23 days of distributing a SaaS. Here's exactly what happened.

Really valuable breakdown, Muhammad. Congrats on sticking with it, and thanks for sharing the numbers so openly.

The biggest takeaway for me is that early-stage distribution is less about being everywhere and more about getting close enough to users to learn why they would actually pay. The agency owner giving feedback and seeing the feature shipped the same day is probably more valuable right now than thousands of low-intent impressions.

I’d keep leaning into those 1:1 conversations until the same pain and buying trigger starts repeating.

This is a strong post-mortem because it separates motion from signal.

A lot of early distribution feels productive because it creates artifacts: posts, directory listings, reels, threads, newsletters. But the only thing that changed your product direction was a conversation with someone close enough to the problem to create evidence.

The way I’d frame the next 7 days:

• Don’t ask “which channel gets impressions?”
• Ask “which channel creates a conversation with someone who can describe the pain in their own words?”
• Then ask “which of those conversations changes what I build, sell, or stop doing?”

For a tool like ColdCraft, I’d probably measure fewer things at first: number of real sequences generated for a real business, whether the person uses/edits/sends them, and whether they ask for the next batch.

That tells you more than traffic because it tests the job-to-be-done, not just curiosity.

This is also the reason I’m building around validation: founders need a way to distinguish activity, feedback, evidence, and actual decision-changing signal. They look similar during the grind, but they are very different.

This matches what I hit this week trying to post about a small $19 guide I put together — the post just disappeared into a subreddit where I had zero history, no traction at all. Your line about broadcast only working once you already have an audience is the part that stings but is obviously true. Going to try your approach and actually talk to a handful of people instead of spraying it across more places.

The focus on AI detection scores is a unique value prop, but it might be worth reconsidering if that's the primary pain point your target users are willing to pay for. It proves the tool works, but does it solve the core business problem of booking more meetings or closing more deals? I suggest picking 5 leads, manually generating sequences for them using your tool, and offering them the output for free in exchange for a quick call. It's much easier to convert someone when you've already delivered the value and can prove it directly impacts their bottom line, rather than just showing them a lower detection percentage.

This is one of the more useful launch post-mortems I've seen because you separated activity from actual distribution.

The biggest takeaway for me is the distinction between broadcast distribution and relationship-driven distribution. Posting everywhere can create the illusion of momentum because you're constantly shipping content, but if the people seeing it aren't your ICP or don't have enough context to trust you, it rarely converts.

The 24,000-member community conversation is particularly interesting. One person with genuine access to your target users can be worth more than hundreds of directory submissions.

I also think there's a second lesson here: when someone gives you feedback and you ship it immediately, the conversation changes from try my SaaS to let's solve this problem together. That seems much more valuable at this stage.

So much respect for insights and sharing the raw numbers. It’s a great reminder that distribution without targeting or an existing audience is just noise. Focus on those direct conversations - that’s where the real signal is

Your directory line is the one I can put a number against, because I ran that experiment this week on one of the platforms you listed.

Launched a dev tool on Peerlist's weekly board on Monday. Over the full week it drew 34 views, 10 upvotes and zero comments, ending around rank 150. Not a single signup I can attribute, and worth being precise about why: the project's website field was the bare domain with no campaign parameters, set weeks before I thought about tracking, so if anyone had converted it would have landed in direct anyway. "Zero I can trace" and "zero" are different claims, and on directories most people only ever have the first one.

Which I think supports rather than softens your conclusion. A directory row is a permanent, indexed page that costs an hour once. It's a decent artifact and a bad channel, and the mistake is budgeting it as distribution. Your beta-tester post landing 12 people on the site at once did more in a day than 22 listings did in three weeks, and that ratio is the actual finding here.

sadly the link is 404: NOT_FOUND, but you can make a redirect or edit the link on this post.

The line that stuck: posting daily gives you something to do, it does not give you customers. Without an audience, broadcast channels mostly measure activity, and your own numbers already show where the signal was (same day community push, then real 1:1 demos). If I were planning the next 23 days I would cut the channel list hard and only keep experiments that can answer one question: how many people in the actual ICP saw this, and of those, how many tried it.

The core lesson is right: broadcast only amplifies existing reach. Without an existing audience, content is a slow drip into a dry channel. That observation is worth writing larger.

But I'd push back slightly on the framing. "Find 10 people to have real conversations with" is correct but incomplete. The better version is: find 10 people who each know 50 of your exact customers. The community admin with 24,000 members in your target market isn't a warm conversation — that's the whole distribution strategy. If you close that person, not as a customer but as a partner, you've unlocked what 6 months of content couldn't. That's not a sale, it's a multiplier.

The irony I'd flag: you built a product that improves cold outreach, and the data you just shared proves that warm relationships outperform cold sequences for early-stage distribution. That's not a contradiction — that's your entire case study and it belongs in your marketing.

"We tested 22 channels including the product we built. Direct warm conversations won. Here's why that changes how we think about sequences." That post would drive exactly the audience that needs what you've built.

Building Genie 007, I hit the same wall at day 30. The turn came when I stopped posting content and started having conversations with the people who already had the audience I needed. How close is the agency owner to becoming a channel partner rather than just a customer?

The "zero traction across 22 channels" number tells a specific story - it's not a distribution problem, it's a targeting clarity problem that distribution just made visible.

When you post a product that solves "cold outreach at scale" on Instagram Reels and Threads, you're reaching people who aren't building companies, aren't hiring, aren't running sales teams. They see it, know it's not for them in 3 seconds, keep scrolling. That's not a funnel leak - that's correct filtering.

The real question isn't "which channel works." It's "of the people who actually raise, hire, and run sales processes, how many saw this?" That's your real denominator. If you reached 50k founders and got zero signups, that's a product clarity problem. If you reached 2k people who run sales (even across 22 channels), and got zero signups, that's a targeting problem revealing a product problem.

The measurement system you built (channels tried, daily posts, directories submitted to) is measuring distribution effort, not distribution effectiveness. What you need to measure is: "How many people in my actual ICP saw this, and of those, how many tried it?" That number would tell you whether the problem is reach, targeting, or product.

You’ve collected a lot of signal across different channels in only 23 days.

What made you decide that the direct conversations were the signal to follow rather than just another early distribution experiment?