How to Search Companies on LinkedIn for Prospecting

    Short answer

    Start from the accounts you already won, check the industry tag against the website, and qualify every company before you look for a single contact.

    Artem Smirnov
    Artem Smirnov

    Last updated · 12 min read

    Artem Smirnov in a dark suit against a charcoal studio backdrop, next to the line 'SEARCH COMPANIES, NOT LABELS. The industry tag is a guess'

    The order I follow to find target companies on LinkedIn

    To search companies on LinkedIn, type a keyword into the search bar, open the Companies tab and narrow the results with the filters at the top of the page or under All filters (LinkedIn Help). For prospecting, Sales Navigator's Account Search adds deeper account filters, such as headcount growth, department headcount and headquarters location, and lets you save companies into account lists.

    I work through 5 steps, always in this order:

    1. Define the target company from the clients you have already won.
    2. Pick a search method that fits the size of your market.
    3. Check the industry yourself, against the company's website.
    4. Qualify each account with a short keep-or-drop check.
    5. Only then search for people, using the titles that actually sign deals in that niche.

    When outreach stalls, founders rewrite the email, since it is the one part they can see. I ask for the list before the copy. Wording is not the lever when the list is wrong. A perfect email to a company that was never going to buy is wasted.

    List building is one stage of a bigger process. If you want the whole picture first, what outbound prospecting is and how its parts connect lays it out.

    How do you search companies on LinkedIn and Sales Navigator?

    Free LinkedIn search. Type a keyword or a plain-language query into the search bar at the top of any LinkedIn page and press Enter. Choose Companies from the tabs above the results. That is enough to see the rough shape of a market or to find one company you already have in mind.

    Sales Navigator Account Search. Inside Sales Navigator, use the + beside Account filters. That opens Account Search. Only a few filters do most of the work on a target list. Definitions come from Sales Navigator Help:

    FilterWhat LinkedIn says it doesHow I use it
    IndustryFilters accounts by industryA first cut only, checked against the website later
    Company headcountCounts the account's employees who are on LinkedInMatch the size band of past clients
    Company headcount growthHow much staff numbers changed in the past 12 monthsSeparate growing companies from shrinking ones
    Headquarters locationRegion, country, state, city or postal code of the headquartersKeep the list inside the markets you serve
    Department headcountTotal employees in a specific departmentCheck that the buying team exists at all
    Job opportunitiesOnly accounts with open jobs posted on LinkedInA hiring trigger for the qualification checks

    One caution on headcount. It only counts people who have a LinkedIn profile. In trades where field crews rarely keep one, a company can look smaller than it is. Before you drop an account for size, look at its website, its projects and its office list.

    Account lists. When a company passes your checks, click Save and add it to a list you already keep or start a fresh one (Sales Navigator guide). Later, a people search limited to that account list returns contacts only from the companies you approved.

    If you are still deciding whether the paid tool is worth it for your team, what Sales Navigator does beyond free search walks through its features and limits.

    Which search method should you use for your market?

    Two things you already know decide the method: how many companies could buy from you, and how many clients you have closed so far.

    MethodWhen it fitsWhat it missesVerdict
    Lookalikes of past clientsYou have a handful of clients who paid well and stayedSegments you have never sold toStart here whenever you have a client history
    Keyword and Boolean search for the verticalThe vertical has its own vocabulary: trades, product types, service namesCompanies that describe themselves in generic wordsUse it to widen the lookalike list inside one vertical
    Sales Navigator account filtersYou know the size band, region and growth pattern of a good accountCompanies with thin pages or a misleading industry labelThe main LinkedIn route once the target is defined
    X-ray search through a search engineA niche so narrow that keyword search keeps returning noiseThin or outdated pages, and anything the engine has not indexedA fallback for a few hard-to-find companies
    B2B data toolsA large market where you need thousands of accounts and contactsRecent changes, and the reason a company fitsCombine more than one for volume, then clean the result

    Every list I build starts from the client base. Take the past and current clients who paid well and stayed, then find companies that look like them. A lookalike list carries its own proof: someone like this has already bought.

    Keyword search widens that list inside the same vertical. It only works with the words the vertical uses about itself, so collect those words before you type anything.

    X-ray search means asking Google or another search engine to find LinkedIn pages for you. A query like site:linkedin.com/company "precast concrete" Ohio returns public company pages that mention both the trade and the place. You get names fast and still check fit by hand. The full method, with more query patterns, is in the guide to LinkedIn X-ray search.

    Data tools are where volume comes from. A list from proper B2B data sources beats a scraped one, and I rarely trust one source alone.

    One client of ours, a UK social media marketing agency selling into 3 niches, ran on lists we built by combining Sales Navigator, Apollo and UpLead. The calendar filled a little more each month: 44 booked sales calls in month 1, then 57, then 68. Across LinkedIn and email, with no paid ads, the campaign reached 714 sales calls in 12 months.

