Reconsidering Qlik

Written By:
Content Copyright © 2016 Bloor. All Rights Reserved.
Also posted on: The IM Blog

I have always liked Qlik. In a sense its problem – if that is the right word, it is not as if the product and company are not successful – is that historically it has been ahead of its time. At least from a technical perspective.

Firstly, consider that it (they, Qlik Sense is now the primary product) is an in-memory product and always has been. This used to be a limiting factor but it isn’t any more now that everything is 64-bit. So, shouldn’t Qlik get some kudos for being ahead of its time? After all, in-memory is a seriously hot topic and Qlik doesn’t haven’t to retrofit in-memory capabilities like so many other vendors and products (not just in business intelligence and analytics). And designing something from the outset rather than retro-fitting it, probably means that it works better.

Secondly, Qlik products use an associative model of data. I need to be clear about this: the model used for analysis is associative and so is the user experience but data is actually stored using a relational model. In practice, most people don’t understand why the associative model is important. But consider that an associative model is an abstraction (or extension) of a graph model. In a graph model you have entities and relationships, in an associative model you have entities and associations; the difference is that in an associative model you can have associations between associations, where graphs do not allow relationships between relationships. This makes the associative model about as flexible as it is possible to get, and when it comes to exploring data that is really important.

Anyway, that was all a bit of a digression. The point is that graph databases represent the fast growing sector within the database environment. According to Forrester, 25% of leading enterprises will have implemented a graph database by the end of next year. What that means is that people are starting to understand what graph databases are and how useful they can be.

The point here is that Qlik is again well-positioned and right up-to-date, given the interest in graph databases. Provided, of course, that it can and will position its associative approach as akin to graph processing.

In other words, despite that the fact that Qlik is over twenty years old, it is based on technologies that are absolutely up-to-date. That, of course, begs the question of why, when I talk to people about BI and analytics, the first vendor that gets mentioned is almost invariably Tableau? Partly, this is marketing and partly this is because Qlik shot itself in the foot when it introduced Qlik Sense. Not that Qlik Sense isn’t a good product – it certainly is – but the company didn’t do a good job of explaining how Qlik Sense related to QlikView and how the two products would move forward. Fortunately, this has now changed and while Qlik Sense Enterprise is the company’s lead product that company also has a coherent story both for QlikView users that want to continue with that product and for those that want to migrate to Qlik Sense.

So, if you dropped Qlik off your radar because they seemed to be a making a mess of things, you should put the company back on your BI map.

This Post Has One Comment
  1. Very well written article. Qlik has/ continues to be a pioneer in the space for a very long time. Although they appeared to struggle with their dual product strategy positioning, it now seems that they are disrupting the market again as they did back when they first introduced QlikView. I’m expecting big things from Qlik as they invest heavily into R&D and continue to prove that they can meet market demands as they do quarter releases for Qlik Sense.

Comments are closed.