The channel you love is not the channel your buyers are in
Sponsoring the community you belong to feels like marketing. It is usually a way of spending your marketing budget among people who are not your customers, and the tell is a question you can ask in ten seconds.
A software company was sponsoring events for the framework its product is built in. The team are part of that community, they enjoy it, the sponsorships were not cheap, and the whole thing looked exactly like marketing. Then somebody asked the obvious question about the people in the room:
If I polled them and said "how many of you identify primarily as a Laravel developer" — it's not gonna be most of them, it's probably not gonna be even five percent of them. So I don't understand why you're fishing in that pond.
Jason Cohen
That is not where the fish are. So this is not helping. It's actually removing time that you could be spending developing relationships and brand and trust where the fish are.
Jason Cohen
The second sentence is the expensive part. A channel that does not work is not neutral — it consumes the hours that the working channel needed.
The question that exposes it
For any channel you are spending on, write down the poll you would run in that room:
"How many of you are the person who decides about {the thing we sell}?"
Then estimate the honest answer. Not "could they benefit" — are they the buyer. Under about a fifth and you are subsidising an audience, however warm it feels.
Affinity is a genuinely good reason to be somewhere. It is not a marketing reason, and the two get confused because the channel chosen from affinity produces the most enjoyable feedback: friendly conversations, people who already know your name, nobody asking hard questions. Buyer channels feel worse and convert better.
The concession that keeps this honest
The rule is not "cut everything you enjoy":
How much time and energy can you afford on things that you just love and give you life? And the answer is: not zero.
Jason Cohen
So run a budgeted column. Give the community you belong to an explicit, small, capped share of your time — and stop calling it pipeline. Once it is labelled correctly, it stops competing with the work that produces customers, and you stop feeling guilty about the part of it you should keep.
The menu you are actually choosing from
It is a shorter list than the internet suggests. The approaches that reliably work in B2B software:
Typical B2B SaaS marketing approaches can work. I call them my big five: content, SEO, cold outreach, integrations and partnerships — and then of course, advertising.
Rob Walling
Six things. Not thirty. And the selection question is about your buyers, not about your preferences:
Which ones will reach my customers? Where are my customers? Where do they exist? How can I reach them with marketing? I narrow that list down to the top two, three or four.
Rob Walling
Two to four, chosen from six, on the basis of where the buyers are. Most struggling go-to-market plans fail one of two ways: they run one channel and call it a strategy, or they run all six badly.
Which ad platform your knowledge licenses
If advertising is one of your two to four, the platform is not a taste question. It is decided by what you actually know about the buyer:
| If you know… | Use | With the caveat |
|---|---|---|
| The buyer's job title | The expensive one — "tougher to make work, you have to do a lot more manual stuff" | |
| Demographics or psychographics | Cheap reach, weak intent — the creative does the qualifying | |
| Real search intent, with volume | Only if the searches exist; check volume before you write a word |
Two failure modes follow directly. Running Google ads for a problem nobody searches for spends money against no demand — if buyers do not know the category exists, they are not typing it. And running LinkedIn ads without a precise title filter pays LinkedIn's premium and throws away the only thing you are paying it for.
Where the fish are, when the pond is small
The uncomfortable case is a buyer who is not clustered anywhere: no conference, no subreddit, no obvious list. This is more common than the channel literature admits, and it has a specific answer rather than a shrug. When there is no pond, the channel is one at a time — outbound to named companies, and introductions through people who already know them. That is not a fallback for teams too small to do real marketing. For a high-value offer it is often the highest-converting channel available, and it is the one you fully control.
Four things to do
- List every channel you spend time or money on, and write the poll question next to each with your honest estimate. Anything under a fifth moves.
- Pick two to four from the six. Write down why each reaches your buyer. "We should be on X" without a reason is affinity wearing a strategy costume.
- Label your fun column and cap it. Keep it. Stop counting it as pipeline.
- Check what you know about the buyer before choosing an ad platform — title, demographics, or search intent. If the answer is none of the three, you are not ready to buy ads; you are ready to go and talk to twenty of them.
Keep reading
- The audience you would spend a year building is worth less than ten emails Three of the most successful bootstrapped founders in software launched to audiences of twenty to thirty thousand people. Here is what those audiences produced, and what they did instead when it did not work.
- The customer who cannot leave, and the one who was always going to Some customers churn because the product failed them. Others churn because they finished. Those are not the same problem, and averaging them together hides the only segmentation decision that matters.
- Narrowing your market is how you get the customers outside it The objection to niching down is always the same: we would be turning away everyone else. The observed effect is the opposite, and the mechanism is not the one most people assume.