Back to all Blogs

Essential “Laws” for Analytics and Data Mining

Published 01 Feb 2016 by Tim Langley, CANDDi
Read this in about 1 minute

Analytics tools are now so common and so affordable that every business could - and should - be using them. However, big data is useless if you don’t use that data to ask and answer the right questions. One prominent data mining expert, Mr Tom Khabaza, has put together nine laws that he believes are the foundation of good analytics.

Laws for Analytics and Data Mining

Those laws are not just rules to be followed - rather, they are observations that help people understand what data can be used for and how to use it. Let’s take a quick look at the core of the laws.

1 . Your business objectives should be at the foundation of every attempt at data mining. You can’t solve a problem if you don’t know what it is.

2 . You need someone who knows your business to be leading the data-mining project.

3 . You should invest time into preparing your data before you analyse it. Analysing bad or outdated data will produce bad or outdated results.

4 . If you’re trying to use analytics to predict trends, expect a lot of trial and error along the way.

5 . There will be patterns in your data. If you can’t find any, either you don’t have enough data or you’re not looking at it in the right way.

6 . Analysing your data will help you to understand your business better. Be prepared to question things you thought you knew were true.

7 . Good analytics will give you the confidence you need to predict what will happen under specific circumstances, so you can handle any crisis.

8 . You should judge data mining by the results that the business produces in the real world, not by whether a specific prediction was right or wrong.

9 . You need to keep on collecting data and revisiting your analytics. The patterns that are present today may change tomorrow.

Analytics should be an ongoing thing. Make collecting data a part of your routine, and focus on improving the quality of your predictive models. Revisit everything seasonally, and don’t make any assumptions. Your markets will change and your competitors will change, and with those changes there will be differences in customer behavior. Only businesses that are open-minded and willing to adapt will survive. Don’t make the mistake of treating analytics as if it should be ‘one and done’. Keep investing in it to get results.


Back to all Blogs