Apify: web scraping, data extraction, and automation tools for AI

What do you like best about the product?

Apify offers an incredible variety of scrapers, but what really stands out is how seamlessly it connects via MCP. Being able to trigger actors and pull high-quality data directly inside AI-driven workflows and agents has made data extraction faster, more practical, and truly automated end to end. They offer $5 in monthly credits, which is already enough to run some tests and extract quite a bit of information. For new users, you can get an extra $5 by completing tasks like following their page, among other things.

What do you dislike about the product?

The initial learning curve can be steep. At first, the dashboard may feel overwhelming and confusing because of the sheer number of features, settings, and available actors. This can make it difficult for newcomers to quickly figure out which specific actor or setup best matches their particular use case.

What problems is the product solving and how is that benefiting you?

I'm using it for various purposes in my workflow, from extracting potential contacts to integrating with specific tools within my applications. Before standardizing on Apify, our web data extraction was fragmented and high-maintenance. Apify is solving problems on the business side, with market research and lead extraction, as well as extracting public information that is used in our application.

What do you like best about the product?

What I like most is how quickly I can set up and run web scraping projects without managing the infrastructure myself. The ready-made Actors, scheduling, proxy options, and API integrations make it very convenient for regular data collection. Overall, the AI and automation features have been useful for getting scrapers running faster and handling repetitive tasks. They work well for common scraping needs, though complex websites still require some manual adjustments and testing.

What do you dislike about the product?

Pricing can get expensive when you’re doing high-volume scraping, and troubleshooting failed Actors sometimes takes a bit of time. I also think some of the more advanced features could be explained more clearly, since they can be harder for new users to understand.

What problems is the product solving and how is that benefiting you?

Apify helps us automate web scraping and data collection that would otherwise require a lot of manual work and custom infrastructure. It saves development time, keeps data updated automatically, and lets us focus more on our core application instead of maintaining scraping systems.

What do you like best about the product?

Aside from the fact that it has a rich library of actors, I love their MCP connectors, which make it very easy to get AI agents to do things for me.

What do you dislike about the product?

The store is deep but uneven per source. For Canadian real estate listings I ended up wiring three different community actors for the same site, because no single one returned everything I needed - one for search, one for enrichment, one for detail pages - each with its own input schema and field shapes. And there's no way to know what fields an actor actually returns until you pay for a run to find out, so I wrote my own probe script for that.

Cost visibility is the bigger gap. run-sync-get-dataset-items hands back the items but no usage figure, so in a multi-tenant app where I attribute spend per tenant, every dollar number I report is an estimate rather than a billed one. A place-name normalization bug on my side once tripled my paid runs on a daily cron and nothing flagged it. Returning usage on the sync endpoint, plus per-actor daily ceilings and a spend anomaly alert, would fix both. Synchronous runs are also slow enough that two of them plus enrichment crowds a 300s serverless function.

What problems is the product solving and how is that benefiting you?

It gives my agents a reliable way to get structured listing data out of sites that have no public API. Instead of maintaining my own scrapers and fighting anti-bot measures, I point a scheduled job at an actor and get clean records back. That turned a nightly data-collection problem into a config change, and let me spend my time on the agent logic and the domain math instead.

What do you like best about the product?

"What I like best about Apify is how easy it makes web scraping without needing to build everything from scratch. I use the Google Scraper actor as part of my n8n automation workflow, and it integrates smoothly — I just call the actor and get clean, structured data back that I can use directly in my automation. Setup was quick, and I didn't have to maintain any scraping infrastructure myself. The reliability and consistency of the results have made it a core part of my workflow. It's saved me a lot of development time and effort."

What do you dislike about the product?

"The only downside I've experienced is that the documentation could be a bit more beginner-friendly, especially for someone integrating Apify actors into external automation tools like n8n. It took me some trial and error initially to get the setup exactly right. Pricing for higher usage volumes could also be a consideration for smaller projects. That said, once everything is set up, it runs smoothly and reliably."

What problems is the product solving and how is that benefiting you?

Before using Apify, I struggled with building and maintaining my own web scraping scripts — handling proxies, rate limits, and anti-bot measures took a lot of time and constant maintenance. Now, I use Apify's Google Scraper actor as part of my n8n automation workflow, which handles the entire scraping layer reliably. This has resulted in significant time savings on development and maintenance, more consistent and structured data output, and a much more scalable automation workflow overall. What used to take hours of scripting and debugging now runs automatically and reliably in the background.

What do you like best about the product?

The store is the whole thing for me. I'm in sales, not engineering, and I had a working scraper running in about twenty minutes without touching code — searched Google Maps, picked the actor with the most runs, typed in a location and hit start. First usable CSV came out before I'd finished my coffee.

The part I didn't expect to care about is that it just handles blocked requests, retries and pagination on its own. I ran one job that pulled a few thousand records over an afternoon and never had to check on it. Output columns are consistent every run, so my contact sheet doesn't need cleaning up each time, and our developer wired the API into it in an afternoon so results land automatically now.

What do you dislike about the product?

Pricing is the main one. It's usage based and I still can't tell what a run will cost before I start it — same actor, same search, different week, and the compute units come out way off. There's no estimate on the input screen and no warning while it's running. I got a couple of bills I wasn't expecting before I worked out to set memory and timeout limits on everything.

The other thing is that store actors go stale. Two I was relying on broke when the target site changed its layout. One got patched within a week, the other still comes back empty instead of throwing an error, which took me a while to even notice. You can see run counts and star ratings before you pick one, but not when it was last updated or whether it's currently working.

Docs are also written for developers. Plenty of Node examples, not much for someone who just wants to configure an existing actor.

What problems is the product solving and how is that benefiting you?

Prospect list building. I run corporate catering sales for a restaurant group, and finding the offices and hotels worth calling in our delivery area used to mean hours of copying from Google Maps by hand. It was painful enough that I'd put it off or only get halfway through, so a lot of the accounts in our catchment were never actually reached.

Now that's an afternoon instead of a project, and I'm working off a much bigger contact base than I'd have built manually. Early days still, but it's running at around two new corporate bookings a month off lists that started in Apify — small numbers, but those are bookings we wouldn't have had otherwise.