Most developer tools get used a handful of times. Someone finds your API, tries it on a test case, maybe runs it a dozen more times, then moves on. That is the normal pattern. Out of 38 actors I have running on Apify, most average 5 to 20 runs per user. Respectable numbers.
But five of them break the pattern completely. These five average 100 to 260 runs per user. Not because of better marketing or a viral tweet. Because they solve problems that require bulk processing by design.
The Numbers
Here is the actual usage data from my Apify dashboard:
| API | Runs Per User |
|---|---|
| Domain WHOIS Lookup | 262 |
| Google Scholar Scraper | 230 |
| AI Content Detector | 132 |
| Website Tech Detector | 126 |
| Email Validator | 105 |
Compare that to something like the LinkedIn Employee Scraper, which has 37 users but averages about 17 runs each. LinkedIn users grab the data they need and stop. WHOIS users feed in hundreds of domains every single session.
Why These Five?
The common thread is not the subject matter. It is the workflow. Every one of these tools plugs into a process where the user already has a list and needs to process all of it:
Domain WHOIS Lookup (262 runs/user): Security researchers and domain investors run this on batches of suspicious domains. When a phishing campaign registers 10,000 domains with similar naming patterns, someone needs registrar data, creation dates, and nameservers for every single one. That is not a one time task. New domains appear daily.
Google Scholar Scraper (230 runs/user): Academic researchers doing systematic literature reviews or bibliometric analysis. They need every paper matching a query, with citations, h index scores, and author profiles exported as structured JSON. One research project can require pulling data on thousands of papers across multiple search terms.
AI Content Detector (132 runs/user): Content moderation teams, academic integrity offices, and publishers who need to scan entire content catalogs. Checking one essay at a time is pointless when you have 500 submissions or 2,000 product descriptions to verify. The bulk API call is the only thing that makes this practical.
Website Tech Detector (126 runs/user): Sales development teams that need technology intelligence on their entire prospect list. If you are selling a React migration service, you need to know which of your 3,000 target companies still run Angular or jQuery. Feed in the list, get back frameworks, CDNs, analytics tools, CMS platforms in clean JSON.
Email Validator (105 runs/user): Cold outreach operators who clean their lists before every campaign. A 5% bounce rate destroys your sender reputation, so smart operators validate 500 to 5,000 emails before hitting send. They do this before every single campaign, not once.
What This Means for Builders
The lesson is simple: if your API solves a problem that people encounter once, you need constant marketing to keep new users flowing in. If your API solves a problem that people encounter in batches, repeatedly, you get sticky users who come back on their own.
None of these five APIs went viral. None of them got featured in a newsletter. The WHOIS lookup has 7 total users. But those 7 users have collectively run it 1,837 times. That is revenue without marketing spend.
The best APIs are not the ones with the most users. They are the ones where each user cannot stop running them.
Try Them
All five are live on the Apify Store under my profile (george.the.developer), priced per call with no monthly subscription. Domain WHOIS at $0.005/lookup, Scholar at $0.004/paper, AI Detector at $0.003/text, Tech Detector at $0.005/site, Email Validator at $0.002/email.
Built in Nairobi. 38 actors, 700+ users, 14,000+ total runs.
Top comments (0)