From $0 inbound to $1.42M pipeline in 6 months

When Loomwise AI first reached out to me, they were in the spot a lot of really good applied-AI startups land in, which is that the product genuinely worked, the early customers genuinely loved it, the retention numbers were genuinely strong, and almost nobody outside their existing accounts had any idea they existed. They had raised a seed round of about $3.1M roughly ten months prior, they had eleven people on the team, seven of whom were engineers, and they were selling a workflow-automation copilot built specifically for product teams, the kind of tool a head of product or a senior PM would use to turn messy roadmaps and scattered Slack threads and half-written PRDs into something structured and shippable, and the thing about that buyer is that they are smart, they are skeptical, they have been burned by AI demos that fall apart in week two, and they do not respond to noise. So the challenge was never the product, the challenge was distribution, and more specifically the challenge was that their entire go-to-market depended on two motions that had both quietly stalled out.

The first motion was founder-led outbound, which had carried them from zero to about $640K in annual recurring revenue, and it had stalled because the founder was spending close to twenty hours a week writing cold emails and doing demos and that obviously does not scale past a certain point, and the reply rates had been sliding from a respectable 4.2% down to about 1.8% as the inboxes of every product leader on earth filled up with AI cold-email spam. The second motion was paid acquisition, where they were spending around $18K a month across LinkedIn and Google, pulling a blended customer acquisition cost of about $9,400 per closed customer against an average contract value of roughly $14,000 a year, and that math works for exactly as long as your runway lets it work, which in their case was not very long, because a CAC payback period north of eight months on a seed-stage balance sheet is the kind of number that makes the next board meeting tense.

Underneath both of those problems was a deeper one, and it is the problem I see with almost every technical-founder-led startup, which is that they had an enormous amount of genuine expertise locked inside their heads and inside their codebase and inside their customer calls, and absolutely none of it was leaving the building. The founder could talk for forty-five minutes about why most AI copilots fail at multi-step product reasoning, he could explain exactly how they handle context windows and hallucination guardrails and the specific evaluation harness they built, and that explanation was the single most compelling sales asset the company owned, and it lived nowhere. It was not on YouTube, it was not on LinkedIn, it was not in a single piece of written content, it was not findable by the product leader who at that very moment was googling whether an AI copilot could actually be trusted with their roadmap. So they were invisible at the exact moment of highest intent, and they were paying $9,400 to buy attention they could have earned, and the pipeline coverage going into the quarter was about 1.4x against a target that needed to be closer to 3.5x, and that gap, that specific and very expensive gap between what they knew and what the market could see, was the thing I was hired to close.

My whole approach with Loomwise started from the same place it always starts, which is that I do not believe a startup with one credible founder and a working product needs to post five times a day across nine platforms and burn itself out chasing an algorithm, I believe it needs one really good shoot a month turned into thirty-plus platform-native assets that get distributed everywhere the buyer already spends time, so that the expertise compounds and the leads start arriving already warm, and that is exactly the engine we built. The way I describe it to founders is that we are not making content, we are making a distribution machine, and the content is just the fuel, right, so the first thing I did was sit the founder down for what I call the source shoot, which is a single focused half-day where we capture the raw expertise on camera, and from that one session we pull the entire month of assets, and that is the unlock, because the founder gives me four hours and I give back thirty-plus pieces, and his time cost stays flat while the output compounds.

Before we shot a single frame though, I spent the first two weeks doing what I think most agencies skip, which is the positioning and message work, because if you distribute the wrong message efficiently you just lose money faster. I went through forty-one of their recorded sales calls, I read every churned-deal note, I interviewed six of their happiest customers, and I pulled out the exact language product leaders used when they described the pain, and what kept coming up was not features, it was trust and it was time, it was product leaders saying they did not want another tool to babysit and they did not trust AI with anything customer-facing without a human in the loop, and so the entire content thesis we built was around earned trust in applied AI for product teams, which conveniently was the thing the founder was best in the world at talking about. That became the spine, and every asset hung off it.

