← Builds
Public data

Reach companies before their public deadline

A public dashboard lists every company that committed to a public industry target. Committed companies get 24 months to get the target validated. We calculated who is running out of time, and flagged the ones that gave up.

Works ifYour buyers make public promises with a deadline.

Built for · B2B SaaS · Europe
PythonPrefectTrigger.devNeon
Get the prompt for your coding agent

The system

~/builds/public-deadlines
01Source

Public file

Target dashboard export

One row per target, for every company that committed.

Python
251 in scope
Group rows into companiessome have 5+ rows
02Resolve

Tiers 1-3

Domain, LinkedIn, exact name

Tier 4

Cleaned + fuzzy name

Legal suffixes stripped, trigram match.

Tier 5

Create

Then enriched.

Resolved 26.7% → 99.6%
03Timing

Overdue

Past the 24 months

Committed, never validated.

Urgent

Under 3 months

Approaching

3 to 6 months

Withdrawn

Commitment removed

Invested and failed. Warm, not dead.

04Run

On demand

Scrape, bridge, deadline

Trigger.devNeon
~90s per run
Deadlines bucketed by urgency.

The signal

A company committed to a public target and the 24-month validation window is closing, or it withdrew the commitment.

Why it predicts a purchase

They already decided this matters and put it in public. A closing deadline creates urgency, and a withdrawn commitment means they tried, hit a wall and still have the pressure. Both are warmer than a company that never started.

How it works

01

Parse the file the way it is built

One company can have five or more rows, one per target. Group them into one company without losing which target is in which state.

02

Match like you mean it

Big corporates in a public file rarely match a CRM on name. Legal-suffix stripping and trigram matching, then creating the companies that still didn't match (and enriching them after), took resolution from 26.7% to 99.6%.

03

Do the date math

Committed plus 24 months is the validation deadline. Anything inside 6 months becomes a signal, bucketed by urgency. Withdrawn commitments get their own flag.

Build notes

  • The no-code version silently returned zero rows from the Excel file. That single failure is why every pipeline for this client moved to Python and later TypeScript.
  • Postgres date minus date returns an interval, not a number. Wrap it before you compare it to 180 days.
  • The context note needs to explain the signal. A note that only says the status and classification assumes the rep already knows what the framework is.

Questions

Why target companies that withdrew a commitment?

They already decided it mattered and spent effort on it, then hit a wall. The pressure did not go away. That makes them warmer than a company that never started.

How do you match companies from a public file to a CRM?

In tiers: domain, LinkedIn URL, exact name, cleaned name with legal suffixes stripped, trigram match, then create the company if nothing matches. Every match stores its tier. That took resolution from 26.7% to 99.6%, mostly by creating the companies the CRM never had.

More builds

All builds →
Public data

Detect 163 B2B tools from a company's website and DNS

Your product replaces, integrates with or sells to users of specific tools.

Open→
Public data

Reach companies hiring for what you sell

You sell a service that companies also try to hire for.

Open→
Public data

Find buyers in certification registries

Your buyers can hold a certificate or standard that proves they care.

Open→

The next step

Let's get started.