01 Aug 2022 · 6 min read

Building an index of recently funded startups

Companies that just raised are the most obvious buying signal there is - they have budget and a mandate to spend it. This project turned that signal into a browsable, daily-updated index using scraping plus a stack of low-code tools. Not a tutorial, more a demonstration of what one person could assemble in a weekend.

The stack

n8n - the automation engine, self-hosted
Airtable - the database
Softr - the front end on top of Airtable
A hosted browser service - for the scraping step

The pipeline

Enumerate the category URLs to cover - gaming, health, fintech, travel, education and a couple of dozen more.
For each category, scrape the funding round listings on a schedule.
Normalise the records: company, round, amount, date, industry.
Deduplicate against Airtable and insert only what's new.
Publish the table through Softr as a filterable directory.

The interesting engineering wasn't the scraping - it was the deduplication and normalisation. Funding data arrives inconsistently: the same company appears with different spellings, amounts arrive in different currencies, and rounds get restated. Without a normalisation step the index looks impressive and is useless.

n8n workflow fetching companies, checking Airtable, enriching and appending funding rows — The pipeline in n8n: fetch → parse → iterate → dedupe against the base → enrich → append.

Airtable base with company name, funding, investor, website, socials, article and industry columns — The normalised record: company, round, investor, links, article, industry.

The Softr front end showing a company card with funding amount, investor and news links — The front end - a searchable card per company, built on top of the same base.