Simon Price | Managing Director
For any business analysing information can seem complex, but it needn’t be. Here are my top ten tips to avoid over-complicating your data.
- More data isn’t necessarily better. You don’t find the needle in the haystack by adding more hay.
- If you handle personal data, make sure you’re clued up on GDPR (General Data Protection Legislation) before it comes in next May. The new rules restrict how long you can hold information and how you use it, and not being prepared could result in a huge fine.
- Data can answer questions, but it doesn’t ask them. You need to know what you want to find out before you start work.
- It’s important to understand how noisy your data is. If a number improves by 10% is that a real change or could you be reacting to a fluke?
- If you don’t understand where a number has come from, there’s huge risk in applying it. Do you know exactly how all of the numbers on your key performance index (KPI) report are calculated?
- Always chart your data. Unexpected data and errors that can hide in spreadsheets will leap out when you visualise the same information.
- Beware of analysis salesmen who sound clever because they make things complicated. “If you can’t explain it to a six year old, you don’t understand it yourself – Albert Einstein”.
- If you’re still using Excel for all of your data needs, take the time to have a look around. There’s a revolution happening in data analysis and many of the best tools are free.
- Any time you see a % stat in the newspapers, flip it upside down and see if it still looks dramatic. “20% of people do x!”, so 80% don’t.
- We’re really bad at guessing what the UK is actually like and often get it wrong, so we need surveys to help. Recent research by Ipsos found, when you ask people to guess what proportion of the UK is happy, we guess 47%, but actually 92% say they’re happy.[i]