Then we ran the monthly shoot. One shoot, half a day, and out of each shoot I produced a long-form anchor piece, which was usually an eighteen to twenty-six minute deep explainer or a recorded teardown of a real product problem, and then from that anchor I cut the spine into platform-native assets, so we are talking eight to twelve short vertical clips for LinkedIn and shorts, four to six written LinkedIn text posts that stand on their own and do not just beg people to click away, two long-form written breakdowns repurposed for the company blog and for a Substack-style newsletter, a handful of carousel posts that worked the same idea visually, and a set of quote-cards and audiograms, and every one of those was built to live natively on its platform rather than being one asset awkwardly stretched across all of them. That is the part that matters, because a vertical clip that was clearly cut for vertical performs three to five times better than a landscape video squeezed into a phone, and over six months that multiple is the difference between invisible and unavoidable.

Distribution was the other half, and this is where I do not let the work just sit on a publish button and hope. We posted the founder-led assets from his personal LinkedIn, because in B2B the human face dramatically outperforms the brand page, and we cross-posted a brand-voice variant from the company page, and we seeded every long-form piece into the three Slack and Discord communities where their buyers actually congregated, and we set up a lightweight newsletter so that the people who engaged once got a reason to come back, and critically we tagged and tracked everything with UTM parameters and a simple attribution sheet so that when a demo got booked we could trace it back to the exact asset that warmed them up. I also rewired their intake so that every piece of content pointed not at a generic homepage but at a specific resource or a soft-CTA that captured intent without scaring off the skeptical buyer, because product leaders do not book a demo off one video, they lurk, they read three things, they check you out, and then they raise their hand, and the job of the engine is to be there for all three of those touches.

The last piece of the approach was the feedback loop, because the whole point of doing one disciplined shoot a month is that you get a clean signal on what is working, so every two weeks I reviewed which assets drove saves and comments and profile visits and actual booked calls, and I fed that straight back into the next shoot brief, so by month three we were not guessing what the buyer wanted to hear, we knew, and the founder was walking into the source shoot with a list of the exact questions his market was asking, and that is when the numbers really started to bend.

Six months in, the way I judge whether the engine worked is not whether the views looked nice on a slide, it is whether the money moved, and with Loomwise the money moved hard, so let me walk through it the way I walked the founder through it on our final review call. We started from a company that had logged essentially zero warm inbound in its entire history, three demos a month if you were generous and most of those were referrals, and by month six the engine was producing 121 qualified inbound demos a month, all of them people who had consumed multiple pieces of the founder's content before they ever raised their hand, which is the entire game, because a demo with someone who already trusts you closes at a completely different rate than a demo with a cold prospect. On the cumulative pipeline number we crossed $1.42M generated across the six months, and I want to be precise about what qualified means here, because it does not mean a form fill, it means a fit-checked opportunity that sales accepted into the pipeline with a real budget and a real timeline, and that pipeline converted to $486K in closed-won revenue cleanly attributed back to content through the UTM and source tracking we set up on day one.

The ROI math is the part I care most about, because anybody can spend money and generate activity, the question is whether the engine returns more than it costs, and on a total engagement investment of $43,000 across the six months, $486K in closed revenue is an 11.3x return, and that is before you count the pipeline that had not closed yet, which on a normal sales cycle will keep converting for another two quarters off content that is already published and already working. The blended ROAS landed at 11:1, and the reason I call it blended is that this includes the paid spend we kept running, except by month five that paid spend was not buying cold attention anymore, it was amplifying the organic assets that had already proven they converted, and that single shift, moving from cold paid to amplified-organic paid, is most of why the customer acquisition cost fell so far.

That CAC number is the one the board cared about most, and it is the one I am proudest of, because we took blended CAC from $9,400 down to $3,950 by month six, which is a 58% reduction, and the mechanism is not magic, it is just that when a buyer arrives warm off six touches of founder content, the demo-to-close rate roughly doubles and the sales cycle compresses, so you spend less to acquire each customer and you acquire them faster, and on a seed-stage balance sheet that change is the difference between a runway you are nervous about and a runway that lets you raise on your own terms. The CAC payback period that had been north of eight months came down to right around three and a half months, and that number alone reset the company's entire growth model.

