Further to my blog post “You’ve got directional issues”, I think it’s time for an example of looking carefully at your campaign results – which we’re always happy to help you with by the way.

Back n 2015-16, we were working with a charity which needed some decent insights into their Individual Giving: what was working, what wasn’t and what they needed to change.

Up until we got involved, in spite of valiant efforts on the part of the former IG fundraiser, the data selections were based on completely different criteria with virtually every appeal, so it was impossible for the fundraiser to compare like-for-like across campaigns.

We outsourced their data to a secure data handler, and ran a retrospective re-segmentation across two years’ worth of campaigns. Bingo. Insights.

First of all, it was really easy to spot that the charity’s raffles were being mailed far too wide, resulting in a consistently negative return. What was particularly galling was that the charity had just been advised by their raffle-provider to increase the mailing file (!) and increase the prizes(?).

It made absolutely no sense.

To demonstrate, we ran projections based on the advice. We also offered the charity five alternative scenarios, one of which we could guarantee would meet, if not beat, their income target for the year. With our data insights, we also predicted how we could improve the response rate for each segment, through our data recommendations coupled with our targeted creative – all at our usual competitive fee.

Our first warm appeal doubled their appeal response rate, hitting the result we’d predicted bang on the nose. It also clobbered their raffle response rate, and it raised the most income they’d ever had from a standard-value appeal.

So what have “women” got to do with this?

Well, one casualty of delivering a high-performing appeal is that sometimes our clients decide they can reproduce it in-house or with a cheaper freelancer.

Here’s the result of how that went (August 2016 on the right).

The charity had hired a new IG fundraiser, who literally tried to copy our winning concept from the year before, without understanding it or why it worked. They’d even copied word-for-word slabs of text, even though the case study was completely different and it was going to the same audience. (Er…authenticity and integrity, where did YOU go…?)

In the light of their appalling results, the Head of Dept asked us why we thought their ‘cheaper’ appeal had performed so badly. So we took a look.

What we found was that the huge drop in income from the charity’s cheaper appeal had come from a poor response by women.

That meant we then knew what questions to ask, starting with, “What on earth did you do to your women?!”

(By the way, our record-beating May 2015 appeal isn’t in this graph because the [miffed?] fundraiser wouldn’t release us the data to run a gender split.)

Still, we were able, with all of our other insights, to give them four things to check:

1. Had they accidentally addressed all the women as Mr?
2. Had their women supporters been pissed off by something else the charity had just done, outside of IG?
3. They’d somehow forgotten to put a cheque option into their creative (!!) – could that be a women-specific problem, unlikely as it might sound?
4. Where we’d lasered appropriate ask amounts the year before, in their creative they’d printed ask amounts, which were far higher than most of their female supporters gave. Why is anyone’s guess – it didn’t save them any money. So had they considered picking up the phone and speaking to some of them?

The devil, as well as being in daft false economies, is in the detail. And that’s the insight they still needed help with.

But this charity wasn’t and isn’t the only one lacking a data insight role or function.

If you need help, don’t hide. Do ask.

This particular fundraiser drove us nuts by not being brave enough to ask. Good fundraisers will always ask, and yes, sometimes it can take courage. But we’ll welcome any question, there really isn’t such a thing as a stupid one – and if we can’t help you, we might be able to recommend someone who can.