Stick the kettle on
A question that gets asked of me (and of data focused forums) is “Should we get a new database?”
My response will always be “Why do you think you need a new database?”
I know it’s bad form to answer a question with another one, and I’m not usually known for avoiding answering questions, but the response I get tells me so much more about the situation than the actual words used. However the answer, that comes back to me, is phrased it’s usually a variation of the same theme: the data is rubbish, the organisation can’t use it, it’s holding back our business.
Let’s look at this and break it down into non-techie, simple pieces. Think of a database as something warm and familiar: a cup of tea.
Now I usually have my tea in a thermos mug. Its large volume and heat retaining abilities ensure I get a constant supply of my work day fuel. I could however, tip it into any old container: a regular mug, an old jam jar or my boot. It would still be tea and it would still be hot enough if I drank it quickly. It would be reasonably unaffected by the container, although I admit this might not be the case if I really did sup it out of my boot.
But this is assuming my tea was well made in the first place. What if I’d made a particularly awful brew? Rubbish tea bag, stale water and rancid milk. I could put this concoction into any one of these containers and the outcome would be the same – pretty dire, undrinkable.
I appreciate at this point you’re probably questioning my sanity, but what this particularly trite analogy demonstrates is that the relationship between the contents and the container shouldn’t be confused – they are not intrinsically linked.
Going back to the original question, if your data is rubbish and the organisation can’t use it, if you feel it’s holding back your business – then these are all issues with your data. This is the data equivalent of a pretty bad cup of tea. You can buy the best mug (or database) money can buy but it won’t change the fundamental issues with the tea you are drinking or the data you are trying to use. What is more appropriate is to look at re-structuring your data, removing duplicates or carrying out a suppression audit.
I’m not ever going to rule out a new database, but for most organisations it would be the answer to a different set of issues that probably don’t exist yet. But I’m more likely to start with something a little easier to manage: a fresh pint of milk, refilling the kettle with some clean water and a nice new box of teabags.
So if you can’t get your data to work for you, if you really do feel it’s holding back your business then give me a ring, stick the kettle on and I’ll be round in a bit with the Hobnobs.