Under Construction! Please Visit Reserve Page. Page Will Be Available Shortly

Data modelling is at once simple and complex. But it is a fantastic tool for digital planners as well as non-specialists.

For example, you are planning a nationwide digital campaign for Guinness, aimed at drinkers who buy cans from off-licences and supermarkets.

The first question to be answered is: “'What could impact on sales?”'

Intelligent guesswork and experience would lead you to a list which looks something like this:

• Price
• Promotions (money-off offers, BOGOFs, etc)
• Events such as St Patrick's Day, Christmas, an Ireland v England Rugby match
• Perhaps a 'Visit Ireland' TV ad campaign
• An existing or parallel above-the-line Guinness campaign
• Competitor activity (what are Murphy’s doing?)

---

Next step is finding the data. You may not be able to get the exact information you require. For example, competitor information might be tricky to get hold of (the full details on the Murphy's campaign may remain out-of-bounds), but you can make educated guesses and use widely available data from bodies like BARB or Nielsen.

Then you turn to your data analyst, who runs a few simple correlations, a transformation or two with the advertising spend, eyeballs the data and comes up with a model. Then you can build your model, test it (the wonderful thing about digital is that you can test quickly, accurately and safely); refine it; re-test it. Soon you’ll have an accurate data model around which to build your campaign – which in its turn can be tested and refined.