# Search Filters and Fields for Targeted B2B Prospecting *Guide — 2026-03-21 — by Mahmoud Zalt* The search filters that actually narrow a B2B prospect list to a usable shortlist: industry, company size, job title, keywords, intent, and the fields most people skip. **Short answer.** Targeted B2B prospecting comes down to layering four filter families: firmographic (industry, company size, revenue, geography), people (job title, seniority, function), behavioral (technology stack, hiring signals, recent funding), and intent (keyword mentions, content downloads, site visits). Apollo, ZoomInfo, and LinkedIn Sales Navigator all expose the same fields under slightly different names. If you do not want to wire those filters yourself, Sistava translates your ICP language into the filter stack automatically, no manual tinkering. ## Which firmographic filters narrow a list the fastest? Firmographic filters describe the company, not the person, and they do the heaviest lifting in any prospecting query because one bad firmographic match kills every downstream signal. The four that move the needle most are industry (NAICS or SIC code, plus the platform's own taxonomy), employee count band (the single best proxy for buying maturity), annual revenue, and headquarters geography. Most sellers stop there, but the underused ones (number of locations, year founded, ownership type, growth stage) often separate a fundable startup from a slow legacy shop inside the same industry-size bucket. The order matters: industry first, then size, then revenue, then geography, because each layer prunes the candidate pool by roughly a factor of ten when you pick a tight band. Skip the industry filter and you are just searching the entire economy with a job-title query, which is how unqualified lists balloon to fifty thousand rows of noise. ## At a Glance - **10x** List shrink per well-chosen firmographic filter - **4** Filter families that matter (firmographic, people, behavioral, intent) - **50k+** Rows a missing industry filter typically produces - **<500** Target shortlist size for a focused outbound week ## Which people-level fields actually identify the buyer? Once the company is right, the people layer decides whether the message lands on a decision-maker or rots in an intern's inbox. The mistake almost everyone makes is filtering only by job title text, which catches synonyms badly and misses entire functions. The cleaner approach combines four fields in parallel: seniority level (C-suite, VP, director, manager, individual contributor), department or function (sales, marketing, engineering, operations), job title keywords with both inclusion and exclusion lists, and tenure in role (anyone in a seat under six months is usually still onboarding and unlikely to buy). LinkedIn Sales Navigator exposes all four cleanly; Apollo lets you stack them too but the field names differ. The exclusion list is the secret weapon: filtering out 'assistant', 'intern', 'consultant', and 'recruiter' inside the title field cuts noise more than any inclusion list ever could. ## Benefits ### Seniority level C-suite, VP, director, manager, IC. Always filter on seniority before title to avoid synonym misses. ### Department or function Sales, marketing, ops, finance, engineering. Catches the buyer even when the title is unusual. ### Title keywords (in + out) Inclusion list narrows; exclusion list removes assistants, interns, recruiters, consultants. ### Tenure in role Filter out under six months and over eight years. The first is onboarding, the second is checked out. ### Years of experience Useful when seniority labels are missing or inconsistent across small companies and agencies. ## Which behavioral and intent filters separate good lists from great ones? Behavioral filters describe what the company is doing right now, and they are the difference between a static list and a live one. The three that consistently pay back the effort are technology stack (companies using Salesforce, HubSpot, Shopify, AWS, or whatever your product complements), hiring activity (open roles in your target function signal both budget and a current pain), and funding signals (recent raise, recent acquisition, recent leadership change). Intent goes one layer deeper: content downloads, review-site visits, keyword mentions in news, and third-party intent data from Bombora or G2. Most platforms surface intent as a score; treat the score as a rerank, not a filter, because intent without firmographic fit is just noise wearing a halo. The realistic stack for a small team is firmographic plus people plus one or two behavioral filters, then sort by intent inside that already-qualified bucket. ### How to layer the four filter families 1. **Start with industry** — Pick your top two or three industry codes (NAICS, SIC, or the platform taxonomy). Everything else assumes this is right. 2. **Add company size and revenue** — Employee count band first, then revenue. Together they encode buying maturity better than either alone. 3. **Layer seniority + department before title** — This catches synonyms job-title-only searches miss. Then add inclusion and exclusion keyword lists. 4. **Add one or two behavioral signals** — Technology stack and hiring activity are the highest-yield. Funding only if your offer cares about cash on hand. 5. **Sort by intent, do not filter by it** — Intent boosts rank inside an already-qualified pool. Filtering on it alone produces a list of curious tire-kickers. The catch is that every platform exposes these filters under different names, surfaces them in a different order, and hides the most useful ones behind premium tiers. Apollo names the people-side filter 'job titles' and the seniority side 'management level'. LinkedIn Sales Navigator calls them 'current job title' and 'seniority level'. ZoomInfo calls them 'job title' and 'job function level'. Same idea, three vocabularies, and a fourth platform usually shows up by the time you have configured the first three. The cost of switching tools is mostly the cost of relearning the filter names, which is one reason a lot of solo founders end up using a sales AI Employee to abstract