“Although I do find it odd that Linda finds the numbers/data thing sexy”

As I (maybe somewhat clumsily) attempted to point out, it's not necessarily the data itself but rather the images it conjures up and what you can do with it…

Suppose there was one typo in the example I quoted for Sarah and the data should been

36 – 36 – 37…

I rest my case (for now, anyway)
Twitter: Linda_Mattacks

And as for “Any fool can gather data” – too blinking true and what a hash they can make of entering even the simplest of it…let alone do anything intelligent with it…
Twitter: Linda_Mattacks

It doesn't matter if the question is “who are my best customers”, or “is this data valid”, the process is the same.

To answer either question requires a set of decisions to be made. For a question that is answerable a set of rules can be created and the answers found. The trick is making the process of rule definition quick and cheap. This is where existing rule systems often fall down – slow to run, and expensive to build. By removing those two barriers the software is suitable for both small and large tasks. Areas of application that previously would not have been suitable to the rules approach become feasible.

The system doesn't just dump data – it incorporates a domain specific framework of rules that provides a context, so that it makes data into information.

Volume of data is only a part of the equation – and so although we have experience of sorting out millions of records for the largest companies, we've also found that smaller companies have complex problems that we can resolve too. For example, we recently did some work for an IFA, where the records were in the low thousands – but the complexity and demands for consistency were very high – and the cost of error was correspondingly high.

In the end it's the risk of error that is the key issue. At one level data clean up is an insurance policy – clients are often surprised by the kind of things that show up in their data.

Human beings are good at spotting some kinds of error and very bad at others. Our software is designed so that as well as performing the common types of checks – like key matching and field format – you can create rules that mimic human common sense.

Any fool can gather data or worse, give clients data dumps but few can make sense of the data in ways that are meaningful or useful. That's what Qualia Systems does – makes sense of the data and further, uses rules to process it and turn numbers into unbiased decisions that would take humans much longer and be more subjective.
Twitter: maverickny