What Lead Researchers Do and When a Tool Is Cheaper

    Short answer

    What a B2B lead researcher actually does, what the seat costs against a data tool, a freelancer or an agency, and a sizing table that tells you when a tool is enough

    Artem Smirnov
    Artem Smirnov

    Last updated · 11 min read

    Artem Smirnov in a dark suit against a charcoal studio backdrop, next to the line 'WHO BUILDS YOUR LIST. The tool is cheaper. The judgment is not'

    A lead researcher in B2B sales builds the target list before anyone writes a word of outreach. The job has three parts:

    1. Find companies that match your ideal customer profile (ICP).
    2. Work out who really decides at each one.
    3. Confirm a working contact for that person.

    The finished rows then go to whoever runs outreach.

    You can fill that seat four ways: hire a researcher, buy a data tool, give the work to a virtual assistant or freelancer, or outsource it to an agency. Tools now handle most of the searching and the address checks. What they cannot supply is judgment: which titles actually sign, which rows to delete, and whether the list is big enough to be worth sending.

    A disclosure before my opinions. My team books meetings for clients through cold email and LinkedIn, and we run no dialing floors. Phone is a narrow add-on: we can call people who have already replied or opted in, never a cold list. List building is part of what clients pay us for, so read my view on hiring versus outsourcing with that in mind.

    What does a lead researcher do in B2B sales?

    Every message you send later is only as good as the row it lands on. Research on inside sales teams at B2B firms puts lead research first in sales development, with every lead judged on how well it fits the ICP (Journal of Business & Industrial Marketing). In complex sales, reps did deeper background research before first contact.

    Research also eats a real share of the selling week, which is the argument for giving it its own seat. A 2013 paper on predicting B2B prospects cites earlier work putting prospect selection at about 20% of a rep's time (ScienceDirect).

    Newer numbers point the same way. Salesforce's 2026 State of Sales statistics say sales reps spend 60% of their time on non-selling tasks. Research is only part of that 60%, but it is one part you can hand off: a researcher, a tool or a list building service builds the list, and the rep spends those hours talking to buyers.

    A note on the term. Most search results for "lead researcher" describe the person who runs an academic or user research project. That role shares the name and nothing else, and this article is about the sales seat.

    Which parts need a person and which can a tool do?

    TaskBetter done byWhy
    Choosing which companies to model the list onPersonIt starts from your own client history, which no database holds
    Naming the titles that sign in your nichePersonTitles shift between niches, and generic ones pass every filter
    Pulling companies and people that match the filtersToolFast and repeatable once the filters are right
    Finding an email address or profileToolData providers cover this at scale
    Verifying every addressTool, with a person reading the resultSoftware runs the check; a person decides what happens to risky rows
    Keeping CRM records currentBothEnrichment updates fields; a person removes people who left or objected
    Deleting rows that fit the filters but not the buyerPersonEvery row needs a reason to stay

    I start with the client's past and current customers and look for lookalikes only after that. On the same pass I find out who signed each of those deals. That one name tells me more about the buyer than any persona slide.

    After that, I search using the job titles that niche actually has. Assistant, Associate and a bare Manager come out on purpose, because they slip through every filter and rarely sign anything. Signing titles change from niche to niche, so lift them from your own closed deals.

    Inside Apollo, separate filters combine with AND, while the options within a single filter combine with OR. Results stop at 100 pages, so if you hit that cap, tighten the filters before you export (Apollo search filters).

    The same help page warns that a broad keyword such as "manufacturing" drags in companies that merely mention the word, because a keyword search matches any company that uses the word, while each company carries only one industry label. A lookalike search accepts up to 5 seed companies, so your best existing clients can seed the first search.

    Verification software removes invalid addresses, helps correct typos and drops duplicates. ZeroBounce, a verifier vendor, calls an address valid when it exists and is in use on the mail server, and invalid when it does not exist or is wrongly formatted (ZeroBounce verifier).

    Apollo says its built-in check is 91% accurate. That is the vendor's own figure, so treat it as a claim, not a measurement (Apollo's verification page). Mailbox providers expect bounces at 2% or lower, says ZeroBounce. My team holds bounces under 0.5% of sends, and that only works when nothing leaves unchecked.

    If you are writing a lead researcher job description, these five points are its core:

    • They can explain in one sentence why each row is on the list.
    • They know your buyers' titles, or they ask before guessing.
    • They write boolean title searches (strings that join titles with AND and OR) instead of scrolling through results.
    • They record a verification result and date for every address.
    • They delete rows without being asked.

    On my team, lists and data stay with senior people. Juniors, interns and students never touch that part of the work.

    How big does the researched list have to be?

    In the first 30 days of outbound for an SEO company in Singapore, we contacted about 16,000 people. Around 800 replied, a 5% reply rate. 78 of them booked a sales meeting with the client.

    Read it backwards: 78 meetings in a month -> about 16,000 researched contacts at the top. And 5% was a strong month. Not every month looks like that.

