Your Ideal Customer Profile Is Probably Too Broad

When a campaign underperforms, the instinct is to rewrite the message. It is the visible part, so it feels like the lever.

It is almost never the message. A good message to the wrong list fails, and a mediocre message to exactly the right list works. Targeting is where the outcome is decided, and it is settled before anything is written.

Why owners resist narrowing

Every business owner has been told to niche down and most instinctively resist, for a reason that is entirely rational.

You can serve a wide range of clients. You have done good work across several industries. Narrowing feels like turning away revenue, and in a slow quarter that is uncomfortable.

But defining who you target in outreach is not the same as deciding who you will accept. Referrals and inbound still arrive from anywhere. Outreach is the one channel where you choose, and choosing badly wastes the whole budget.

Build it from your own data

Most profiles are built from opinion. Somebody in a meeting says the target is professional services firms between twenty and two hundred people, and that becomes the criteria.

The better version comes from the deals you already closed. Look at the last twenty and ask what the good ones have in common. Not the ones with the largest invoice, the ones that were straightforward to sell, profitable to deliver, and stayed.

The patterns are usually more specific and more surprising than the assumption. It might be a stage rather than a size, a particular trigger event, a certain tool already in use, or a structural situation like having two locations and no central system.

The criteria that actually predict

Industry and headcount are the easiest to filter on and among the weakest predictors. More useful:

• A trigger event. Recent funding, a new location, a leadership hire, a run of job postings for the same role. Something changed, which is when businesses buy.

• Observable evidence of the problem. Hiring three coordinators suggests a coordination problem. A careers page full of admin roles says more than a headcount band.

• Structural fit. Whether their model actually needs what you do, rather than whether they could theoretically use it.

• Who signs. A profile that does not name the role that decides produces conversations with people who cannot buy.

Write down who you are not for

The exclusion list is more useful than the inclusion list and almost nobody has one.

Too small to afford it. Too large to move quickly. In a sector where you have lost three in a row. Looking for the cheapest option. Writing this down stops arguments internally and stops your outreach reaching people who will consume time and not buy.

Where AI changes what is possible

Good targeting has always meant researching each company properly, which is why nobody did it beyond the first forty prospects.

AI reads at a scale that changes the economics. It goes through a company’s site, its careers page, its announcements and its public material, and judges against criteria you set. Not just does this company match a size band, but does the evidence suggest they have the problem you solve, right now.

It also works backwards. Given your closed deals, it can find what those companies genuinely share, including patterns nobody in the business had noticed, and then find more of them.

That turns a profile from an opinion written in a meeting into something derived from your own results and applied at volume.

Test narrow, then widen

The practical approach is to run the tightest definition first, on a small list. If it works, widen one criterion at a time and watch what happens to reply quality.

Starting broad and narrowing later almost never happens, because a broad campaign produces enough activity to look busy while producing nothing worth having.

What to measure

• Positive replies as a share of all replies. The clearest signal of targeting quality.

• Booked to held rate. Wrong fit prospects book and do not show.

• Close rate by segment. Compare against inbound to see whether outreach is reaching the same quality of buyer.

• Time to close by segment. Your best segment is usually also your fastest.

Interview five customers

The fastest route to a real profile is not analysis. It is asking.

Pick five clients you would want ten more of and ask three questions. What was happening in the business when you started looking. What else did you consider. What nearly stopped you buying.

The first answer gives you the trigger event, which is the highest value targeting criterion available and almost never appears in a profile written in a meeting. The second tells you who you are actually competing with, which is usually not who you assumed. The third gives you the objection your outreach should pre empt.

Half a day of calls produces a better profile than a week of analysis, because the pattern you are looking for is in the situation rather than in the firmographics.

Segments should be different messages, not different lists

Once the profile is tight, the useful next step is not to widen it but to split it.

The same service sold to a construction firm and to a professional practice solves the same underlying problem and is described completely differently. Same offer, different language, different examples, different objection to handle first.

That is what segmentation should mean. Most businesses use it to mean different lists receiving the same message, which delivers none of the benefit and all of the administrative cost.

Revisit it every six months

A profile is a snapshot of what worked, and what works moves.

Your capability changes, the market changes, and the segment that was ideal two years ago may now be the one with the longest sales cycle and the thinnest margin. Rerun the exercise against the most recent twenty closed deals rather than against the original assumption, and be willing to find that your best segment is not the one you built the business around.

AI Optimize builds the profile from your closed deals, verifies every prospect against it, and runs the outreach on top. That work sits under Cold Email Outreach and LinkedIn Outreach.