Not every strategic bet pays off. The ones that do not are the ones you learn the most from, if you are honest about why they failed.
I want to write about a channel experiment I ran over approximately three months that consumed a significant portion of my time and a meaningful fraction of my runway. It did not work. I am writing about it because I think the failure contains more useful information than most of my successes, and because I see other founders making the same bet without the visibility into how it plays out.
The bet and the reasoning behind it
My thesis was that startup coaches and independent business advisors — people who work with early-stage founders on a paid, professional basis — would be a high-leverage distribution channel for my product. The reasoning was straightforward: these professionals have direct, trusted, ongoing relationships with exactly my target customer. They speak to multiple founders every week. If I could get ten coaches to recommend my product to their clients, I would have a referral engine that reached thirty or forty founders a month with a warm introduction.
The economics looked attractive. CAC through this channel would theoretically be near zero — a small revenue share or affiliate fee rather than the cost of direct acquisition. The quality of leads would be high because coaches would pre-qualify the fit. The conversion from coach recommendation to purchase would be strong because the trust transfer from coach to product would reduce the evaluation barrier.
I built a partner programme. I created onboarding materials for coaches. I developed a presentation about how my product would complement their client work. I contacted 47 coaches and advisors over the course of six weeks. Twenty-three responded. I had substantive conversations with sixteen. I signed four as active referral partners.
What went wrong — in layers
The first problem was incentive misalignment that I had not adequately modelled. Startup coaches and advisors are protective of their client relationships in a way that I had understood intellectually but underestimated emotionally. Several coaches who were genuinely enthusiastic about the product became hesitant when they thought through what it meant for their positioning. If my product could systematically answer many of the questions founders pay coaches to help them work through, what did that imply about the coach's ongoing role?
I heard this framed several ways. One coach said it directly: I love the product but I worry about recommending something that makes my clients feel like they need me less. Another said more politely that she was not sure the product was right for her clients at the stage she worked with them. What both were really expressing was a worry about competitive encroachment, even though the overlap was not as large as they feared.
The second problem was the sales cycle. I had expected to spend a week or two building a relationship with each coach before they started referring clients. The reality was that the trust required for a coach to put their name behind a product recommendation to clients — people who trusted the coach's judgment — required much longer. The average time from first contact to first referral from my four active partners was eleven weeks. I could have acquired many direct customers in that time at lower cost.
The third problem was referral quality variance. The coaches who did refer clients sent me founders at very different stages and with very different needs. My product is designed for founders at a specific moment — pre-validation, pre-raise, pre-product-market-fit. Several of the referrals I received were from founders who had already moved past that window, or who had specific problems my product was not designed to address. My conversion rate on coach-referred leads was actually lower than my conversion rate on direct outreach, which was the opposite of the hypothesis.
What I salvaged and what I learned
I ended the programme in its current form but maintained the four partner relationships that had generated any value. The framing that worked with those four was fundamentally different from the framing I had started with.
Instead of positioning my product as something coaches could recommend to clients as a standalone tool, I repositioned it as something that would make the coach's sessions more productive — a shared analysis the coach and founder could review together, with the coach adding interpretation and the product providing the underlying framework and evidence base. This framing made the product an enabler of the coaching relationship rather than a partial substitute for it.
Two of the four coaches are now active referral sources under this framing. The programme is small but real.
The lesson I carry from this experiment: distribution channels that depend on other people's trust with their clients are much harder to build than they appear because you cannot control the incentive structure. The fastest path to distribution is almost always a direct relationship with the end customer, even when an indirect path looks more scalable on paper.