Icarus GrowthFoundational Implementations

Marketing · Lead Generation / 07 of 09

Custom Enrichment Tools

Build a small model or API that answers the one question about a prospect that no data provider sells, so that you can filter and target on a signal your competitors cannot even see.

How it runs

A picture becomes a filterable column

Trigger

Row in, address only

Step 01

Look up the coordinates

Step 02

Pull a satellite image of roof

Step 03

Run the JPEG through the model

continues

Step 04

True or false comes back

Step 05

Return it to the calling API

Step 06

Write it onto the row

Outcome

Used as a filter

What it is

A purpose-built model wrapped in an API, answering one question about a prospect that no enrichment provider sells you.

You need this if the thing that actually qualifies a lead in your market is something you can only tell by looking, which caps your targeting at whatever a human can check by hand.

Everyone else is buying identical fields from identical providers and writing identical lines.

If you already enrich, a signal you own is the one input a competitor cannot buy their way to with the same budget.

The actual build

Custom Enrichment Tools — the as-built system.

What it does for you

01

A judgement that used to need a human eye runs across the entire list

so that list cleanliness goes up by an order of magnitude rather than a few percent.

02

Targeting on a signal nobody else can see lifts positive reply rates by two to three times

which carries through to the same lift in closes and cuts acquisition cost by the same factor.

03

The signal is yours rather than rented

so it compounds into a moat instead of a tactic anyone can copy next quarter.

Built and running

Royal Energy

PhaibionLive
  • The question was whether a building already had solar panels on its roof, which no data provider sells.
  • The build took a latitude and longitude, pulled a satellite screenshot of that roof from the Google Maps API, ran the JPEG through a machine-learning model trained on roofs with panels and roofs without, and returned a plain true or false to the calling API.
  • That verdict was pushed through Clay and used as a filter, splitting the market into roofs that were already somebody's customer and roofs that were not.

Want this one built into your business?

Thirty minutes, no deck. We map what you run today and tell you what it would take.

Book a call →