To find leads on LinkedIn, define the companies and roles you want to reach, then review relevant people from competitor engagement, your own posts, creators, customer relationships and events. Record the source and date for each person, check their current role, and choose a useful next step. Public engagement shows attention; it does not establish buying intent.
The five methods below show where to look and how to review what you find. You can start with a small manual sample, then decide whether you need a maintained audience. The market examples retain their original research dates and sample sizes.
The five places your leads already are on LinkedIn
Start with one source and a manageable sample. Review profiles and available public activity manually, or work from records your team already holds. Record where each suggestion came from so another person can check it.
A checklist for qualifying each LinkedIn lead
Use one row per person. These fields help you decide whether to keep researching, follow up or remove the suggestion.
| Field | What to record |
|---|---|
| Person and company | Name, profile URL, current role and company |
| Account fit | Why the company matches your region, sector and use case |
| Observed activity | Source URL, date and what the person actually did |
| Relationship | Existing customer, shared contact or no known relationship |
| Next step | Useful question, introduction, further research or exclude |
A worked example: you sell onboarding software to mid-sized SaaS companies. An operations lead at a relevant account comments on a post about slow handovers. Save the post, confirm their role and read the comment. A reasonable first question is how their team handles handovers today. Do not claim that they are buying software or unhappy with a competitor. This is an illustrative example, not a customer result.
1. Count your competitors' engagers
Pick your six closest competitors. Open each company page, go to their posts, and for the last 90 days open the reaction list on every post. Write down the names, roles and companies of everyone whose title looks like a buyer or an influencer of the buy.
Two counts matter. The first is the raw crowd: how many distinct people reacted at all. The second is the rate: total market reactions divided by the number of posts that vendor published in the window. The rate is what tells you which competitor is actually holding attention, because a vendor posting five times a week will win on raw totals whether or not anyone cares.
Print the window and the sample on your own sheet. Ninety days, six competitors, this many posts. A count without a denominator is not a finding.
2. Count your own page's engagers
Same method, your own page and your team's personal posts. This is the fastest of the five and usually the most uncomfortable, because the overlap between people who engage you and people who engage your competitors is often small.
Do the personal posts separately from the page. Across every market we have pulled, founders and operators out-reach their company pages by roughly ten times. In an ecommerce customer-service AI market pulled in London in August 2026, one vendor's two founders reached 40 of the sampled market between them. The same vendor's company page reached 3. If you only count the page, you will conclude your reach is smaller than it is.
3. Count the voices your market follows
Take the list of named people from steps one and two. Open twenty of them at random and look at what each one has engaged with in the last month that was not a vendor post. Write down the author each time.
Tally the authors. The result is usually not what people expect. In a revenue-tooling market in the Nordics, pulled 27 August 2026 over a 90-day window with the 170 most active of 642 sampled people analysed, the top voice was Brian LaManna, an account executive at Gong who writes the newsletter "Closed Won", reaching 29 of 170. Then Abhijit Mitra, CEO of Outreach, at 24. Laurie Pratt Ehrbar, CMO of Clari and Salesloft, at 20. Amit Bendov, Gong co-founder, at 17. Steve Cox, CEO of Clari and Salesloft, at 16. Nadia Rashid, CRO of Outreach, at 16.
The person a market follows most is rarely a vendor CEO. In the London ecommerce pull, the single biggest voice was the CX Professionals Association, reaching 50 of the 170 most active people.
4. Count past customers who moved company
Export your closed-won list and your churned list, then open the LinkedIn profile of every named contact on both. Note who is no longer at the company you sold to.
Every one of those people is a lead with a fact attached. They know your product, they know whether it worked, and they are now inside a company you have no relationship with. This is the cheapest of the five counts and the one most teams have never run.
Sort the result by whether the person's new company is in your category. Someone who took your tool from a 40-person company to a 400-person one is a different conversation from someone who left the sector.
5. Count the rooms your market gathers in
Go back to your named people and look at what they have posted about attending, speaking at, or organising in the last year. Tally the conferences, the associations, the certifications and the recurring dates.
Count at company level, not person level. You want to know that eleven companies in your market show up around a given room. You do not want to claim that a specific person attended, because a public post is not proof of attendance and nobody sells a verified attendance list that is honest.
Rooms are not all conferences. CX Day, run by the CXPA and falling on Tuesday 6 October in 2026 as published on cxpaglobal.org in September 2026, is a date rather than a venue, and a whole profession posts on it.
