Vivify VenuesEducation / facility hire
Finding the hirers who were already there: a venue-discovery system for Vivify Venues
Vivify's best prospects were community groups already hiring space at non-customer schools, invisible to standard data providers. We built a multi-source discovery and enrichment system that finds them, scores the evidence and prices every search: 107 evidence-backed prospects from the first live run at a data cost of £0.14.
hirer groups found in the first live search
- hirer groups found in the first live search
- 107
- prospects new to Vivify's database
- 89
- total data cost of that search
- £0.14
Vivify Venues helps schools turn their halls, pitches and studios into revenue by hiring them out to community groups outside school hours. Their growth problem was sharper than most: the hirers they most wanted to win were already renting space, just at schools that are not Vivify customers. A dance class meeting every Tuesday at a non-customer school half a mile from a customer school is a near-perfect prospect. The trouble is finding them.
Those groups do not show up in company databases, because a community group's registered address is rarely where it actually meets. The signal Vivify needed was the session venue: the school named on a club finder, an activity directory listing, a Facebook page, a local league fixture grid. Gathering that by hand meant a salesperson trawling sites one at a time per target school. It did not scale, so it mostly did not happen.
What we built
We built Vivify a venue-discovery and hirer-prospecting system: a React front-end (Radiuswell) on top of an n8n orchestration layer and a Supabase database that acts as the master record of every organisation found.
The discovery engine runs two modes. An area sweep finds community groups across a postcode radius. A venue-hirers search targets one named school and hunts for groups with evidence of meeting there. The venue search pulls from three sources in parallel: Google Places, search-engine results via DataForSEO using venue-tagged queries, and Facebook page data harvested via Apify. Results are deduplicated across sources, and records that arrive without a Google place ID are assigned a deterministic synthetic one so nothing gets orphaned downstream.
Every group then passes through a shared enrichment chain: website scraping, a search-engine fallback for groups with no site, Facebook contact extraction, email verification, geocoding, AI classification of the activity type, and a quality score, before merging into the master organisations table. Each result carries a confidence tier: confirmed, where the evidence text names the venue; likely, where web or social activity matches; and proximity-only, where the group is nearby but no venue evidence was found. The evidence URL is stored with the record, so a salesperson can see exactly why the system believes a group hires that school.
Because Vivify wanted to charge searches back internally, the system records real per-source spend (Google, DataForSEO, Apify) on every search and shows it in the interface and in CSV and Excel exports.
How it works day to day
A team member types in a target school, gets a cost and volume estimate, and runs the search. Multi-source discovery and enrichment run in the background while the results page polls to completion: no babysitting, no spreadsheets. Minutes later they have a ranked list of hirer groups with contact details, activity type, confidence tier and the evidence link for each. Sales starts with the confirmed tier, where the pitch writes itself: we know you currently meet at this school, here is a better-equipped venue nearby.
Results
The first live run targeted a non-customer school in Stockport that sits across the road from a Vivify customer. The search returned 107 hirer groups, 89 of them new to Vivify's database. The search-results source added 27 groups that Google Places alone had missed, 5 of them with venue-confirmed evidence. Zero records were orphaned through enrichment, and the total data cost of the search was £0.14. (All figures are from the system's own search and cost records for that run.)
That is the economics of the whole idea in one number: a fraction of a penny per qualified, evidence-backed prospect, against hours of manual research that previously produced a fraction of the list.
The pattern
This is a productised pattern, not a one-off. Most B2B companies have a version of Vivify's problem: their best prospects are visible somewhere on the open web, but the signal that qualifies them lives in messy, scattered sources that no standard data provider indexes. The reusable system is the same every time: multi-source discovery tuned to the qualifying signal, automated enrichment and verification, confidence scoring with evidence attached, and per-search cost tracking, all behind a front-end a sales team can run themselves. Swap the sources and the qualifying signal and the same machine finds hirers for venues, tenants for landlords, or buyers for anyone whose market advertises where it shows up rather than where it is registered. If your prospects are findable but not listable, this is the build that lists them.
The part nobody sees: the data engine
Discovery is only useful if the database behind it stays clean. Under the search system sits a cleansing engine that every record passes through: live website scrape for current contact details, search-result fallback for organisations with no site, Facebook enrichment, email verification before anything is trusted, geocoding, and an AI pass that classifies what each organisation actually does against a controlled vocabulary. Then the cleanup: junk emails and dead websites stripped out, phone and mobile numbers separated, quality scores assigned, and duplicates merged into a single master record with the evidence retained.
The result is a database Vivify can actually act on, where every record says what it is, how reachable it is and why the system believes it.
How we did it
Related product
Per-prospect research feeding outreach is the same engine Emailwell productises.
See EmailwellNext case study
Inform Holdings
From spreadsheet to sequence: an AI lead engine for Inform Holdings