    Do not plan on that. In my campaigns, replies usually land around 1% of people contacted, and at thousands of emails a day they drift toward 0.5%.

    Between 10% and 50% of replies turn into meetings. Very specific, heavily personalized campaigns can reach replies of 15% to 17%, with about 75% of repliers interested. The bigger the volume, the lower those averages sit.

    Here is the sizing that follows from the 1% average and a 10% to 50% share of replies turning into meetings. The last column is my answer to when a tool is enough.

    Meetings you want per monthVerified contacts needed per monthWho should own the research
    51,000 to 5,000A data tool plus one careful reviewer can cover it
    102,000 to 10,000A tool works if the titles are settled and the market is large; otherwise give one person ownership
    30 to 356,000 to 35,000; my working rule is at least 10,000A full-time researcher or an agency; a tool login alone cannot carry it

    At the 0.5% rate typical of large volume, double each figure. My first list for a new campaign starts at 20,000 contacts.

    So I want research that is accurate AND big enough. A short, perfect list of 50 or 100 names a month still collides with people who are busy, mid-contract or not buying this quarter. That is a bet on timing. Volume evens out the luck.

    The reverse also holds. If the first 1,000 rows start no conversations, 10,000 more rows built the same way will not either. Copy cannot rescue a wrong list. Fix the list first.

    How much does a lead researcher cost compared with tools?

    No official pay series exists for the title "lead researcher". Every row below is a proxy, and each one has a limit you should know before you budget from it.

    OptionProxy figureSource and dateLimit of the proxy
    In-house hire, USMedian $78,760 a year; lowest 10% under $43,390, highest 10% over $155,480BLS, market research analysts, May 2025A broader research and analysis occupation, not a list-building title
    Freelancer or VA, California$25 to $89 per hourUpwork, at the time of writingMarketplace data with no stated method; covers lead generation specialists in general
    Data tool, ApolloA free plan, plus paid plans priced on the page that vary with seats and creditsApollo pricing, October 2026It finds and checks contacts, yet someone still has to define the buyer
    AgencyNo independent price series exists. Our own fees start from USD 3,000 (about EUR 2,500) a month and can pass USD 10,000 when the scope is largerWhat agencies charge for lead generation shows the published ranges and fee modelsThe fee usually bundles research with sending and reply handling, so compare whole systems

    I left Europe out of the table. The only EU pay data I had was from 2022 and covered all jobs, which is too old and too broad to budget a seat from.

    Apollo's pricing page also says plan credits reset at the start of each billing cycle and unused credits do not roll over. So size any paid plan to the credits your researcher will actually spend each month.

    What those rows mean for your budget:

    • As of May 2025, the BLS median works out to roughly $6,560 a month in salary alone. Payroll costs, software and data come on top.
    • For a rep-level benchmark, BLS puts the May 2025 median annual wage for wholesale and manufacturing sales representatives outside technical and scientific products at $72,080. That group includes inside sales roles.
    • The Upwork range shows what freelancers charge. It tells you nothing about whether their lists are any good.
    • A free tool plan costs nothing in money. Its real cost is the reviewer's hours, so price those before you start.
    • If a sales manager does that review, those hours are expensive. BLS reports a May 2025 median annual wage of $148,270 for sales managers.

    Whichever route you pick, software and data are a second bill. On our side, clients pay for tools, data, domains and mailboxes directly. None of it is inside our monthly fee, and the total moves with your market, your send volume and how many reps you run.

    Should you hire a researcher, buy a tool, use a VA or outsource?

    OptionFits whenWho verifiesMain risk
    Data tool run by your own teamYour market is small enough to search in one pass, and someone in-house can name the signing titles and write filtersYour team, with a verifierNoisy keywords and generic titles slip through
    Freelancer or VAYou have a written spec and someone in-house who reviews the workThe freelancer first, then youCheap lists recycled from other buyers
    In-house researcherYou need thousands of new contacts every month for several reps, plus constant CRM upkeepThe researcherA salary for the seat, and the seat gets pulled into replies
    AgencyYou want the list, the sending and the reply handling run as one system, with no in-house skill to build itThe agencySome agencies hide who does the work; ask for names and look them up

    Decision rules I would apply:

    • If nobody can name the titles that signed your recent deals, fix that before choosing any option. No tool or hire can do it for you.
    • If your whole market fits in a few thousand companies, a tool plus one careful reviewer usually covers the research. Put the rest of the money into the offer.
    • If you go with a freelancer or VA, put the spec in writing and review their first batch yourself. What else you can safely hand a VA is the question behind whether a VA or an agency should run your outreach.
    • If the sizing table put you in the bottom row, someone has to own the list as a job.

    Two patterns hold in my campaigns. Cheap lists bought from freelancers give themselves away fast, through weak open rates and a pile of bounces. When my team builds a list, we stack several data sources, such as Sales Navigator, Apollo and UpLead, and that combined list beats a scraped one. Stacking sources is the first step in how my team runs B2B lead generation for clients.

