Hindsight, what a wonderful thing. Reflecting on lessons learnt from a situation, after it has happened. Useful, right? Well, we’re challenging that idea with our concept of ‘pre-mortem KPIs’.

Businesses use key performance indicators (KPIs) to evaluate how effectively they are working towards achieving their objectives. They tell the status of something important to a business, normally after an issue has become apparent. However, we know KPIs present much greater value if they are used to indicate a situation that could happen in the near future.

That’s where pre-mortem KPIs come in. They highlight issues that may arise if certain actions are, or are not, taken. They enable a business to avoid areas of failure, saving them considerable time, energy and potentially money, trying to rectify things again after the damage has been done.

Think about measuring the health of your business. Do you want to be a doctor, who identifies early signs of future issues? Or, a forensic pathologist, who determines cause of death in post-mortem? Why do a post-mortem after a problem has occurred, when you could do a pre-mortem before it happens?

Pre-mortem KPIs Prevent Future Failings

Instead of waiting for issues to occur, pre-mortem KPIs look ahead to establish what could go wrong, before it does. Why wait for problems to happen when you can prevent them?

The value of pre-mortem KPIs is that they give you ‘pro-spective hindsight’ – which is what KPIs should deliver and the type of KPIs that our 4dBI methodology creates. Our unique 4dBI method builds pre-mortem KPIs through the application of its framework, which ensures a business’s goals and processes are reflected in the information it consumes.

In online retail, for example, it’s essential that orders are packed and shipped on time to meet consumer expectations. A common KPI here would be ‘orders pending dispatch’, or something of a similar nature. They’re likely to use ‘red, amber and green’ performance indicators that highlight if there is a backlog of orders that need to be shipped. However, the issue here is that the information provided is reactionary, only showing the problem when it needs to be fixed. Here, it would be far more valuable to know if there is likely to be a backlog of orders before that problem occurs, so the right actions can be taken to prevent it.

Avoid Escalations with the Right Information

When you look at how well your business is performing, you’re likely to look at the problems that have occurred – those which have hindered your progress in successfully meeting your goals.  On reflection, you’ll probably think they should have been ‘nipped in the bud’ before they inevitably escalated.

It’s important to note that this isn’t full-blown predictive analytics (even though data is being analysed to make predictions about future events). In fact, it’s much simpler. It’s the method of combining multiple KPIs to create a deeper awareness of potential situations, so you can stay on track to meet your business goals, evading negative impacts on the way.

A single dashboard often contains multiple KPIs, each individually delivering the information needed to create an outlook on potential future issues. Yet, this requires you to understand the correlation between KPIs, before joining them up into something meaningful. In contrast, pre-mortem KPIs draw on all the relevant dashboard information and their knock-on effects. This allows you to gain a clear picture of what can happen, eliminating the need to manually piece it all together. When you understand how multiple KPIs can reveal a more insightful picture, there’s a much higher chance of spotting, and preventing, problems.

If you are a doctor, you don’t rely on just one symptom for a diagnosis. Your patients wouldn’t be very confident in you if you did. You analyse the whole situation to establish the best methods required to prevent it from worsening. Likewise, combining multiple KPIs enables you to look at the ‘bigger picture’ so you can see how different factors combine to point towards certain situations. You wouldn’t wait for disaster to strike before you relied on a post-mortem to show what the problem was, or to show you how you how easily avoidable it could have been.

So, before the red warning light in the distribution centre starts flashing, look at the other data that’s available such as ‘order volume vs the same time last week’, ‘delivery capacity remaining’ and ‘delivery service issues’. These all play a significant part in the way your business operates and the status of them combined will tell you if changes need to be made.

The Right Information Was Already There

In most cases, your current BI technology (when used the right way) will provide you with all the right information, your people just need to know where it was.

Of course, it can be difficult for them to initially detect the early warning signs of problems being missed before they turn into real business issues. Too often, the crucial joining-up of available information occurs after a disaster has struck, which leads to businesses performing their own version of a post-mortem analysis. Ironically, most of the information used in these post-mortem analyses was available long before the problem occurred. Which means it could have been foreseen.

By considering the kind of negative situations that could occur, you can then think about what information a post-mortem analysis of that problem would need – so plan ahead! That information then needs to be pulled together in advance as pre-mortem analysis – in the form of KPIs that provide early indications of future issues that may be on the horizon. Know the information you need, and use it the right way, so you can align it with your goals, and create BI solutions for every eventuality.

About the Author

Phil Husbands

Phil Husbands

I show organisations of all shapes and sizes how real business intelligence isn't just about technology. As the Data and Analytics Director at Acrotrend, I help businesses to turn their data into truly valuable reports and information, by aligning BI and data technology with their goals, people and processes.

Leave a Comment