The number that shaped Nima Aksoy's approach was not in a dashboard he built. His travel planning app, iPlan, went from zero to 500,000 downloads in a few weeks when ChatGPT launched, with no marketing at all. Investors called from every direction.
He exited before revenue, after he and his co-founder could not agree on the way forward. The lesson he took was about timing, and it shapes how he decides which data to trust today.
The short answer: Which data deserves trust depends on the product's stage: market demand first, the activation drop-off next, financial reporting at maturity.
What Nima Aksoy trusts, and for what
| Evidence | Trusted for | Not trusted for |
|---|---|---|
| Competitors' revenue, users and market size | Deciding whether to build at all | How users behave in your product |
| Registrations and activation drop-off | Seeing where self-serve users fall out | Proving there is demand |
| Paid self-serve conversion | Killing a feature or a whole motion | Judging a product before it charges |
| Monthly board reporting | Running a mature company | A product at the frontier |
Demand before dashboards
At zero users, product analytics is noise. Before he builds, Nima studies the market: what competitors earn, how many users they have, how big the space is. The go decision has two conditions, and both have to hold: the market wants it, and he likes the project.
A dashboard for the product doesn’t make sense at the beginning, because we don’t have many users.
Once users arrive, watch the drop-off
Bowora, his platform for surfacing startups to funders, began with a bare list of signups and no charts. A year in, the team tracks registrations and every user who starts a submission and does not finish it. That progression, from a list to a completion funnel to drop-off analysis, is the activation stage of a product-led business in miniature.
We control the registrations. We control all the drop-off users
Paid conversion is the kill switch
Features live or die on whether self-serve revenue appears. An automated content feature ran for a few months without a single paid subscription, so it was shut down. Social media scheduling did not work either, so it stopped. There is no sentiment in the decision and no long debate.
A spike is not a habit
The risk in Nima's world is not a wrong figure on a chart. It is building tracking before demand is proven, or reading a distribution wave as product-market fit. iPlan's downloads were real, and they arrived because the market tilted toward everything AI at once.
He now builds with very small teams. Vocaler.ai, which he estimates would normally take ten to fifteen people for five or six months, is two people working for two months.
With AI, two people is more than enough.
What to do with this
Match the tracking to the stage, as Nima does, and do not build heavy tracking before demand is proven. The one exception we would add is ours, not his: the events for your activation step. Retrofitting a funnel after the first wave means that wave is a cohort you can never measure.
For seven checks drawn from across the series, see the questions to ask before trusting a number.
Frameworks this interview tests
- Aha moment and time to value: The aha moment is the action after which users become far more likely to stay. A borrowed number misleads; your own needs a validated event and a fixed time window.
- Freemium vs free trial: Freemium gives a permanent free tier for volume; a free trial gives full access for a limited time. Dave Boyce adds the reverse trial and an experiment to decide.
Read next in The Question
Source and disclosure
Read Issue 3 on A Founder's Friend. Quotations on this page are copied from the published issue.
Skene sponsored this research and commissioned it on a single question. A Founder's Friend ran the interviews, wrote them up and published them, and the series states that Skene paid for the research and did not direct the questions, the analysis or the conclusions. Each of the thirteen interviews was reviewed and signed off by the person in it, named or anonymous by their choice. Appearing in the series is not an endorsement of Skene, and nobody here is presented as a Skene customer.