One of the results of the rise of ‘big data’ has been to push data-driven decision making higher up the corporate agenda. Whether this is for proper strategic planning, purchasing ROI, or just good old personal development; showing that decisions are based on data is now an accepted way of being seen to be smart.
Of course this is a good thing, and I always endorse decision making that is based on correct analysis on enough of the right data. Using evidence gives certainty; it brings objectivity to the subjective, and cuts through the many competing opinions. However, there needs to be enough of the right data and it needs to be analyzed in the proper fashion.
The push towards using data for more and more business decisions means that people may use wrong or incomplete data simply because it’s easily accessible by existing analysis tools. Ultimately, this causes people to make decisions based on data that is readily available and easy to analyze, such as social media data, but rarely are they looking at the full picture. Over the past few years this has become even easier to do, and once it begins, it becomes part of an ongoing process.
In the worst cases this data is used to develop Key Performance Indicators (KPIs) that require additional resources in order to meet goals. Key members of staff may then end up putting significant amounts of time into these KPIs, but without stopping to re-evaluate if these goals are truly helping the bottom line. In these instances a circular pattern emerges and it’s difficult to dislodge and rethink the strategy and objectives.
The most common example I come across in my day-to-day work is setting marketing strategy based on social media data. There are a few reasons why this is happening.
- Purchasing departments are increasing their involvement in the decisions over which marketing tools and services to use.
- Social media data has ease of access: It is API accessible, and can be neatly extracted and searched.
- Social media data comes packaged with neatly measureable metrics such as mentions, retweets, shares, likes and so on.
- Getting data for marketing has traditionally relied on focus groups, feedback forms and so on which is hard, inefficient and expensive for the quantity of feedback you get.
It is very tempting to pull this data, develop insights from the data and create KPIs based on it. As the majority of employees are allocated goals of increasing KPIs by 20% per quarter, the whole exercise can become a hamster wheel of pointless activity. But common thinking states that since it can be measured and turned into data, everything is operating fine.
How do we avoid this? We have to think about what data would be truly informative and useful, even if it isn’t immediately obvious where this would come from. Once even the most hard-to-get-data is received, what type of analysis should be done on the data? Even if the analysis tool or technique isn’t immediately obvious, still ask the right questions to get the analysis needed.
Good decision making has good data at its core. Great decision making has good data combined with the right analysis. Don’t get trapped in the hamster wheel of pointless activity.