What is the pricing model? (eg per request?) Roughly what is that cost?
Is there an "all-you-can-eat" access level that other organizations have?
How granular are the data fields per record (/person) ?
Do requesters generally cache the retrieved data on their own stores, continually re-retrieve it, do analysis in large batches, or what? Would we expect to see big caches of this data sitting around outside of Lexis's vaults?
> What is the pricing model? (eg per request?) Roughly what is that cost?
Across all the products I've used (web, batch, API), the price is per-request (where a batch is multiple requests). The actual price can vary by several orders of magnitude depending on what you're requesting. If we're talking about list pricing, it ranges from about 10 cents to tens of dollars per request. There are significant volume discounts available, of course.
>Is there an "all-you-can-eat" access level that other organizations have?
I have not seen one myself and I doubt they offer one even to their largest enterprise customers. However, I am fairly certain that they offer it to certain government agencies. For example, I have seen federal government RFP's that require it.
>How granular are the data fields per record (/person) ?
It depends on the kind of response/report you're looking at, but as a general matter they are highly granular. For example, provided the data are available, I could itemize a list of every car you've ever owned by make/model/year, infer whether you're in a same-sex relationship, or see what the grounds were for granting your divorce.
>Do requesters generally cache the retrieved data on their own stores, continually re-retrieve it, do analysis in large batches, or what?
Aggressive caching is an absolute requirement for API users. That is proximately because Lexis's server does not "remember" your requests. They are logged and have unique transaction ID's but if you run the same search twice they will not check the request against a log cache to see if it's the same as one you ran recently. Instead they'll just charge you a second time (a very easy way to accidentally run up a big bill when running tests against a production endpoint). In combination with the high price per request, you'd be stupid not to cache the whole response.
>Would we expect to see big caches of this data sitting around outside of their vaults?
Yeah certainly. You can't arbitrarily run requests and store the data (e.g. in order to resell it), but you can store the results of requests you've made.
I don't think you'll get everything they have on you this way, but I got a printout that included about a dozen individual reports when I made my request.
Is there an "all-you-can-eat" access level that other organizations have?
How granular are the data fields per record (/person) ?
Do requesters generally cache the retrieved data on their own stores, continually re-retrieve it, do analysis in large batches, or what? Would we expect to see big caches of this data sitting around outside of Lexis's vaults?