Marg Diffbot

Impressive database of information curated from across the web

There is a bit of a learning curve to the Diffbot Query Language, but it's worth it!

Recommendations to others considering the product:

What problems are you solving with the product? What benefits have you realized?

We have used Diffbot for several years, their API for text extraction is extremely powerful and accurate. It has become an important part of our data processing pipeline. Their API(s) allow us to convert unstructured HTML data into information we can ingest and store.

Their support is also very responsive and has always provide us with value answers and feedback when needed.

They also provide with a web interface to define custom rules, that functionality has also proved very useful, however its UI can be not very intuitive sometimes.

What problems are you solving with the product? What benefits have you realized?

Diffbot's Extraction APIs and Crawlbot API provide an incredibly valuable, versatile, and simple to use pipeline for acquiring crucial information from web pages that may not have been visited before. The Analyze API makes it a snap to determine if the page in question is a product page or not, and the wide array of elements that Diffbot returns from most pages is exceptionally useful!

In our space, we tend to cover a large percentage of the e-commerce world, and that takes us to many domains that are either irregular, outdated, or less than perfect in terms of function. We've noticed that for those pages, or ones with domains that have sophisticated/aggressive bot blocking techniques that Diffbot will often fail to provide a result (or at least within a minute or two). This can be problematic for a company like ours that explores tens of thousands of domains each day as it can slow down our discovery pipeline that finds new listings and e-commerce domains.

What problems are you solving with the product? What benefits have you realized?

Diffbot can augment data streams for SO MANY industries/use cases. Within ours we're able to keep track of news mentions on universities (from literally all over the web), and enrich leads for outreach. I'm sure there's a ton more we could be doing with Diffbot. But even with those uses the service has paid for itself many times over. It doesn't take many saved work hours to justify the $299 price tag...

To tap into the full power of Diffbots offerings you do need a technical team member. (But for what service is this not the case?) Basically you can deal with pre-extracted sites (of which there seem to be millions) with the Knowledge Graph and Enhance. If you want to crawl a specific site repeatedly you'll need to at least know hot to make an API call.

What problems are you solving with the product? What benefits have you realized?

We've been using both the Knowledge Graph and Enhance products. We use the Knowledge Graph for a wider search, finding individuals with certain job titles at certain orgs. Then we enrich those profiles with Enhance, together it's a great market research and lead enrichment set up.

We don't need all of Diffbot's offerings. (At least for now.) Their APIs and crawler aren't super applicable to our use case at the moment. With that said, seeing what type of well-formed data is returned from other Diffbot products makes us think we could find a use for these down the road. We aren't a technical team. So this aspect of Diffbot's products isn't really applicable to us... but from what I understand we should be able to easily find an individual who can help us make better use of Diffbot's more technical products.

What problems are you solving with the product? What benefits have you realized?