The reach and engagement numbers tell the top-of-funnel story that makes all of the above possible, because monthly content reach went from about 9,000, which was basically the founder's existing network seeing the occasional post, up to 742,000 in month six, an 82x increase, and more importantly that reach was concentrated almost entirely on the exact buyer, product leaders at venture-backed software companies, because we never chased broad reach, we chased the right reach. Engagement scaled with it, the founder's LinkedIn following went from roughly 2,100 to just over 19,000 genuinely relevant followers, his posts started getting saved and shared by the precise people who sign the contracts, and he started getting tagged into community discussions unprompted, which is the moment you know the category presence is real, because the market is now bringing you into conversations instead of you having to buy your way in.

There is one more result that does not show up cleanly in a chart but mattered enormously, which is that the founder got his time back, because the engine replaced close to twenty hours a week of manual outbound with a four-hour monthly shoot, and that reclaimed time went straight back into product and into closing the warm demos the engine was now generating, so the whole company got more leveraged, not just the marketing line. And the outbound that did continue got dramatically more effective on its own, because reply rates on the small amount of targeted outbound the team still ran recovered from 1.8% to 6.1%, since prospects now recognized the name before the email landed, and a cold email from someone whose teardown you watched last week is not really a cold email anymore. Put all of it together and the picture is a company that walked in invisible and dependent on a founder grinding outbound, and walked out six months later with a self-sustaining distribution flywheel, $1.42M in pipeline, $486K closed, and a CAC cut by more than half, and that is what I mean when I say the leads arrive warm.

If I zoom out from the specific Loomwise numbers, the reason this worked is the same reason it works every time a founder has real expertise and no distribution, and it is worth saying plainly, because a lot of startups try to solve a distribution problem by making more content when they actually have a leverage problem, right, they think the answer is volume when the answer is compounding. Loomwise did not need to be louder, they needed the expertise that already existed inside the founder's head to become findable at the moment of intent, and the entire engine I build is designed around that one idea, which is that you capture the source material once a month in a single disciplined shoot, you turn it into thirty-plus platform-native assets, and you distribute those everywhere the buyer already is so they compound on top of each other instead of disappearing in a feed.

The thing I want any founder reading this to take away is that the monthly cadence is a feature, not a compromise, because the discipline of one shoot a month forces clarity, it forces you to pick the few ideas that actually matter to your buyer, and it gives you a clean read on what is working before you commit the next round of effort, and that feedback loop is what bent the Loomwise numbers from month three onward, since by then we were not guessing, we were briefing each shoot around the exact questions the market was asking us. Most agencies cannot do this because they are set up to sell you hours and volume, and I am set up to sell you a machine that returns more than it costs, which is why the number I lead with is the 11.3x and not the view count.

The other piece that made this durable is that we built it to outlast the engagement, because I documented the entire playbook, the shoot framework, the asset breakdown, the distribution checklist, and the attribution dashboard, and I handed it over so the system keeps running on the founder's four hours a month whether I am in the room or not, and that matters because a distribution engine that collapses the day the agency leaves was never an engine, it was a dependency, and I do not build dependencies. Loomwise left the six months with a category presence their bigger competitors are now reacting to, a pipeline that keeps converting off content already published, and a founder who finally has his calendar back, and the assets we shot in month one are still pulling in demos, which is the compounding I keep talking about made concrete.

There is also a financial argument here that I think gets missed when founders only look at the cost of the engagement, because the real comparison is not whether $43,000 over six months is a lot of money, it is what that $43,000 bought against the alternatives, right, and the alternative for Loomwise was either hiring a full-time content and growth lead at a fully loaded cost north of $130K a year who would still take three months to ramp, or pouring another $108K into the same paid channels that were already producing a $9,400 CAC, and against either of those the engine returned $486K in closed revenue and $1.42M in pipeline while cutting the cost per customer by 58%, so the choice was not close, and that is before you account for the asset library being a durable balance-sheet item that keeps producing after the spend stops. Most paid acquisition is a rented audience that disappears the second you stop paying, and the assets we built are owned, they sit there compounding, and a piece that cost a fraction of one month's fee to produce in month one was still booking demos in month six at zero marginal cost, which is the kind of unit economics that actually changes how a company can raise and how it can grow.

So when a product leader now googles whether an AI copilot can be trusted with their roadmap, Loomwise is the name that shows up with a real answer instead of a sales pitch, and when that buyer eventually books a demo they arrive already half-sold, and that is the whole point of a distribution flywheel, you do the expensive trust-building work once, at scale, and then you let it pay you back over and over, and the math just keeps getting better the longer it runs.