Why I won't give a "Top" Metrics List

Modern culture likes to turn things into bite-sized chunks. Part of this is psychological; It's a lot easier to remember a Top 3 or Top 5 than some unordered list. The rise of Buzzfeed is testament to the "top list" mentality.

The problem with metrics are that often there is "background information" that is important and informs the metric. For example, let's take Average Annualized Gift. By taking its background information - total number of gifts, total number of donors, and total amount received - and turning into a ratio, we lose the background. As we saw when I discussed Average Gift, this can completely obfuscate the reasons behind an increase or a decrease.

You can report on this metric, but having that background information is important. How do you know that the increase in average gift is due to an increase in giving, or a decrease in donors? I would argue, quite vehemently, that you are better off reporting on these three than simply on the average value. So where someone may list average annualized gift as a "must-have" metric, I would put three in its place.

I have discussed similar problems with a single retention rate. Your retention rate could not change year over year for subgroups (such as first time donors, repeat donors, etc), and yet your overall database retention rate could change simply due to how many donors you acquired last year. Imagine seeing year over year increase in your retention rate, only to figure out that it had nothing to do with donor love, but everything to do with a year over year reduction in acquisition strategy.

You want a great way to really understand the distribution of donation sizes? Separate your gifts into deciles. That is, break your gifts into ten equal-sized buckets. What are the edges of these buckets? That is, what is the 10th percentile, the 20th, the 30th, etc. There's an extra nine measures for you (10 if you include the highest gift value), and it will tell you far more about "typical" giving than just average or just median.

Of course things can get overwhelming, which is why we do "top" lists. I think you are better off getting a bunch of benchmark numbers, and then picking some specific metrics to target, and focus on strategies to improve those. Is it retention? Measure a few retention rates and let those be your 3 or 5; Maybe first-year donor, monthly donors, and all other donors. Maybe you specifically measure donors who gave two years in a row as well, or the rate of reacquisition of lapsed donors.

We cannot move forward with sophistication in measurement and analysis in the fundraising sector if we focus on 5 nuanced metrics. It's as simple as that.