A Product-Market Fit lesson for SaaS founders
I recently reviewed a SaaS product that looked healthier than it really was.
Traffic was coming in.
People were signing up.
The category had real market interest.
On the surface, that feels like progress. If you are an early-stage founder, it is easy to look at those numbers and think: "The market wants this. We just need to improve the onboarding."
But there was one problem:
Almost nobody was reaching the first meaningful value moment inside the product.
That changed the diagnosis completely.
Traffic validated awareness.
Signups validated interest.
Activation was supposed to validate value.
And activation was effectively zero.
Near-zero activation naturally points to onboarding.
That was my first instinct too.
When we looked closer, the onboarding flow had obvious friction:
So yes, onboarding needed work.
But the deeper issue was not just screen-level friction.
The real question became:
Was the product attracting users who could realistically reach value through the current product path?
That is a different question from "is onboarding confusing?"
The analytics showed where users stopped.
Session recordings showed hesitation.
But support tickets explained why.
The pattern was clear: many users wanted automation outcomes. They wanted speed, clarity, and execution.
But the product required setup, strategy creation, and product-specific learning before the payoff happened.
That created an expectation gap.
The promise sounded fast.
The product path felt slow.
The users were not all failing for the same reason either. Two segments were getting mixed together.
The first segment was beginners.
They represented a larger market, but they needed a lot before they could succeed:
The second segment was automation seekers.
They were narrower, but the path to value looked simpler:
That led to the key product-market fit lesson:
The best first ICP is not always the largest possible market.
Sometimes the best first ICP is the segment with the shortest credible path to value.
A cleaner onboarding flow would help.
Better copy would help.
Fewer steps would help.
But if the wrong segments keep entering the same product path, onboarding improvements can only do so much.
The recommendation became an ICP narrowing experiment.
Not a permanent pivot.
Not abandoning the larger market.
Just a cleaner experiment:
Narrowing the ICP reduces variables.
Messaging gets sharper.
Onboarding gets simpler.
Support content gets more focused.
Product priorities get clearer.
Activation data becomes easier to interpret.
A signup is not product-market fit.
A signup means the promise was interesting enough to create action.
But it does not prove that the user understands the product.
It does not prove that the user can reach value.
It does not prove that the product can create repeatable success.
For this product, healthy signup volume created optimism.
Zero activation created clarity.
The strongest takeaway:
Low activation can expose a product-market fit problem hiding behind what looks like an onboarding problem.
So before buying more traffic, rewriting the landing page, or polishing onboarding screens, I would ask:
Strong interest creates momentum.
First value creates evidence.
Without activation, product-market fit is still a hypothesis.
For founders who have seen healthy signups but weak activation:
Did you fix it by improving onboarding, narrowing the ICP, changing the product promise, or something else?
The frame of 'shortest credible path to value' is the cleanest way I've seen this kind of trap renamed — most founders genuinely do read flat activation as 'we need better onboarding copy' and never get around to asking whether the people walking in were ever the right walkers. The piece I'd love to hear you push on: when you narrow ICP for an experiment like this, you implicitly throw away whatever early signal the other segments were giving — including segments that might activate fine with a slightly different product. How do you decide that cost is worth paying, especially early on when the dataset on any one segment is tiny? Is there a way you've found to keep the narrowing reversible without diluting the experiment itself?
You are so right to call this out. If you look at my earlier article, I kept the ICP intentionally broad as I want to learn which ICP resonates the most with the solution/product. You hit on the exact reason why I kept the ICP intentionally broad at first—I wanted to learn, and narrowing too early can cause a really costly reversal later.But when resources are limited and activation is flat, you have to find a way to get clean data. It's an experiment, not a permanent pivot: I’m not advising anyone to abandon the larger market permanently. It's just about choosing one path to fiercely measure right now. Identify who is trying to do what: Instead of throwing away other segments, you separate them. Finding that missed ICP just lets you direct them to a proper value path instead of letting everyone mix together and fail in the same loop.
In short, the key is ICP + shortest value path. You use it as a temporary test to build confidence. If it works, you double down. If it doesn't, you keep looking.
The 'shortest credible path to value' frame is the right one. But finding that segment is harder than it sounds.
When we did this, the highest-intent segment wasn't who we expected. We assumed power users would activate fastest. Actually it was the beginners who had ONE specific job they wanted done — not the ones who wanted to learn the whole system.
Support tickets told us more than analytics ever did. Users would say 'I just want to do X' but our onboarding taught them about Y, Z, and A first.
