eStar chief technology officer Matt Neale asks, is business intelligence an oxymoron or just misunderstood?
Effective use of data brings clarity to decisions, information to strategy, and ultimately - intelligence to business, but there is a common misconception we need to address, that data equals intelligence.
First, what is business intelligence?
Essentially (and there are a few definitions), it's the ability to understanding and act within the business environment.
Business intelligence is used to inform operations, guide strategy and improve decision making. When done well, it can lead to competitive advantage, increased agility, and provide more confidence and certainty on decision making.
Many times, you’ll hear companies and individuals refer to dashboards, data lakes and clever reporting as being their entire 'business intelligence' practice. Intelligence, however, is much more.
True business intelligence requires using data to gain knowledge and make meaningful decisions, both strategically and operationally, to benefit the business.
So what does that actually entail?
Data on its own holds no value unless measured. Measurement of data without analysis and understanding is also of no value unless decisions are made – and to complete the cycle – decisions not acted upon, do not drive progress.
This is commonly simplified to 'Measure-Understand-Improve,' the business equivalent of the classic OODA loop (Observe-Orient-Decide-Act) used extensively in military strategy (and favoured by strategic advisors who really wanted to be pilots!).
The cycle is self-explanatory and rather elegant in its simplicity.
However, despite the clarity this brings - many BI and data analysis practices fall down, not through a lack of data, but through gaps in this process.
Before we get to that, let’s talk about data, we need to get a few things out the way.
To be relevant and actionable, data needs to be three things;
• Current
• Comparable
• Consistent
Currency of data usually means time. Historic values can be current, if they are used to compare to the current state of play. But if data is out of date, or not comparable (we’ll get to that) then it’s not current, and it’s not relevant.
Data needs to be comparable - just as you can't compare apples to oranges, you can't compare EDM click-through rates to conversion rates.
You can correlate, but you can't compare.
However, if you have a predictable annual sales cycle (as most retail businesses do) then Q1 sales in 2020 can be compared to Q1 sales in 2021.
This also speaks to being consistent.
That means the same measurement method, of the same values on a repeatable basis.
If you measure sales exclusive of GST one year, and change to GST inclusive the next, then you lose consistency, and you lose the ability to compare.
But data is usually the least of the problem, it’s making sense of it where the challenge starts.
Moving through our Measure-Understand-Improve cycle now, ask yourself - what measures do you have in place in your business? Are they current? Can I compare them?
Be clear – these are not KPI’s or targets – they are simple measurements taken from data; sales, AOV conversion rate, DIFOT, for example, are all measures.
On their own they have little value, when we repeat the measurement cycle, we start to grow that value.
To really harvest that value, we need to understand our measures, we do this by comparison, against ourselves, against our competitors and industry, or against targets.
This brings us to the point of understanding, and being able to derive trends, make comparisons and then consider how we might go about changing these measures.
Question time again, how often do you analyse your measures – not merely as a look back at past performance – but to set out new targets?
How often do you decide, actively, on ways to change the measured results?
This brings us to the last piece, and by far the most important – improve.
Improvement requires action.
How many decisions are made in your business that are never acted upon? How many of these are based off analysis and conscious decision making from data and measures?
We should be at a point in our discussion where a picture is emerging of how much intelligence is being derived from your BI practise.
If you’re seeing gaps, great! Improve it.
One of the key outcomes from the cycle is to gain advantage. This may be over a competitor, or even your own past performance.
The ability to iterate over this cycle effectively, is what drives improvement. The ability to do this better (not just more quickly) than your competition, drives a competitive advantage.
Which seems like an intelligent thing to do.