What the count showed in the revenue-tooling market
Here is a complete result from that Nordics pull, dated 27 August 2026, 90-day window, 170 most active of 642 sampled, market built from competitor engagers and then title-filtered. Market reactions, then reactions per post:
- Gong: 327, 13.6 per post
- Outreach: 264, 11.0
- Salesloft: 99, 4.1
- Clari: 68, 2.8
- Revenue.io: 33, 1.4
- Sybill: 31, 2.8
- Avoma: 27, 1.1
- People.ai: 12, 0.5
- Attention: 0 from 24 posts
Attention published 24 posts in the window and got nothing from this market. That is stated as zero with the post count, because zero with a denominator is a finding and zero on its own is a typo.
Gong's best post was "the biggest sales call mistake" at 74 market reactions. Its post about AI discussions rising while teams are not shrinking pulled 29. The craft post beat the industry-trend post by more than two to one, which is the pattern in every market we have pulled so far.
One more thing the count surfaces: the raw crowd is not the market. In a UK pull of AI agent platforms on 20 August 2026, eight anchor accounts over 90 days produced 4,449 raw reactors. Filtered by title, that was 79 people who own a service or lead a team against 1,523 builders and practitioners. Per platform the ratio ran from 11 to 1 on Botpress to 87.7 to 1 on Gumloop. These title groups cover only part of the raw sample. They show why role review matters; they do not establish who has purchasing authority.
How to find champions inside a target account
A possible champion understands the problem and may help you reach the people involved in solving it. Build that hypothesis from an existing relationship, relevant responsibility or a conversation about the work.
Start with former customers, people introduced by a trusted contact and practitioners who can explain the problem. Confirm their current company and role. Then ask who else is involved and whether the proposed change matters to their team.
For example, a past customer who has moved into operations at a target account is a useful person to reconnect with. Ask whether the same problem exists there before assuming they can sponsor a purchase. Keep decision-makers, practitioners and potential champions as separate roles until you have evidence. See how Postelix approaches account champions and the outbound account brief.
What a maintained market does that an afternoon cannot
A manual list needs maintenance as roles change and new people engage with your category. Postelix brings audience research and observed activity into a workspace your team can review. From there, you can plan content, prepare an ad audience, shortlist creators and events, and decide whom to follow up with. Coverage depends on the sources available for your market.
Start with competitor-engagement research, review your LinkedIn audience, then use the outbound account brief to choose a next step. For an event campaign, use the six-week promotion plan.
Explore a fictional market at https://engine.postelix.com/demo after entering your email. Building your own market requires a paid subscription. See [current Postelix plans](/#pricing) for pricing and allowances.
Questions
What is the fastest way to find leads on LinkedIn?
Start with accounts you understand. Review recent posts from a few relevant competitors and your own team, then check each engager's role and company against your criteria. Save the activity link and date. A small, qualified sample gives you a useful starting list.
How do I find champions inside an account?
Start with people who know your product or are close to the problem you solve. Check their current role, relevant work and shared relationships. Use public activity to prepare a question, then establish through a conversation whether they want the change and can help it happen. A job title, like or comment does not establish that someone is a champion.
Are competitor post engagers actually buyers?
Some may be relevant, but a reaction alone does not establish a buyer. In our UK AI-platform sample from 20 August 2026, title filtering identified 79 service owners or team leads and 1,523 builders or practitioners among 4,449 raw reactors. Those labels describe part of that sample; they do not prove who controls a budget. Check role, company fit and the underlying activity.
Can I use a tool to scrape LinkedIn post engagers?
LinkedIn's User Agreement prohibits using software, scripts, robots, crawlers or browser plug-ins to scrape or copy data from the service, as published in LinkedIn's User Agreement, September 2026. Reading public reaction lists yourself is a different route, and so is working from your own customer data.
How many leads should a market contain?
Fewer than most people expect. The markets we have pulled have run from a few hundred highly active people to several thousand named ones. Basic discovers up to 2,000 new people per month and supports 6,000 active people in one market. The number of relevant buyers depends on the market; complete coverage is not guaranteed.
How often should the list be refreshed?
Weekly, if you want the movement. People leave companies, new people start engaging the category, and vendor attention shifts month to month. A quarterly rebuild will show you the market but not who entered it last Tuesday.