The shortest path to value often means ignoring 80% of what your product can do and getting someone to ONE outcome as fast as possible. Everything else comes after activation.
Absolutely True. My take is never based on expectation or assumptions. It is based on hard data. And you hit it hardest, real customer feedback cant be replaced with anything. Again, it is equally important to understand the activation moment for your product, if you get it wrong, you will go on witch hunt.
Same problem here. What worked for me was sending a single email 10 minutes after signup with one concrete action: "run this curl command and see live data." The people who did that converted. The ones who didn't, churned. Onboarding is really just getting someone to their first "aha" as fast as possible.
Bang on. Now, in todays era, email might be too late. Guide them right there in the application. You may be surprised how much improvement you will see.
This really resonated with where I'm at.
I recently had around 27 visitors to Ashive, but almost all of them bounced before reaching what I'd consider the first meaningful value.
My first instinct was to keep improving the onboarding, but I'm starting to wonder if I don't even have enough signal yet to know whether it's an onboarding problem, an ICP problem, or simply not enough conversations with real founders.
Still trying to figure out how to get enough honest users before optimizing the wrong thing.
Lets have a conversation. I can help. Or you can try to go over these questions which will make you think harder - https://product-led-growth.com/saas-growth-diagnosis
The useful distinction here is “screen activation” vs “workflow activation”.
A user can finish onboarding and still not reach value if the product has not helped them complete the job they came for.
I like the ICP narrowing idea because it forces one sharper question: which segment can reach the first real outcome with the least explanation?
That makes activation data cleaner. Otherwise every onboarding test mixes together product friction, wrong audience, weak intent, and expectation mismatch.
So True, you are bang on.
Thanks. I keep seeing the same trap in my own builds.
A user can complete the UI path, but real activation only happens when they finish the job they came for.
Harder to measure, but much more useful than just counting completed onboarding screens.
This feels like one of the most common SaaS problems.
Did you eventually figure out whether it was an onboarding issue, an expectation mismatch, or something else entirely?
Actually, it was more of an ICP issue; however the onboarding is still the issue and needs to be fixed asap. :)
That's actually encouraging to hear.
I think a lot of founders jump straight into rebuilding the product when the real issue is simply getting the right people through the door first. If the ICP isn't right, it's almost impossible to know whether onboarding is actually broken or just being tested by the wrong audience.Now that you've narrowed the ICP, I'm curious whether the onboarding issues look different than they did before, or if they've stayed pretty much the same.
The expectation gap feels like the key here. If the promise attracts users looking for a fast automation outcome, but the product first asks them to learn strategy and setup, even a cleaner onboarding flow may not fix activation.
I like the idea of narrowing the ICP as a diagnostic tool, not a permanent positioning decision. It reduces enough variables to show whether the real issue is friction, audience fit, or the product path itself.
Correct. I decided to go that path to really reduce the risk and validate the newly focused ICPs.
This is exactly my situation too. Signups but
no activation. Have you tried onboarding
emails or in-app guides?
I have ton of user support requests to know for sure that there are other issues than just onboarding flow. Breaking the flow into 2 parts is a simple yet effective solution for my client.
I am open to discussion if you need to brainstorm ideas. It can be icp, messaging as well.
One thing that helped me think about this: define the activation event as the first real workflow completed, not the first screen completed. For a consumer app I’d also split “curious signup” from “came here with a job to do” before touching onboarding, because those two users need totally different first sessions.
Usually activation events should be picked after looking at correlation between retention and the action that gives the highest level of retention aka best use case. In my case, I have to push users to activate first before I can view retention data. It is usually more than one factors. I use my framework to really diagnose the situation: https://product-led-growth.com/saas-growth-diagnosis
Interesting writeup.
The part that always trips me up is that onboarding fixes and ICP fixes can produce the exact same result.
Activation goes up either way.
Figuring out which one actually caused the improvement is where things get interesting.
To be honest , as an advisor I start by looking at fundamentals like ICP, key use case ( North Star ) . It can be onboarding that contributes the growth bottleneck; however I do start by looking from bottom to top.
What you are describing is true for products with product market fit. Onboarding will help only if you are attracting the right audience. Without that, onboarding will yield no results. I am open to discussion. Looking forward.
That's exactly why I commented.
The same activation improvement can justify very different conclusions depending on what you believe caused it.
Happy to share the fuller thought if it's useful.
What's the best email to reach you on?
would you mind sending me a request here ? I am interested in knowing how you approach this as well. We will be able to learn from each other. https://product-led-growth.com/contact-product-led-growth-consulting-services