    Market size decides how deep you go. With a few hundred or a few thousand possible buyers, go named and deep. Put every account on the list by name, research each one and write an offer for that whole group.

    Deep does not mean tiny. A handful of perfect accounts is still a coin flip: many good buyers are simply busy, not ready or not interested this month. You need enough accounts as well as the right ones. Keep the lookalike list wide enough that a quiet few weeks does not empty your pipeline. Volume evens out the luck.

    Once the count reaches tens of thousands, pull from several sources and push for volume while keeping the data clean. My working floor: if you want 30 to 35 booked B2B meetings per month, plan on reaching at least 10,000 people per month. Volume on messy data burns your sender reputation, so cleaning comes before scale.

    Why that many? Instantly's 2026 benchmark report gives an average reply rate of 3.43%, and above 10% only for elite campaigns. And lemlist cites Cleverly data putting meeting-booked rates at 0.5% to 2.5%, depending on deal type, offer clarity and follow-up quality. At those rates, a small list does not fill a calendar.

    Treat the industry on a LinkedIn company page as the company's own description of itself. It can drift from what the company actually sells, especially after a pivot.

    That causes two mistakes. A company with the wrong label shows up in searches it does not belong in. A company with a vague label never shows up in the search that should find it.

    Take a 30-person firm that builds payroll software but files itself under "Accounting". A search for software companies misses it. A search for accounting firms pulls it in. When the label and the website disagree, trust the website.

    Official classification helps you name the segment. NAICS is the North American standard US federal statistical agencies use to classify business establishments, adopted in 1997 to replace the SIC system (U.S. Census Bureau). Its search accepts a keyword or a 2 to 6 digit code, so you can look up the official name for the trade your best clients work in.

    Read the descriptions that come back. Write down the exact words used for that activity and for the activities next to it. Those words become your search keywords and your exclusion terms.

    These are the label mistakes I look for. The examples are invented; the checks are the ones I run:

    Mislabel patternExampleThe check that catches it
    Service firm filed under its clients' industryA marketing agency for builders listed under constructionThe services page sells campaigns and retainers, not building work
    Manufacturer filed under the industry it sells toA maker of commercial kitchen equipment listed under restaurantsThe product range, distributors and a factory appear on the website
    Builder filed under designA design-and-build fit-out contractor listed under architectureCase studies show finished builds, and staff titles include site managers and estimators
    Old label after a pivotA firm that moved from hardware resale to managed IT, still under its old categoryThe latest case studies and current job ads

    A company full of account managers with no engineers sells services, whatever its page says.

    A worked example: from three past clients to 60 accounts

    Picture a firm that does outsourced estimating for building and civil contractors in Ohio. The companies are invented for the example. The steps are the ones I run on a real list.

    To size a market like this, start with the U.S. Census Bureau's County Business Patterns. It is an annual series with data on establishments with paid employees, by industry and employment size. Filter it by state and by NAICS code: 238 for specialty trade contractors, 237 for heavy civil.

    One catch. The National Association of County Executives puts it plainly: "An establishment is a single physical business location, not a company." A contractor with 3 yards shows up 3 times. So the number of companies you can sell to is lower than the count.

    The seller has three past clients:

    Client 1Client 2Client 3
    VerticalCommercial interiors contractorSite work and excavation contractorCivil engineering contractor
    Size bandMid-size, a few dozen staffMid-size, just over a hundred staffMid-size, under a hundred staff
    RegionColumbusClevelandCincinnati
    Who signedVP of preconstructionVP of preconstructionChief estimator

    What they share becomes the definition: mid-size contractors doing commercial and civil work in Ohio, where preconstruction owns the estimating budget.

    Step 1: confirm who decided. Go through past deals and note who actually made the call on each one. The people in the meetings are a different list. Here the VP of preconstruction signed twice and the chief estimator once, so both roles go on the list.

    Step 2: write down the vocabulary. These contractors describe themselves as site work, excavation, civil, tenant improvement, general contractor or design-build. Those words find them far more precisely than the broad label "construction".

    Step 3: build the strings. First, the operators.

    Boolean operators combine words in a search: OR matches any one of the words listed, AND requires both bracketed groups to match, and NOT drops any result containing the words after it. An account string for this seller might read:

    ("site work" OR "excavation" OR "civil engineering" OR "tenant improvement" OR "general contractor") AND ("Columbus" OR "Cleveland" OR "Cincinnati" OR "Ohio")

    In Sales Navigator, put the trade words in the search bar, set Headquarters location to Ohio and set Company headcount to the size bands of the three clients. The people string waits until the accounts are chosen. Here it is:

    ("Preconstruction" OR "Chief Estimator" OR "Estimating Director") NOT ("Assistant" OR "Trainee" OR "Associate")

    I strip generic titles out of every search. "Manager" on its own says nothing about who signs, and neither does "Assistant" or "Associate".

