turns out I was wrong on that. But you can't learn without trying and to me that's all these changes are. - small experiments that hopefully lead to increasing that %.

I was also collecting the company size early on as well which at my stage, isn't really needed. I know who I'm targeting, and for those who I don't there's a trove of valuable information.

General rule of thumb is if it doesn't help you activate the user, remove it. For some, this piece of information my be crucial.

Afterwards, I'd dump them straight into the dashboard, blind, expecting them to just magically again start tracking more events and getting the most out of the product.

Tsk tsk tsk.
---

Well I tried adding the new flow here but I guess there is a word limit on Indiehackers update posts? Here's a video showcasing the new onboarding flow

https://vimeo.com/794442768/6607d2ce35

I'll add a comment below with the thinking behind each step and changes:

, Founder of Icon for Rella
Rella
on January 31, 2023
  1. 3

    So to break down what's happening in the video and the new flow

    Step 1 - Commitment
    This is the most risky part of the onboarding and something I Might remove if it doesn't help. It is that commitment message I talked about earlier - reminding the user why they signed up 'convert, retain and delight more users' and getting them to subconsciously 'commit' to completing onboarding. Again this is an experiment so will see how it goes.

    Step 1.5 - Minor details
    Now I was tossing and turning about where to place this. I decided to leave where it is for now, because the information is important and is not as 'in your face' as pushing them straight into the event tracking screen. I wanted a lighter introduction first.

    Step 2 - Setup
    The new setup screen firstly doesn't require slack to be integrated at all. Secondly, feedback I had gotten was that due to the nature of the product, some users would struggle to get it setup themselves. I wanted to remove as much friction as possible so here I give them a choice. Either set it up your self OR jump on a call with us and we'll help set it up for you.

    So far this has been working very well with 2 people already selecting the 'get help'. This also is a great sign to indicate that someone is serious about getting the product up and running AND gives me a chance to learn about why they signed up etc.

    For those that choose self-setup, the process is completely streamlined. Before there was a lot of mental processing needed from the user:

    • What event do I send?
    • How best do I send it?
    • Should I trigger a fake event?

    Worst, the sample event I gave them was 'First event' which in no way is a real life event. Therefore not going to deliver my AHA moment.

    So now I remove all this friction. I decide your first event is your 'User signup event'. I give you three quick steps for how to get this setup and HOW long it will take for this to complete (stole this from June so shout out to them!) and broke it down. Before It was just one big blob of code which was intimidating and confusing.

    I also now don't let them send a fake event. I've decided that if I can't convince you here to properly integrate, then I most likely won't be able to later. Therefore, to proceed you need to complete this step.

    I'm thinking of also supplementing this with a 'Why should I do this?' section below the code that outlines what they will get from tracking this event.

    Step 3 - Slack

    Now I add the slack integration, with a skip button. It's not essential to use Rella, but is there. By now they've gotten their aha moment so it's ok to add some extra steps here to help push them to a converted user.

    Step 4 - Dashboard

    Now once they've done onboarding, I don't just dump them in the dashboard with no direction. The first thing they see is a alert at the top giving them suggestions for other things to start tracking that will be helpful for them to convert, retain and delight their users. This again is to remove any friction and thinking required by the user as to 'what to track'.

    Lastly - this is an experiment. All of this will change most likely in the coming weeks.

    Right now our conversion rate is ~3%. Will update on how this improves / declines from the new flow.

  2. 1

    Hi, thanks for sharing, I read your post when you first published it and I liked a lot, I came back now and looks like the website is not working and video is not available, can I ask what happened?

  3. 1

    Thanks for sharing!

    Any update on how this effected the user activation percentage?