I tried various parsing methods through this service, and the results were mixed. In terms of functionality, it hardly differs from many other services, so it should be evaluated based on its speed. It is excellent for simple tasks, especially considering the free usage. For example, parsing the audience of one group took just a second, while a more complex task—parsing 10 communities with an audience of about 10-15 thousand subscribers—took about 3 seconds. That's not too long, but the service is clearly struggling. When it comes to intersecting audiences, it works very slowly. I tested searching for an audience that is a member of three public groups simultaneously from ten selected large groups (with 2-3 million subscribers each). Out of curiosity, I also compared parsing in Chrome and Firefox. In both cases, it took 6 minutes, even though there was a video open in Chrome and 10 tabs. The search for intersections with 5 entries in the same groups took the same amount of time. So for the system, there’s no significant difference; it just lags when searching for intersections: the larger the audience, the longer it takes. During the test, the script for finding similar communities from the top without various keywords just froze. The search by post works great, parsing 500 people in half a second. However, when searching for activity, there’s no logic: it looks for likes and reposts from the last 10 posts in the group and filters communities of 2 million people in 5 minutes, while running a similar script on the last 30 posts takes 4 minutes. The service is clearly lagging, especially when dealing with large volumes of data and complex tasks. But for analyzing small communities, everything goes quickly and easily. And considering that payment is not mandatory, that’s simply wonderful.
