The customer ask From a discovery call · Nooks
Find meUS B2B companies, $20M+ revenue, with 3+ SDRs/BDRs — verified through current title AND historical job experience — using Salesforce, Outreach, Gong, or SalesLoft, in sales / RevOps / marketing verticals, excluding SDR teams in offshore-only locations.
All four tools get the same ask. Watch what happens.
Processing query
filter: revenue · ok
filter: tech stack · ok
count SDRs / company · unsupported
title vs. job history · unsupported
filter: tech stack · ok
count SDRs / company · unsupported
title vs. job history · unsupported
✕ Cannot execute
Filters operate per-record only. Can't roll up "3+ SDRs per company." No way to match titles against historical experience.
Processing query
persona: SDR · ok
tech filter · ok
title · job history · current only
nested AND/OR/NOT · limited
tech filter · ok
title · job history · current only
nested AND/OR/NOT · limited
✕ Cannot execute
Title filters match current role only. Misses anyone who used to be an SDR. Persona filters can't compose with company-level tech + vertical + count in one query.
Processing query
tech filter (via ZI/Apollo) · passthrough
SDR count loop · manual
title array · no native UDF
runtime · ≈ 35 min
SDR count loop · manual
title array · no native UDF
runtime · ≈ 35 min
~ Partial · slow
Inherits the ceilings of underlying providers. Title-history must be hand-coded per row. Hits rate limits at scale.
Processing query
Spark SQL editor · ready
UDFs (match_phrases, title_array) · native
transform(experience, exp -> exp.title) · supported
aggregate SDR count · native
UDFs (match_phrases, title_array) · native
transform(experience, exp -> exp.title) · supported
aggregate SDR count · native
✓ Executes · 4.2s
Spark SQL with user-defined functions over the full 200M-company graph. Title arrays, lambda transforms, nested aggregations — one query.
Walk through how Landbase runs the query against the buying graph.
Why this works on Landbase: a fundamentally different approach. Other platforms force every prompt into a fixed filter UI; Landbase lets the customer write the search criteria themselves — flexible enough to express any condition, then narrow 213M down to the qualified set.
FUNNEL · QUALIFICATION
idle
running…
SPARK
3,247 qualified accounts. Every clause satisfied. Each row: company + verified SDR count + matched tech + buying-group contacts. Ready to score, tier, and push to your CRM.
3,247
accounts
Unlimited possibility — every customer ask becomes one query.
Each of these came from a real customer call. None of them can be answered with ZoomInfo or Apollo filters. All of them are one query in Landbase — Spark SQL with user-defined functions over the full 200M-company graph.
Industrial · Energy
"Find mid-market US manufacturers on 5+ acres, with multi-shift operations, no public automation signal on their site or LinkedIn, and an EPA Tier-II permit on file."
Hiring signal · Data
"Find SaaS companies that hired 5+ data engineers in the last 90 days, where the new hires came from FAANG or Series B-D startups, using Snowflake + dbt."
Buyer journey · RevOps
"Find companies whose Director of RevOps was previously a Salesforce customer and switched to HubSpot in the last 18 months — and is now hiring SDRs."
Regulatory · Healthcare
"Find healthcare orgs with 10+ NPI providers, multi-state license, no PE backing, in states with new telehealth reimbursement rules this year."
Intent + tech stack
"Find Series C+ B2B companies where the CFO posted on LinkedIn about cost optimization in the last 30 days AND the stack includes Snowflake + dbt + Looker."
Negative signal · CRM gap
"Find companies not yet in our CRM, 51-500 employees, in our true-fit ICP, with a CRO in seat for less than 12 months."
STEP 1 / 5
A real ask from a Nooks discovery call. Six conditions, three data sources, two aggregations.