ROI: Return of Intelligence

by | Nov 28, 2016

Over the years I’ve become inured to the request to “…find the ROI…” for x, y or Bob campaign. I’ve given up arguing the point about it. It’s got its uses, but to be honest, it never answers the question you’re really asking.

It’s like playing the American TV game show jeopardy where you’re given the answer. “So, Darlene, the answer is ‘I am john Rambo’, and the question is..?” “The question, Hank, is ‘Who was Liberace?’” It is a question and it is an answer, but they just don’t belong together.

The problem we have, is people don’t see they are asking for an ROI as the answer to a question they have posed, but not voiced.

To get to the bottom of why ROI doesn’t really work, let’s look at what the question usually is. What will normally be required is to know is how the campaign performed. People will also want to know who performed. Again we have two questions masquerading as one: the first is a question about financial performance and how it affects budgets and forecasts, and the second is a question about customer performance, which can be used to plan tactics and strategies going forward.

I know I’m not a technically trained statistician, but even I can see that you can’t answer both these questions with one measure. It’s really not possible to kill two birds with one stone, despite what your Granny repeatedly told you.

We can take the first one as this is the easiest one to tackle. As the person holding the purse strings, what I really want to know is whether I can rest easy because the money’s coming in, or if I need to fill a gap in my projected annual income. And to make it even clearer, what I really need is to know how much money this activity is going to put into the kitty. The easiest and most straightforward measure here, is net income. Net income is a no-fuss approach to life, and does exactly what it says on the tin. This is how much money your efforts have earned.

What I like about net income is that it dispenses with the layer of mystery that you have with ROI and shows you where the problem lies: cut your costs you get the same amount of money. When my suggestion to reduce the outgoing volume of an activity is met with fierce resistance, you can guarantee that people are only looking at ROI. For me this is a quick win; a simple spreadsheet will prove that the amount of money going back to the organisation is exactly the same when you cut out all the dead cost.

The performance of customers or supporters (if you are a charity) is a harder nut to crack. I personally use a score, which was shown to me many years ago. It’s the average gift x the average response of a segment. Sometimes you have to multiply the answer by 10 or 100 to get a figure that you can properly differentiate, but it’s a great way of seeing which segments are performing best, without getting distracted by money. For instance, a small segment may be working really hard, but because of low volume not be contributing much to the total income. Well, now you know these are good responders, you just need to get more of them.

Always easier said than done, but it’s worth bearing in mind that you might want to put some effort into these people, finding more of them or creating more of them – put the effort into getting the second purchase or signing them up for a specific service.

Whilst this blog is really just my rant about ROI and how I don’t like it, what it also shows, in a roundabout way, is that you need to understand what the question is before deciding on how to answer it.

I had to learn the hard way, but now make sure that when I’m working with someone who is new to the Data Nerd game, I really challenge them and get them thinking about what’s being asked and to spot the times when it’s an answer disguised as a question that’s being posed. But I can only do so much – after all, I neither speak to everyone nor train every Data Nerd. For you non-data bods, you should be thinking about what you actually want to know, rather than blurting out the bog standard request. You might be surprised at what you find and how it can change your view of your campaigns.

If this all sounds too much and you need some help sorting out the wheat from the chaff, then give Julie or Charlie a call on 01462 713444 to chat through what you want to know and how we can get you the answers.

In the meantime, I’m going off to start a Data Nerds Are Sexy campaign, whilst Googling statistical theories and SQL scripts. Alas, I may be fighting a losing battle.

Share This