Set up a keyword alert for "sales pipeline" or "cold outreach" and the posts start arriving. Some are from VPs of Sales at companies that look exactly like your customers. Some are from recruiters hiring for a sales role. Some are from a student writing up a class project. All three match the keyword. Only one is worth your time.
This is the blind spot in most LinkedIn engagement tools and most manual routines. They answer "what is this post about" and skip "who wrote it". If you use LinkedIn to find customers, the second question is the one that matters.
What is the prospect-match problem?
The prospect-match problem is the gap between posts that mention your topic and posts written by people who could buy from you. Keyword search fills your feed with the first kind. You want the second kind. Every post in the gap costs you time to open, read and reject.
It is a matching problem, not a volume problem. Finding more posts does not help. Finding the same number of posts from the right authors does.
Why do keyword alerts return the wrong people?
Keyword alerts return the wrong people because a word has no job title. The same phrase means different things depending on who typed it. Here is what one alert for "sales pipeline" typically pulls in:
| Who wrote the post | What they meant | Worth a comment? |
|---|---|---|
| VP of Sales at a software company | Their deals are stalling at one stage | Yes, this is the buyer |
| Recruiter | They have candidates in a hiring pipeline for sales roles | No |
| Sales coach | A tip, written to attract the same buyers you want | No, a competitor for attention |
| Student | A summary of a course module | No |
| CRM vendor | A product announcement | No |
| Founder, 8 employees | They do all the selling themselves and it is a mess | Maybe, depends who you sell to |
One match in six is a buyer. The other five are not noise in the sense of being off-topic. They are exactly on topic. That is what makes them expensive: you have to read each one to find out.
What does keyword-only searching cost you?
It costs you the time you set aside for engaging. You open the list, see a post that matched, read it, check the author, realize they are a recruiter, and move on. Do that a dozen times and a quick fifteen-minute check has become forty minutes, with most of it spent filtering.
There is a second cost that is easier to miss. When you do comment on a wrong-fit post, you build familiarity with someone who will never buy. The comment may be good. It is still pointed at nobody. We cover what a well-aimed comment does in how B2B sales leaders turn comments into warm leads.
How do you find LinkedIn posts from your ideal customers?
Reverse the order. Start with the person, then the topic.
- List the job titles your customers put on their profiles. Use their words, not yours. "Head of Talent" and "VP People" are different searches from "HR".
- Search for posts by those titles. This is the step keyword alerts skip.
- Rank what comes back by topic. Now keywords are useful, because every result is already a possible buyer.
- Remove the lookalikes. Add the titles that keep slipping in, such as students, consultants who sell to the same buyer, and vendors.
- Narrow by place if it matters. If you only sell in one country, check where the author is based.
- Keep only fresh posts. A comment on a post from last month reaches nobody.
Done by hand, this is a profile check on every post. That is the work a tool should take off you.
How should you choose job titles and keywords?
Titles decide who you reach. Keywords decide which of their posts you see first. The most common mistake is writing keywords that describe what you sell.
| You sell | Weak keyword (what you sell) | Strong keyword (what the buyer says) |
|---|---|---|
| Recruiting services | "recruiting" | "time to hire" |
| Bookkeeping | "bookkeeping services" | "month end close" |
| Sales training | "sales coaching" | "ramp time" |
| Paid media | "Facebook ads agency" | "cost per lead" |
A weak keyword finds your competitors, because they are the ones posting about the service. A strong keyword finds buyers describing the problem. Three to six good keywords beat ten vague ones. Allen's guide on keywords that work has more examples, and worked audience examples shows full setups for recruiters, accountants, agencies, coaches, financial advisors and B2B sales.
How does Allen match posts to prospects?
Allen searches by author first. Each setting on its Prospects page does one job:
- Job titles find people. Allen looks for posts written by people who hold the titles you list. This is the most important setting.
- Keywords rank posts. You can save up to 10. They do not change who Allen searches for. They help it pick which of those people's posts are worth a comment.
- Target country filters by location. Set one and Allen checks each author's real location before showing you the post.
- Industries nudge the ranking toward people in those fields.
- Minimum followers sets a floor, so your comment is seen by a bigger audience if you want that.
- People to Skip and Words to Skip remove the wrong authors and the topics you never want to touch.
Every post then gets a fit score for how well the author matches your audience. Posts with a reasonable fit show up on your Today page. Autopilot is stricter and only posts on strong matches.
Allen only looks at posts from the past week, and it remembers who you commented on recently so you do not appear to follow one person around. Public posts come from third-party data providers, so nothing is read through your own LinkedIn account. The full explanation is in how Allen decides which posts to show you.
Can you skip keywords completely?
Yes. Once the search starts from job titles, keywords become optional. In Allen you can search a saved audience "talking about anything", which ignores keywords and returns your buyers' freshest posts on any subject.
This matters more than it sounds. Buyers do not post about your category very often. A dental practice owner might post about hiring, a conference, or a new location long before they post about the thing you sell. Each of those is a fine place to be seen. If you wait for the exact keyword, you wait a long time.
What if you want specific people, not a type of person?
Use a watchlist. Some sales depend on a short list of named accounts, and no title search is precise enough for that. Allen lets you add named people from its Connections page, upload your LinkedIn connections as a file, or add anyone from their profile with the Chrome extension. It then checks those people's posts on every run. The help guide on using the watchlist as a dream 100 shows how to set that up.
How do you know your targeting is working?
Look at the first ten posts in your list and count how many authors you would be glad to have a sales conversation with. Under five, the titles are too broad. If the list is nearly empty, they are too tight.
Then watch replies. Allen's own guidance is to give a new setup twenty weekdays, and it notes that zero replies by week three almost always means the audience is wrong, not the comments. If the wrong industry keeps showing up, the fix is nearly always in job titles and skip lists. See fixing wrong-industry results.
If your routine involves a lot of scrolling past posts that do not quite fit, that is the keyword problem showing up in real time. The fix is not more posts. It is the right authors. You can see a list built around your buyers during Allen's 7-day free trial.