After building and maintaining 35 Apify actors in production, I've learned which ones consistently deliver and which ones eat your credits.
This post covers the 10 actors I actually use every month, with real stats from my usage dashboard: run count, success rate, and cost-per-result.
Why This Matters
Apify has hundreds of actors in the store. Most are abandoned side projects or demos that fail on any real input. Finding reliable ones takes painful trial-and-error — and failed runs still cost credits.
Here's what 12 months of production use taught me.
1. Contact Info Scraper
Runs last 30 days: 800+
Success rate: 99%
Cost: ~$0.005 per result
What it does: Extracts emails, phone numbers, and social handles from any website or list of URLs.
This is the one that pays for itself. Feed it a list of company websites and get structured contact data back in minutes. Works on JavaScript-heavy sites. Handles pagination automatically.
Best use case: B2B lead generation at scale — I've used it to enrich CRM records from company LinkedIn URLs at $5 per thousand contacts instead of the $100+/mo SaaS alternatives charge.
2. Google SERP Scraper
Runs last 30 days: 250+
Success rate: 98%
Cost: ~$0.002 per result
What it does: Pulls organic results, featured snippets, ads, and "People Also Ask" from Google Search.
Most SERP scrapers break within 72 hours of a Google layout change. This one doesn't. The key is it uses residential proxies by default so Google treats it as real traffic.
Best use case: Rank tracking for 500+ keywords without paying $200/month for a SaaS rank tracker. I run it every Monday morning on a scheduler.
3. Twitter/X Profile Scraper
Runs last 30 days: 230+
Success rate: 96%
What it does: Scrapes public profile data, follower counts, recent tweets, and engagement rates without the API.
Twitter's official API now costs $100/month for basic access. This gets you everything the API offers on public profiles at a fraction of the cost.
4. YouTube Channel Scraper
Runs last 30 days: 180+
Success rate: 100%
Cost: ~$0.003 per video
What it does: Pulls all video metadata, view counts, engagement, upload schedule from any YouTube channel.
I use this for competitive research — pull all videos from a competitor's channel and find their top-performing content topics. Takes 2 minutes for a 500-video channel.
5. TikTok Shop Trending Products
Runs last 30 days: 64+
Success rate: 95%
What it does: Surfaces trending products in TikTok Shop before they hit mainstream awareness.
This is the early-warning system for dropshippers and brand scouts. A product that appears in trending data Monday morning is usually all over Instagram by Thursday. First-mover advantage is real here.
6. Facebook Public Scraper
Runs last 30 days: 76+
Success rate: 100%
What it does: Extracts posts, comments, engagement, and page data from public Facebook pages.
Facebook has progressively locked down its data. This actor navigates their current anti-bot setup and returns clean structured data from public pages — ads library, business pages, public groups.
7. LinkedIn Job Scraper
Runs last 30 days: 33+
Success rate: 91%
What it does: Pulls job listings with full descriptions, requirements, company data, and posting dates.
Beyond job searching: I use this to track competitor hiring velocity. When a competitor posts 5 senior ML engineer roles in a week, they're about to ship something. That's a sales and competitive signal worth knowing.
8. Reddit Post Scraper
Runs last 30 days: 110+
Success rate: 99%
What it does: Extracts posts, comments, upvotes, and metadata from any subreddit or search query.
Excellent for sentiment analysis, trend detection, and competitive research. Run it on subreddits where your customers hang out and you have a real-time window into their pain points.
9. Google Maps Scraper
Runs last 30 days: 60+
Success rate: 97%
What it does: Pulls business listings, reviews, hours, contact info, and coordinates from Google Maps.
The official Google Places API costs $32 per 1,000 results. This costs about $1 per 1,000. I've used it to build local business lead lists for sales teams and market research databases.
10. Amazon Product Scraper
Runs last 30 days: 50+
Success rate: 66% (known issue — Playwright dependency being fixed)
What it does: Extracts product data, prices, reviews, rankings from Amazon.
Currently has a build dependency issue that causes failures on roughly 1 in 3 runs. The fix is deployed — waiting on a platform rebuild. Once resolved, this should hit 95%+ like the others.
Note: Don't use this for production workloads until the fix ships.
What I Skip
Price-Drop Alert actor — Playwright dependency issue, 0% success rate. Skip it entirely for now.
Healthcare Provider Scraper — Base image mismatch causing most runs to fail. Under repair.
The Real Numbers
Here's the honest cost breakdown for running all these at scale:
| Actor | Runs/Month | Cost/Run | Monthly Cost |
|---|---|---|---|
| Contact Info | 800 | $0.005/result | ~$8-32 |
| Google SERP | 250 | $0.002/result | ~$2-10 |
| YouTube | 180 | $0.003/video | ~$1-5 |
| 230 | ~$0.004 | ~$2-8 | |
| TikTok Shop | 64 | ~$0.005 | ~$1-3 |
| Total | ~$15-60/month |
Compare that to SaaS alternatives that charge $200-500/month for equivalent data access.
Getting Started
All actors I've mentioned are in the Apify Store. Most have a free tier that gives you 50-100 results per run — enough to test the output format before committing to paid usage.
If you want to skip the discovery phase and get pre-configured inputs for every actor above (including the exact field mappings that consistently work), I've packaged them into the $29 Apify Scrapers Bundle — 35 scrapers with quick-start configs, one-time download.
Common Mistakes
1. Not setting ACTOR_MAX_PAID_DATASET_ITEMS
Without this limit, a single run can consume your entire monthly budget if the target site has more pages than expected. Always set this to a safe upper bound.
2. Running on shared datacenter proxies
Sites that block scrapers check proxy reputation first. Residential proxies cost 3-5x more per GB but have 10x better success rates on protected sites. Worth it.
3. Ignoring the "Runs" count in the store
Actor popularity is a quality signal. An actor with 800+ monthly runs has been battle-tested by real users. One with 3 runs is probably a demo project.
4. Not checking the build date
Some actors in the Apify Store haven't been rebuilt in 6+ months. Web scraping targets change constantly. If the last build was over 3 months ago, test it carefully before relying on it.
n8n AI Automation Pack ($39) — 5 production-ready workflows
If you're building a data pipeline and want to compare options before choosing actors, the Web Scraping Tools Comparison post covers the broader stack.
For anti-detection specifically: Web Scraping Without Getting Banned in 2026.
Top comments (0)