Clavis: extending data quality

In the last few years there has been a trend amongst data quality vendors to move towards what is typically known as pre-emptive data quality. That is, to have proactive capabilities that enforces (as much as possible) data quality at the point of entry. This is typically accomplished by having a front-end that integrates with appropriate ERP, CRM or MDM software for the likes of SAP and Oracle.

However, this is somewhat limited. There are a number of functions, such as customer on-boarding and new product introduction, where substantial business processes are involved prior to the data ever being entered into SAP, Oracle or whatever. And these processes involve significant data quality issues which, if you get errors in the data, can lead to significant rework. As an instance of this, suppose that you have already sent the design of your new product on to the packaging department for design of the appropriate packaging, and then you discover that the height of the product is wrong, then packaging will have to rework their requirements. Of course, not all examples of rework are so clear cut as this, but whatever the particular concern, poor data quality within these sorts of processes will inevitably mean lost time.

Okay, so I hope you can see the value in applying data quality inside these sorts of processes. But until just recently there was no way that you could apply conventional data quality products within these sorts of processes because there were no vendors that offered this sort of functionality. Well, as you may have guessed, there is now. This is what Clavis Technology does.

Clavis is Irish and it was set up by the same team of people that brought you Similarity Systems (now part of Informatica), which means that they have a strong pedigree in this space. The company was established in 2007 and released its software late in 2008. It is SaaS-based, which I think is convenient in this sort of environment.

There are two main things you need to know about Clavis’ technology. The first is that it is based on a glossary that can be used by both business and technical personnel. It contains, or may contain, all the relevant reference data and rules pertinent to the processes concerned. I say ‘may contain’ because you can also call out to other sources of reference data via a web service if you want to but typically you will load standard reference data (GS1 for example in manufacturing) into the Glossary, which you can then augment with your own in-house definitions and standards. Note the glossary is hierarchical and supports inheritance, which makes maintenance much easier.

The second thing you need to know about Clavis is where it is getting the data from that it is cleansing. Bear in mind that we are not talking about traditional environments such as databases or applications here. The data hasn’t got that far yet. No, here we are talking about Excel spreadsheets, web forms, Adobe Flex, rich Internet applications and so on: anywhere that anyone might be initially entering data—and note that that data may not be entered by in-house personnel but by customers, suppliers, partners or other external personnel.

That should give you an idea. What Clavis is doing is pushing back the boundaries of data quality in an entirely innovative manner. There is no other company doing this (I usually put caveats on statements like that but this time I think I can pretty sure that Clavis really is unique). If you are interested in pre-emptive data quality, and you have any sort of relevant processes that go on before data is entered into your ERP and other applications, then you really should take a look at Clavis.