My most expensive lesson

I love the certainty of data. Data provides the guiding light, gives confidence in the strategy, reduces the influence of emotion.

So long as you view the data honestly.

Data is remarkably easy to either manipulate or find the single point that proves the outcome you want. This is fraught with danger. 

Rather than stop when you find the data point that proves your hypothesis, greater effort should be placed on continuing to search for the data point that disproves your hypothesis.

This will save you time and potentially a lot of money.

I’ve always been very aware of confirmation bias and take great delight in finding the data point that screws the proposal. If you can’t find that data point, then full steam ahead.

I had seen many senior company execs present skewed data to paint a rosier picture and was always so disappointed that they weren’t strong enough to give the honest view.

And then suddenly I found myself sucked into that murky world.

One of most important markets had been slowly deteriorating; high double-digit growth…high single digit growth…low single digit, you get the idea.  We knew we were at a price premium but as a company convinced ourselves that as a premium brand this was not the issue.

To reduce the price would have killed the year (it was only March). It would have been a disaster.

Then we found the saviour data point – the market is shrinking! They audience size is declining and there are less new entrants – hooray! Market share isn’t declining so the issue isn’t the price – phew.

Normal service resumed.

A month later, the spanner-in-the-works data point comes to light (as it usually does). Sure, there is a decline in audience volume but market is still growing and we are losing market share FAST.

The smack of reality stung. All hands-on deck. More detailed performance analysis showed we were haemorrhaging low value transactions. This was actual good news since we could target the price action and reduce the revenue impact. All credit to the marketing team (you know who you are) for launching a state-based test, supported with a 360-degree marketing campaign in record time. After just 2 months the test was rolled out nationally. Performance flipped from -9% to +30%. 

We missed our target for the year, just, but set ourselves up for strong year-on-year growth in subsequent years. It still pxsses me off that I allowed myself to succumb to confirmation bias, pressure and consequences can do that to you. A painful and expensive lesson learned.

Data is the guiding light, but that light can lead you over a cliff if you are not honest.