    Intent data earns its place only when the market is large. In a niche of roughly 2,000 companies, a signal feed that lights up on about 10 of them leaves almost the whole market dark. In a market that small I skip intent data and work on the offer instead, so every company on the list has a reason to reply.

    How is a lead researcher different from an SDR?

    The Journal of Business & Industrial Marketing field study cited above describes sales development reps as back-end salespeople who use digital tools to nurture and qualify leads before passing them to outside sales. A researcher decides who goes on the list and proves each person is reachable. An SDR talks to those people and qualifies them.

    One person can hold both seats, but then the research needs protection from the reply inbox. Book research time in fixed blocks and count the new rows added each week.

    How do you check a researcher's list before anyone sends?

    Before you pay for a full list, ask whoever builds it for 50 rows of the real thing: a hire on trial, a freelancer, a tool export or an agency. I would ask any agency for that sample before reading a word of their copy. Once you have seen the rows, ask the questions that matter before hiring a lead gen agency.

    Run each row against these 10 points:

    1. ICP fit: each row says why the company matches your profile.
    2. Real titles: each person decides on or signs for what you sell.
    3. Generic titles removed: none of the low-signing titles named above is left.
    4. Source column: you can see where every row came from.
    5. Verification status: every email carries a result and the date of the check.
    6. Catch-alls flagged: Apollo describes a catch-all server as one that accepts mail for any address at its domain, real or not, so a "valid" result there proves little.
    7. Duplicates removed: no person appears twice.
    8. Sole traders flagged: UK rules treat them as individuals, as the next section explains.
    9. Company size checked: every company sits inside the headcount range you actually serve.
    10. Date of pull: people change jobs, and a list without a date cannot be judged.

    Then pick 20 rows at random and check each person by hand on LinkedIn and the company website. Anyone who left, holds the wrong title or works at a company outside your range counts as a research fault. Count the faults before you pay for the rest.

    The date matters because lists go stale fast. ZoomInfo, which sells B2B data, cites industry benchmarks putting yearly B2B data decay at 22.5% overall and above 70% for some fields, email addresses among them. Treat that as a vendor's number with an interest in refreshes. Even a fraction of it is enough to sink an undated list.

    In the UK, the rules sit in the ICO's guidance on business-to-business marketing, which the regulator is reviewing after the Data (Use and Access) Act (ICO). These are the points a researcher's work touches:

    • Corporate subscribers (companies and similar bodies) sit outside the consent rules in PECR, the UK's Privacy and Electronic Communications Regulations. You must still identify yourself, and each message has to carry a working opt-out.
    • Sole traders and some partnerships are treated like private people. Emailing them needs their consent or the soft opt-in. The soft opt-in only covers details you gathered while selling, or negotiating to sell, them similar products, and every message must offer a way out.
    • Once your researcher writes a person's name against a company, that row holds personal data under the UK GDPR. If that person says no to direct marketing, you stop, with no exceptions.
    • A public profile is not consent to marketing. If you took the details from a third party or a public page, tell the person how you use them within one month.
    • If anyone on your side phones these contacts for marketing, screen the numbers against the Corporate and Telephone Preference Service (CTPS and TPS) opt-out registers and your own do-not-call list.

    In the EU, recital 14 of the GDPR places personal data about legal persons, including a legal person's contact details, outside the regulation (GDPR). That covers the company and its general contact details. The named employee your researcher found is another matter: their details remain personal data.

    This is not legal advice.

    Deciding between a hire, a VA and an outside team for list research? Book a call and send me a sample of your current list, and I will give you my honest read.

    Questions founders ask about lead research

    Do lead researchers need a degree?

    This sales title has no formal requirement. The closest official comparison, the BLS profile for market research analysts, says those roles typically need a bachelor's degree, as of May 2025. For list work, test the skill instead: hand a candidate your ICP and ask for 10 rows, each with a written reason to stay.

    Should a researcher collect emails for prospects in Germany?

    My team sends cold email to buyers in Germany, Austria or the Czech Republic only where a compliant basis exists. When a client needs German accounts, we apply stricter house rules and reach those buyers on LinkedIn. So for those three countries, a researcher should put accurate LinkedIn profiles ahead of email addresses.

    Should a lead researcher work in the CRM directly?

    Yes, if the CRM records an owner for every account. My team gives each account and each rep a separate territory, so two people never write to the same contact. A researcher importing into a shared CRM with no owner field can drop one decision-maker into two sequences in the same week.

    How often should a researched list be refreshed?

    Verify again right before every send, whatever the calendar says. Apollo refunds an email credit only when the address bounces within 30 days of the reveal. A list pulled last quarter and sent today gets no refund, and nobody has checked whether those people still hold the job.

    So settle the signing titles, size the list from the table, check 50 rows, and only then decide who builds the rest.

    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.