Synopsis
If your marketing isn’t working, the problem almost never started with the marketing. It started earlier, in a decision nobody made: who exactly this company sells to, and what it is for. Every campaign, post, ad and AI-generated paragraph that came afterwards was built on a guess.
This is the post I wish someone had handed me fifteen years ago. It explains why “more” doesn’t fix marketing that isn’t working, what the skipped decision actually is, and how to make it before you spend another dollar.
The Three Things Everyone Tries First
When marketing isn’t producing results, most companies reach for one of three fixes, in roughly this order.
More content. If two posts a week didn’t work, maybe five will. The calendar fills up, the team is busier, and the numbers don’t move — because the audience that wasn’t responding to two posts isn’t going to respond to five of the same.
A new agency. The last one didn’t deliver, so a different one will. Sometimes that’s true. More often, the new agency inherits the same missing foundation and produces a fresher-looking version of the same problem.
An AI tool. This is the 2025 version of the first fix. The content comes out fast, and it comes out sounding like everyone else’s — because the tool was given nothing specific to work from. It did exactly what it was asked.
None of these are stupid. They’re the obvious moves. They fail for the same reason: they add output on top of a question that was never answered. And adding output to an unanswered question doesn’t clarify it. It makes the confusion faster.
The Decision That Got Skipped
Here is the question, in the plainest words I can find: who is actually buying from you, and why?
Not who you’d like to buy. Not the segment on the pitch deck. The real person who signed the last twenty contracts, what was going wrong in their business the week they went looking, what they typed into Google, what they said on the first call, and what finally made them trust you enough to pay.
Most growth-stage companies I meet can’t answer this on paper. The founder can answer it — roughly, from memory, differently each time — and that’s the problem. If the answer lives only in one person’s head, then sales describes the customer one way, marketing another, the website a third, and every new hire gets a fourth version in their first week.
That’s what “marketing not working” usually looks like from the inside. It isn’t a skills problem. It’s an alignment problem, and it traces back to one skipped decision.
The second half of the decision is just as unglamorous: what is this company for? One sentence that every person in the building would say the same way. Not a mission statement. A promise. If your sales team and your homepage would answer differently, you don’t have one yet.
What Happens When You Make It
I’ve watched the same thing happen enough times to trust it. When a company finally writes down who the customer is and what the promise is — really writes it down, with the buyer’s own words in it — three things change within a quarter.
The website gets shorter, because you stop hedging for people who were never going to buy. Sales calls get faster, because the prospect hears their own problem described back to them in the first two minutes. And the content, whether a person or a tool produced it, starts sounding like your company instead of the category, because it finally has something specific underneath it.
That last point matters more every year. AI is excellent at producing and hopeless at choosing. Give it a clear customer and a clear promise, and it will write usable first drafts all day. Give it nothing, and it will faithfully produce the average of the internet. The tool doesn’t need to be better. The foundation does.
“Marketing not working” is almost never a marketing problem. It’s a decision that hasn’t been made yet.
How to Make the Decision in a Week
You don’t need a workshop or a consultant to start. You need a week and some honesty.
- Day one: list the last twenty deals you won and the last ten you lost. For each, write one line: what was going wrong for them when they came looking.
- Day two: read your last thirty sales-call notes or emails, and copy out the exact phrases customers used to describe the problem. Not your words. Theirs.
- Day three: look for the pattern. Usually two or three kinds of buyer show up, and one of them is clearly the best — closes faster, stays longer, refers others. That one is your customer.
- Day four: write the promise. One sentence, in their language, that tells that buyer what changes when they work with you.
- Day five: read your homepage, your LinkedIn page and your sales deck against that sentence. Everything that doesn’t serve it is a candidate for deletion.
That’s the whole thing. It’s the first week of every engagement I run — I call it a marketing audit, because that’s what people search for, but it’s really just this decision, made properly and written down. Everything else in marketing is execution, and execution is easy to buy. The decision isn’t.
Key Takeaways
- When marketing isn’t working, more content, a new agency or an AI tool add output on top of a question nobody answered — and make the confusion faster.
- The skipped decision is two questions: who is actually buying from you and why, and what is this company for. Both must be written down in the buyer’s own words.
- Make the decision in a week from your own won and lost deals, then read every piece of marketing against it. The tools don’t need to be better. The foundation does.
FAQ
Consistency multiplies whatever is underneath it. If the message was built on a guess about the customer, posting more often produces more of the same result. Check whether sales, marketing and the website describe the customer the same way — if they don’t, that’s the problem, not the frequency.
Both, and they come from the same source. Strategy decides who you’re for and what you promise; messaging is how you say it. When the first decision hasn’t been made, the messaging can’t be fixed by rewriting it — it has nothing to be consistent with.
AI can produce faster, more consistent content once the customer and the promise are defined. It cannot make that decision for you. Used before the foundation exists, it produces generic output at scale.
Start with a marketing audit — a short, bounded look at where your message breaks between your website, your sales calls and your team. At Octopus & Son it’s the first week of every Marketing Foundation engagement, and it’s a 20-minute call to find out if it’s the right fit.
