Measuring What Matters

The Vietnam War was tragic, it went on from 1955 to 1975. By 1966, the United States had dropped three million tonnes of bombs on the Ho Chi Minh Trail – more than what was dropped on Germany and Japan during World War II. So how did the US continue this disastrous campaign for over 20 years? Behind this was Robert McNamara ‒ Secretary of Defense in the US government ‒ and the system of metrics he created. Success was measured in terms of body counts and kill ratios. Robert McNamara was a numbers guy, he believed numbers could solve all problems. During the war, McNamara tracked every combat statistic he could, the Pentagon set up dashboards to measure progress, the primary data being “kill ratios”, expressed as a ratio of casualties ‒ Viet Cong against the US. McNamara felt that he could comprehend what was happening on the ground by staring at spreadsheets so he created a mountain of analytics to guide the war’s strategy. In one of the episodes of the Netflix documentary ‘Vietnam War’, a despondent war veteran lamented: “If you can’t measure what’s important, you make what you can measure important”. This is also true of business and the metrics we measure.

And the paradox today is that with an unprecedented amount of data, tools, and analytics at our disposal, we are overwhelmed with reports and find it difficult to make meaning of it all. For instance, measuring impact of marketing interventions has never been easy and the digital revolution has compounded complexity, drastically increasing the number of touch points between companies and consumers while raising the expectations of senior management, among others, that everything is measurable. New tools and techniques are invented daily. So how does one wade through this complexity and the dozens of reports we are being bombarded with? Here are some tips I’ve found useful.

*Do three simple checks. Does the data offer value ‒ short-term or long-term? Can the insight or data be actioned upon and do all stakeholders recognise the data source and comprehend the information. If a report or dashboard fails this test, be ruthless with culling it.

*Create a measurement and dashboard architecture. To escape the tyranny of random facts and avoid senior leaders or wider stakeholders reviewing data they don’t understand fully, it’s useful to have a measurement and dashboard architecture. Grouping data and reports into three buckets – strategic, functional, and operational – may be helpful. Key business outcomes, important to the CEO or CFO, may be grouped in the first bucket, for example, revenue, spends, ROI, long-term trends in traffic or brand health, price competitiveness or margins. The functional bucket could have data that drive specific functional actions, for example, campaign traffic, cost-per-sale or brand-lift etc. The operational bucket would have the minute details needed to manage day-to-day or weekly operations. 

*Choose your tools wisely and stick with it. The tools you select need to be consistent, relevant and add business value. For instance, push for consistent tools to be used across business units and establish a common currency of metrics that will be reviewed. The tools chosen need to be relevant and fit your organisation’s philosophy and strategic direction.

*Beware of report creep. A report may have been requested for in a specific business context but it continues to be circulated even after the context has changed and is no longer relevant. In many large organisations that have multiple legacy systems, compiling reports or creating dashboards need time and resources to be invested in. Report creep can be wasteful. It’s useful to regularly review reports being circulated to check its relevance. In one of my previous roles, when circulation of some reports (that had long outlived their usefulness) was stopped, the recipients didn’t miss these reports and this simple action freed up resources for some other useful projects.

*Invest in building the right culture.  All data is of no use if the culture to use it is not created. Culture eats strategy for breakfast. Create the right forums for review, get the right people in and standardise templates – the last thing you want to do is to spend time on explaining templates or have people hunting for data.

Data is important because winning means keeping score and keeping score requires a scorecard. We have learnt this from our childhood, even a simple game of galli cricket uses a score card. Here are some pitfalls to avoid.

*Setup what success looks like. Do this so that you’re not distracted by random data points.

*Set the right benchmarks. This helps set expectations and you don’t make frequent and random course corrections. Put adequate stretch but also be conscious of the business environment and the assumptions you have made.

*Be Roughly Right versus Precisely Wrong. Don’t spend an inordinate amount of time in generating data and metrics.

And finally remember that you still need to make the decision. Data doesn’t talk – people do. Data can’t decide – you do. Not everything can be explained by data. Not everything that’s important in marketing (or business) can be measured, not everything that can be measured is important. Sometimes a fetish for data can distract us from developing a richer picture, from asking questions or having conversations to understand what’s really going on.

---