over the underlying tool. Once the filter stack is right, the next question is which fields you actually pull off the matched records. A list of two hundred well-targeted companies is useless if the only field you have is company name. Real outbound needs verified work email, mobile number when allowed, LinkedIn URL, recent role change date, and at least two business-context fields you can reference in the first line. Pulling those fields cleanly is where most prospecting workflows quietly leak hours, because each platform charges credits per enriched field and most teams over-buy on fields they never use. ## Which data fields should you actually pull off a matched record? The fields that get used in real messages are smaller than the field list any platform offers, and that gap is where prospecting credits disappear. The minimum useful row has eight columns: full name, work email (verified), LinkedIn URL, company name, company domain, role title, role start date, and one business-context field (recent funding, recent product launch, or a public hire). Everything else is optional and should only be pulled when a specific play needs it: direct phone for cold calling, company HQ city for event invites, technology stack for product-led plays, headcount growth rate for expansion outbound. Pulling every available field by default looks thorough on paper but quietly burns enrichment credits and slows the export. The discipline is to define the row schema before the search, not after, so you know exactly what each column is for and which message it feeds. ## Benefits ### Full name + role + start date Identity, fit, and timing in three fields. The start date filters out anyone too new or too tenured. ### Verified work email Verified, not guessed. A bounce rate over three percent will get your sending domain throttled fast. ### LinkedIn URL The fallback channel when email fails, and the source of truth when other fields disagree. ### One business-context field Funding event, product launch, recent hire, growth signal. Powers the first line of every cold message. ## How do you keep filters and fields consistent across tools and teammates? Two people on the same team building lists from the same ICP in the same tool will produce different lists, because everyone applies a slightly different filter stack and grabs a slightly different field set. The fix is a single written ICP definition (industry codes, employee bands, revenue bands, target seniority, included and excluded titles, required behavioral signals) checked into a shared doc and copy-pasted into the platform every time. A short pre-flight checklist on top (did I exclude assistants, did I exclude under-six-month tenure, did I cap the list at five hundred rows) catches the obvious misses. The deeper fix is to automate the translation from ICP language to filter stack, so you brief once in plain English and the search runs the same way every time. That is the layer most platforms still leave to humans and Slack threads. ## Frequently asked questions ## FAQ ### What is the single most important filter in B2B prospecting? Industry, followed immediately by employee size band. A wrong industry filter invalidates every downstream signal, and the size band encodes buying maturity better than any other firmographic field. Get those two right before you touch job titles. ### Should I filter by job title or by seniority? Both, in parallel. Seniority catches the buyer when the title uses an unusual word; the title keyword list narrows inside the seniority band. Always add an exclusion list (assistant, intern, recruiter, consultant) which often removes more noise than any inclusion list adds signal. ### Is intent data worth paying extra for? Only as a sort, never as a filter. Intent without firmographic fit is curious traffic, not buyers. Use intent to reorder an already-qualified shortlist so the warmest accounts get worked first. ### Which platforms expose these filters best? LinkedIn Sales Navigator has the cleanest people filters; Apollo has the best price-to-coverage ratio; ZoomInfo is strongest on revenue and direct phone; Clay sits on top and lets you mix sources. Sistava abstracts over the search layer so your AI sales employee uses the same filter logic regardless of underlying provider. ### How big should a focused prospecting list be? Under five hundred rows for a focused outbound week. Anything larger means the filter stack is too loose, the personalization will be shallow, and the bounce rate plus reply quality will both suffer. Tight lists outperform big lists at the same effort. If you want the deeper read on how the data behind these filters actually gets sourced (where verified emails come from, how phone numbers get cross-checked, why one platform shows a hit while another shows nothing), the companion piece walks through the tradeoffs between accuracy, speed, and cost for B2B contact data. It is the layer underneath this one, and reading both together explains why two seemingly identical lists from two different tools can produce wildly different reply rates. The honest summary is that targeted B2B prospecting is not a tooling problem, it is a discipline problem dressed up as a tooling problem. The platforms (Apollo, LinkedIn Sales Navigator, ZoomInfo, Clay, and a dozen smaller ones) all expose roughly the same filters under different names, and the difference between a list that books meetings and a list that burns credits is the filter stack you bring to the search, not the search engine itself. Write down the ICP once, layer the four filter families in order (firmographic, people, behavioral, intent), pull only the eight fields you will actually use, and keep the shortlist under five hundred rows. If you would rather spend that effort talking to the buyers instead of configuring the filters, a sales AI Employee inside Sistava does the translation for you, runs the same stack every week, and hands back the shortlist already enriched. Pick the path that frees the hour you need most. **Tags:** b2b-prospecting, search-filters, icp-targeting, lead-data-fields, sales-prospecting, ai-sales-employee, data-fields