How to find businesses that need marketing help

A business worth pitching is one whose marketing is visibly underperforming what it already has, and you can spot that from a public profile in about five minutes. Four things tell you: how far a post travels, whether anyone replies, how often they post, and whether they already pay somebody for promotion. No dashboard access, no tools, no conversation.

This is the manual version of what our engine does before it writes anything. Run it by hand a few times first, because the numbers stop being abstract the moment you have pulled them yourself.

What "needs marketing help" actually means

Not "they are bad at marketing". That question has no answer you can act on, and a cold message built on it is an insult with a price list attached. The useful question is narrower: is there a gap between the audience they already have and what that audience is doing?

A company with forty thousand followers and four replies a post has a problem they can already feel. That is the difference between a prospect and a stranger: you are not telling them something new, you are naming something that has been annoying them for months. A company with no audience at all usually has no budget either, which is why the second half of this guide is about telling those two apart.

The four checks

1. Reach against audience

Open their last ten posts and look at views. Divide by the follower count. Anything under about five percent means most of their own audience is not being shown their posts, which on most platforms is a signal the account is being suppressed for low engagement rather than a mystery.

Write the number down but do not put it in your email as a percentage. "Your posts are reaching about three in a hundred of your followers" lands; "3.1% reach rate" reads like an export.

2. Replies, separately from likes

Count replies on the same ten posts, and count them separately from likes. This is the check most people skip and it is the most diagnostic of the four. Likes are cheap and can be bought. Replies mean somebody had a thought. An account pulling three hundred likes and two replies has an audience that is watching and not engaging, which is a different and more fixable problem than not having an audience.

Our rule, learned by sending bad messages: never point at a likes count above ten as the problem. The founder knows twelve likes is not a catastrophe, and saying it is costs you the rest of the message.

3. Cadence, and the gaps in it

Look at the dates, not the average. An average hides the thing you want: a feed that posted daily through a launch and then went quiet for six weeks is a team that ran out of capacity, and that is a specific, true, sellable observation. "About one post every few weeks" is how a person says this. "0.2 posts per week" is how a tool says it.

4. Whether they already pay for distribution

Check for paid posts, sponsored mentions, or influencer placements. This changes the entire conversation. A company already buying promotion has a budget and has made the decision that distribution is worth paying for. You are not selling them the idea, you are competing on execution, which is a much shorter sale.

Reading the four together

The combinations matter more than any single number:

• Good reach, few replies. The audience exists and is not being activated. The strongest case to sell into, and the easiest to say kindly.
• Poor reach, good replies. A small engaged core. Growth problem, not a content problem. Do not tell them their content is bad, because it is not.
• Both poor, posting constantly. Effort is going in and nothing is coming out. Usually the most receptive of the four, and the most demoralised.
• Both poor, barely posting. Nobody is working on it. Often no budget and no owner internally, which makes it a slow sale however obvious the problem looks.

What not to conclude

Everything above is visible from outside, and that is the limit of it. You cannot see whether they have an agency already, whether a rebrand is two weeks out, whether the quiet stretch was a legal hold, or whether the person who ran the account left. Guessing at any of it in a first message is how a good observation becomes an insulting one.

The discipline that makes this work is the same one we hard-coded into the engine: if you did not see it, you do not say it. State the four things you actually counted, and stop.