Most B2B operators spend two days building a list that should take 10 minutes. They jump between LinkedIn, a scraping tool, a spreadsheet, and a CRM, hand-cleaning data that any modern prospecting platform can return in a single query. If your list-building process takes a full day, the process is the bottleneck, not the data.
This is the exact sequence we use to build a 10,000-record, ICP-filtered, enriched, and verified B2B list in under 10 minutes using Apollo as the primary source. It works for any vertical where your ICP can be described in filterable criteria, which is to say almost every B2B vertical.
Step 1. Define the ICP in filterable criteria
Every field in your ICP needs to map to a filter inside Apollo, ZoomInfo, or Sales Navigator. If it does not, it is not an ICP criterion for the purpose of list building. Write it down as a secondary qualifier to use after the list is pulled, but do not let it block the initial query.
The four required fields
- Industry or sub-industry. Specific. Not 'technology' but 'B2B SaaS serving mid-market marketing teams'. Filter on Apollo's granular industry taxonomy, not the broad tier.
- Employee count band. Tight enough to capture only one buyer profile. A 10 to 50 headcount founder is a different buyer from a 200 to 500 headcount VP.
- Revenue band. Loosely filters on who can afford your offer. Set a floor that matches your entry-level pricing.
- Location. Country at minimum. State or metro for local targeting. Filter on company HQ, not employee location.
The optional fields that sharpen the list
- Tech stack or technographic. If your offer integrates with a specific tool, filter on it. Cuts the list in half and doubles reply rate.
- Funding stage. Early-stage and recently funded buyers move faster. Late-stage and public buy through procurement.
- Headcount growth rate. Companies growing 30 percent plus year over year are hiring, buying tools, and have budget moving.
Step 2. Build the Apollo query
Open Apollo. Go to the People search. Do not start with the Companies search, because the goal is a list of named decision makers, not a list of accounts to be enriched later.
The filter order that avoids over-narrowing
Apollo filters compound, so the order you apply them in changes the result count. Apply the broadest filters first so you can see how each successive filter cuts the list. If a filter drops the count below your target, loosen it before applying the next one.
- Apply industry. Confirm the count is in the right range (typically 40,000 to 400,000 for a major vertical).
- Apply employee count band. Expect a 50 to 70 percent cut.
- Apply revenue band. Another 20 to 40 percent cut.
- Apply location. Usually light, 10 to 20 percent.
- Apply job title or seniority filter. Target director, VP, or founder-level depending on the offer.
The seniority filter is the one most operators get wrong
Do not filter on a specific job title like CMO. Titles vary by company size. A 40-person SaaS calls their marketing lead Head of Growth. A 400-person SaaS calls them VP of Marketing. Filter on department (Marketing) plus seniority (Director and above) and you capture both. Apollo supports this natively.
Step 3. Pull and enrich the list in one pass
At this point the Apollo list shows 8,000 to 15,000 records. Export the records with the Enrich option selected, so Apollo pulls emails and mobile numbers on the export rather than leaving empty fields. This single toggle saves a full enrichment pass later.
Apollo enrichment typically returns emails on 60 to 75 percent of records on the first pass. The remaining 25 to 40 percent will need waterfall enrichment through a second provider. Do not stop at the Apollo export. A list that is 70 percent enriched is not ready for sending.
Running the waterfall in under two minutes
Drop the unenriched records into Clay or Floker with a pre-built waterfall recipe that hits Prospeo, Contact Out, and FindyMail in sequence. A 3,000 record unenriched residue runs through this in about 90 seconds and lifts total coverage to 85 to 92 percent.
Apollo initial enrichment
60 to 75%
After waterfall
85 to 92%
After verification
82 to 89%
Cost per final record
$0.08 to $0.14
Step 4. Verify and segment
The enriched list goes through two verification providers and Scrubby for catch-all resolution before anything touches a sequencer. This is not optional. A list that is 85 percent enriched but contains stale catch-alls bounces at 8 percent and kills the sending domain inside a week.
Segment by sub-industry, not by the full list
The 10,000 record list is not one campaign. It is five to 10 sub-segments of 500 to 2,500 records each, because reply rate comes from writing copy that is specific to the sub-segment. A segment of 2,000 mid-market B2B SaaS marketing directors converts at two to three times the rate of the full blended list, because the first-line can reference a problem that only they have.
- Segment on sub-industry (fintech, healthtech, martech, etc.) within the broader vertical.
- Segment on company size band. A 50-person buyer has different problems from a 500-person buyer.
- Segment on geography when the offer is locally compliant or locally delivered.
- Do not segment on seniority unless you are running the same offer with a different angle per level.
Step 5. Ship it without touching a spreadsheet
The last step that kills most list-builds is manual import. Exporting from Apollo, opening it in Sheets, filtering, re-exporting, uploading to the sequencer. Every touch is a chance to corrupt the list and every touch adds 15 minutes to the timeline.
If your sequencer supports direct integration with Apollo or Clay or Floker, use it. Smartlead, Instantly, and Lemlist all support this. The list flows from source to enrichment to verification to segment to sequencer without a spreadsheet in the middle. Total build time from query to campaign-ready list runs eight to 12 minutes once the workflow is set.
The takeaway is not that this process is clever. It is that modern prospecting platforms already support every step, and the slowdown is always human process. Eliminate the spreadsheet, eliminate the manual touches, and a 10,000-lead build drops from a half-day task to something your team does on a Tuesday morning before standup.