Apify's Long Tail: 57% of Actors Serve Under 10 Users

Apify · marketplace signal

Across 3,155 tracked Apify actors (May 2026): 57% serve under 10 users in 30 days, median is 7, only 7 actors clear 10,000.

apify.com →

The Signal Census pulse tracker measures 3,155 Apify Store actors with continuous data history through 2026-05-16. Their combined 30-day active users total 405,438. How those users distribute across the catalog is what determines every other dynamic on the Store.

The median actor delivers 7 users in 30 days. The top actor — Google Maps Scraper from the compass creator — delivers 27,806. The ratio between them is roughly 4,000-to-1. That spread is not unusual; it is what a digital marketplace looks like when nothing forces demand to flatten.

The demand buckets

Three cumulative thresholds tell the structural story:

• 57% of tracked actors serve fewer than 10 users in 30 days. The bottom of the distribution is the dominant population of the Store.
• 89% serve fewer than 100 users. Almost the entire catalog is at sub-100-user demand levels.
• 0.22% (7 actors) clear 10,000 users. The “household name” tier is a single-digit population.

The power-law math

Concentration of demand at the top of the distribution is sharper than catalog-share alone suggests.

The top 10 actors capture 113 times their proportional share of demand. The top 100 capture more demand than the remaining 3,055 combined — 77.7% versus 22.3%. By the time the curve reaches rank 100, the marginal actor is adding so little to total demand that the next 3,000 listings together account for less than a quarter of all measured usage.

This is a textbook Pareto distribution with a relatively steep alpha. Apify is not unusual in producing this shape — every digital marketplace with low publishing friction produces something close to it. App Store, NPM, Wikipedia article views, YouTube channel subscriptions, and Spotify track plays all follow similar curves. What is specific to Apify is the magnitude of the spread (4,000-to-1 between median and top) and the size of the absolute floor (median = 7 users per month).

Where the bottom of the curve actually comes from

The 1,786 actors at the bottom of the distribution — those serving 0 to 9 users — are not a single population. The composition matters for publishers deciding where to compete.

Abandoned listings. Actors published in 2022–2024 by creators who have since stopped maintaining them. The 308 actors at exactly 0 users are mostly here. Their listings remain in the Store but the underlying scrapers either break on first run or no longer target a useful surface.

Spray-pattern publishers. A small number of creators publish dozens or hundreds of low-traffic actors as a portfolio strategy — the Q1 2026 lead-extractors census on this site documented contacts-api running 210 actors at 2 users each. The same publishers appear across categories. Their absolute contribution to the long tail is meaningful: publishers running 50+ listings each account for several hundred of the bottom-bucket actors.

Genuinely niche actors. Tools targeting small audiences — a regional job board, a specific software vendor’s API, a particular product catalog — that work but serve a buyer pool of a few customers. These are the long-tail listings that genuinely belong on a marketplace: the addressable market is small, the demand is real, the actor is functional.

New actors not yet found by buyers. Roughly 30–70 new actors ship to the Store per week. New listings start at zero demand and take weeks to months to find their audience if they find one at all. Some fraction of the bottom bucket is new actors in their discovery period rather than failed listings.

The four populations are not separable from public metadata alone. Distinguishing them requires looking at publish date, publisher portfolio size, target type, and recent demand-trend slope — which is the analytical work the pulse tracker does on a daily basis.

Where the top of the curve concentrates

The seven actors clearing 10,000 active users in 30 days as of 2026-05-16:

Four of the seven are Apify-published (Instagram, RAG Web Browser, Instagram Profile, Google Search Results). Three are not — compass Google Maps, clockworks TikTok, curious_coder LinkedIn Jobs. The Apify-published share at the top of the curve is roughly proportional to its share across the rest of the distribution: Apify is a heavy publisher across all tiers, not specifically at any one of them.

What is true at the top: the household-name tier is dominated by general-purpose, broadly-targeting actors against the largest possible surfaces. Google Maps, Instagram (full + profile), TikTok, LinkedIn Jobs, and Google Search — five of the seven — are the largest scraping targets on the public internet by buyer demand. Two more (RAG Web Browser, Google Search Scraper) are general-purpose primitives for the AI-agent buyer that is currently driving the steepest growth on the Store.

Publisher portfolio shape mirrors the actor distribution

The actor-level power-law has a publisher-level analog. Across 209 tracked publishers, the 61 “concentrate” publishers (≤5 actors each) operate 1.8% of the catalog and capture 13.5% of demand — a 7.4× over-indexing on per-actor demand. The 37 “spray” publishers (50+ actors each) operate 72.9% of the catalog and capture 42.9% of demand — a 0.59× under-indexing.

The per-actor density spread by publisher class is 12.5×: concentrate publishers average 339 users per actor, spray publishers average 27. The inflection point is at roughly 20 actors per publisher. Below it, each actor over-produces; above it, each under-produces.

The compass creator illustrates the top of the curve. Five actors, 34,578 users, averaging 6,916 per actor — close to 1,000× the median actor on the Store. The largest spray operators run hundreds of actors averaging 1–2 users each. Both shapes ship to the same Store; the per-actor economics differ by three orders of magnitude.

What this implies for new publishers

The distribution shape forces a specific strategic choice.

Below 100 users is the default outcome. A new actor with no marketing, no SEO investment, and no specific buyer commitment lands in the 1–9 user bucket and stays there. That outcome is 47% of all listings. Publishing alone is not a strategy.

The 100–999 bucket is where craft becomes legible. Actors with clean schemas, accurate descriptions, well-priced PPE, and competent targeting end up here. The 257 actors in the 100–499 band and 41 in the 500–999 band are the population of “actually used” listings that are not yet category leaders. This is where a serious publisher should expect to operate after 6–12 months of consistent work on a single target.

Above 1,000 users requires category dominance or platform distribution. The 53 actors above 1,000 users are there for one of two reasons: they are the canonical actor for a specific target and own the SEO surface for it, or they are Apify-published and benefit from first-party distribution weight. Both paths exist; both take multi-quarter investment.

The top 10 is functionally closed. New actors do not break into the top 10 in a single quarter. They climb through the 100, 500, 1000 brackets first. The household-name tier is stable across consecutive quarterly censuses.

Default to concentrate. A new publisher with no portfolio should ship one or two well-targeted actors first, not ten. The per-actor density math favours fewer, better listings. Spray publishing works only as an SEO-volume play, not a per-actor revenue model — and the underlying economics depend on aggregate platform fees, not buyer demand per listing.

Strategy on the Store has to be designed against this distribution, not against the assumption that publishing is sufficient to capture demand.

Sources

• Signal Census pulse: pulse_details.json (data window 2026-05-16, 3,155 tracked actors, 405,438 sum 30-day active users)
• Apify pricing documentation
• Signal Census Q1 2026 lead and contact extractors census
• Signal Census: Apify Store May 2026 — 25,787 actors, 480k MAU