    Step 4: filter down. Say the account search returns 150 companies. The label check removes the design studios, equipment makers and agencies that used contractor words in their descriptions. Qualification then removes firms too small to need outsourced estimating and firms big enough to keep it in-house.

    Suppose around 60 accounts remain, each with a written reason to be there and two titles to look for. They go into one account list, and the people search runs only inside it.

    How do you qualify a company before you look for contacts?

    A company can match the vertical and still be a bad account. Run every account through 4 checks before anyone searches for a person:

    1. It sells what your past clients sold. Look for the same type of work and the same kind of customer. A site work firm and a homebuilder both pour concrete, but only one of them buys estimating the way your clients did.
    2. It can afford your deal size. Headcount, the number of live projects and the job sizes in its case studies show roughly what it spends. If your smallest past deal would eat a big share of that, drop it.
    3. It has a buyer role you have already closed. If your deals were signed by preconstruction leaders and this company has nobody in that seat, you would be selling into a structure you have never sold into. Go back through your past projects, check who really made the call on each one, then look for that seat here.
    4. There is a visible trigger. Hiring for the roles you support, a new leader in the buying seat, a new office or a run of contract wins all count. The Job opportunities and headcount growth filters surface some of these for free.

    If 2 checks fail, drop the company. If 1 fails, keep it and note which one, because that note shapes the first message.

    Set the thresholds from your own past deals, not from this article. Mine would be wrong for you. Qualification is what lets you write one message for one group, because every company left on the list has the same reason to buy.

    How do you move from the company list to the right people?

    Now you search for people, inside your account list only, by role. Use the titles that niche really gives the person who signs. In the Ohio example, that means the VP of preconstruction and the chief estimator, nobody else.

    How many people to contact per account depends on how the company buys. At a firm with one clear budget owner, start with that role and add a second person only if the deal needs it. If the deal has several buyers, how many contacts to reach at each company covers buying committees.

    Before the first send, every address goes through NeverBounce or DeBounce, and bounces stay below 0.5%. A high bounce rate marks your sending domain as risky, and every later campaign pays for it.

    The mailbox providers say much the same. Yahoo's Sender Hub advises senders to watch hard bounces, soft bounces and inactive recipients.

    Then split the list. Each account goes to one rep and one sending account, so nobody on your side contacts the same person twice. It is the same split my team uses when we run LinkedIn outreach for clients.

    How do you check a list before you trust it?

    If you inherited or bought a list, audit it before a single message goes out. In my experience, a list bought cheaply from a freelancer usually shows the same pattern: few opens and a lot of bounces.

    If you hire someone to build lists, ask for a sample before you look at any copy. Request 50 rows with company, person, title and the reason each one is there. Run 20 of them through the qualification checks above yourself. If several of the 20 fail, treat the full list as suspect and ask for a rebuild.

    If a company on your list has no written reason to be there, delete the row before you search for a single person.

    If your LinkedIn company search keeps returning the wrong firms, book a call and share your filters. I will tell you which ones I would change.

    Questions founders ask about searching companies on LinkedIn

    Should I target companies in my own city first?

    When your offer allows it, yes. In our campaigns, prospects who know you are based nearby, and could meet you in person, are easier to start a conversation with. That makes a local slice a cheap first test for a new list. Start with the region you work from, give it a few weeks of replies, then move to the next region.

    Do I need buyer intent data to pick target companies?

    It pays off when the market is large. My team uses intent data when it fits, as one tool among several. Say 2,000 companies fit your target and only about 10 show an intent signal. A signal on 10 names tells you nothing about the other 1,990, so in a market that small we skip the tool and write one offer for the whole list.

    Does the Sales Navigator industry filter include sub-industries?

    Yes, selecting an industry in Account Search also includes its child industries (Sales Navigator Help). Choose the narrowest industry that fits, then tighten it with the trade words you collected from your best clients' websites.

    How do I find a person at a specific company on LinkedIn?

    Work from the firm outward. Open its Account Page in Sales Navigator before you look at any person. It tracks hiring, team size by department, and which decision-makers have recently arrived or left (Sales Navigator guide). For more names, run a lead search with Current Company set to that firm and narrow by Function or Seniority. Keep only the people whose role matches the buyer you closed before.

    Want to see what we would build for your company?

    I am Artem Smirnov. I help established B2B companies build and run the go-to-market system behind their pipeline, from outbound to AI search visibility. Book a call and we will see what fits yours.

    Artem Smirnov
    Artem Smirnov

    I help established B2B companies build and run their go-to-market systems. 9 years in outbound, 10,000+ campaigns.