Content Copyright © 2015 Bloor. All Rights Reserved.
This blog was originally posted under: IM Blog
Alation, from the company of the same name, is an interesting product. The company calls it a “data accessibility platform” in the sense of making data accessible to users rather than in the sense of disabled accessibility, but in any case it isn’t exactly clear what “making data accessible to users” means.
Fundamentally, Alation crawls through your existing environments – (multiple) databases, APIs, documents, source code and so forth – to discover relevant contextual metadata about the data you have, how it is used, for what purpose, who uses it, whether it is up-to-date, whether it is trustworthy, and so on. It can also import information from existing metadata management tools. The results, which can be updated on a scheduled basis – daily or weekly would be typical – are stored in a relational database but logically manifested as a graph, so that you can see the relationships that exist between and across the data and its context. The system also creates a catalogue against which you can perform natural language search, and which also support a social collaboration layer that analysts, business users, data stewards and IT personnel can all share.
The company sees four use cases. The first is for data search and discovery where you want to find data that is relevant to a particular task. In this example, the software not only allows you to discover tables, reports, attributes and so forth, based on natural language processing, but also to identify subject matter experts within your organisation. The second use case is collaborative analytics. Here the product provides a power query tool that provides “smart suggestions” based on machine learning principles to discover and share queries. The intention is to bypass the lengthy learning curve that new users have who are unfamiliar with your data.
While those two use cases are user oriented the remaining two are enterprise oriented. The first of these is for data optimisation where the software introspects existing queries and reports and derives statistics about duplicates, query performance, unused data and so forth. Finally, the machine learning and discovery capabilities in Alation can be used to support data governance and compliance initiatives.
The company has actually been in business since 2012 and has been acting in stealth mode until now. Its most notable existing client is eBay.
The difficulty that Alation faces is that it is in a market of one. Evangelism is always difficult in such circumstances. Also, I am not too keen on the description of the product as a data accessibility platform although I confess that I have not thought of anything better. One particular issue is that Alation might be seen as overlapping with data preparation products. This is true, but only at the margins. Indeed, Alation sees data preparation vendors (and, indeed, BI suppliers) as potential partners. I agree. In an ideal world I would actually like to see a single product that combined the facilities of Alation with those of a data preparation platform. The problem with the latter is that while they allow you to manipulate the data prior to analysis they don’t actually help you to find what data you might need in the first place, which is exactly what Alation does.
So, it will be interesting to see how Alation develops. I think it has a significant future.