ActorStack.dev
ActorStack.dev
Startup job marketExplainer

The column we measured at 0.1% and did not ship

Wellfound exposes a maximum years-of-experience value. It was filled on 2 of 2,229 jobs. Shipping it would have added a field that looks like data and is not.

By Oswaldo Carabano5 min read
A white measuring tape curving across a dark background, showing the numbers 15 to 45.
Photo: Semanur Çoban on Pexels

Short answer

Wellfound exposes a maximum years-of-experience value alongside the minimum, and across 2,229 measured jobs it was present on 2 of them — a 0.1% fill rate. It is not in this Actor's output. A column that is null on 999 rows in 1,000 still counts in a field list, still appears in a schema, and still gets planned around by somebody who assumes a field exists because it was advertised. The minimum, at 31.6% overall and between 12.8% and 75% by category, does ship, with its range stated.

Key points

  • `yearsExperienceMax` was filled on 2 of 2,229 jobs: a 0.1% fill rate.
  • A field list is a promise. Every entry on it suggests a column somebody can build against, and a 0.1% column cannot be built against.
  • Counting fields is a marketing metric in this category, which is exactly the incentive that produces schemas padded with empty columns.
  • `years_experience_min` does ship at 31.6% overall, with its 12.8% to 75% category range published alongside.
  • The same reasoning removed six attachment fields from a different Actor in this catalogue after they measured zero.
On this page5 sections

There is a field on Wellfound that this Actor does not return, and the reason is a number worth stating plainly.

The measurement

Across 2,229 unique jobs measured over 25 roles and all 13 categories, the maximum years-of-experience value was present on 2 rows. That is a fill rate of 0.1%.

Why 0.1% is worse than absent

A column that is null on 999 rows in 1,000 is not a sparse field, it is a field that does not exist wearing the costume of one. It appears in a schema. It appears in a field list. Somebody writes a query against it, gets nulls, and spends an afternoon deciding whether their pipeline is broken.

The incentive this resists

Field count is a marketing metric in this category — “58 fields” is a selling line, and 59 would be a slightly better one. That is precisely the incentive that produces schemas padded with columns nobody can use, and the only defence against it is a rule applied before the count is advertised rather than after.

What ships instead

years_experience_min, at 31.6% overall and between 12.8% and 75% by role category, with the range published rather than the mean — the same treatment as ats_source.

The same decision elsewhere

The Telegram Actor in this catalogue removed six attachment fields after 893 posts across 29 channels produced zero positive cases for any of them — that measurement. Different target, same rule: a column that would always carry the same value is not a measurement.

Frequently asked questions

Why not ship a field that is sometimes filled?
Because at 0.1% it is not sometimes filled, it is essentially never filled — 2 rows in 2,229. A consumer who sees it in the field list will write a query against it, get nulls, and have to discover for themselves that the column was never usable.
Do I get years of experience at all?
Yes, the minimum: `years_experience_min`, filled on 31.6% of rows overall and between 12.8% and 75% depending on the role category. The range is published rather than the average, for the same reason as the ATS field.
Where else was this rule applied?
The Telegram Actor removed six attachment fields — documents, voice notes, audio, stickers, locations and round videos — after 893 posts across 29 channels produced zero positive cases for any of them.

Sources

Every URL below was requested and returned a page on the date shown.

  1. Operator claimchecked 9 Sept 2026
    Wellfound Jobs Scraper — Actor README and input schemaActorStack / Apify Store
  2. Operator claimchecked 9 Sept 2026
    Telegram Channel Scraper — Actor README and input schemaActorStack / Apify Store

Related reading

A laptop screen showing a plain text-mode terminal with a command prompt.
WellfoundExplainer

The Wellfound API question

There is no documented public endpoint for startup job listings. What robots.txt permits, what Cloudflare sits in front of, and which surface is actually readable.

6 min
A desk with stacked legal reference books, loose documents and a newspaper.
WellfoundExplainer

Why no people data

Founders, employees, funding rounds and investors are all on Wellfound and none of them are in the output. What the line is, and why it sits where it does.

6 min
Developers working side by side at desktop computers in an open-plan tech office.
WellfoundGuide

Scrape Wellfound jobs

A walkthrough of extracting startup job listings: how Wellfound's URL-path filtering constrains what you can ask for, which caps actually bound a run, and what arrives in each row.

9 min