I think it's a matter of common sense that we shouldn't ask for more
money unless we can credibly demonstrate with stuff like success
metrics and improving trends that we can spend the money we've already
been given effectively.
Risker's comments made me wonder, however, about the more specific
issue of how the WMF is measuring the cost/benefit of banner displays.
The benefit should be fairly easy to figure out as denominated in
dollars- roughly speaking, it would probably look something like the
total amount raised over a defined period divided by the number of
banner displays during that period. But what about those much more
subtle and potentially lagging costs? I assume that the WMF is
measuring stuff like session lengths and return rates. Is the WMF
tracking on anything else for non-logged in users?
In any case, what I would most like to see is a comparison of graphs
of such metrics over the course of a full campaign. It seems like we
all agree that the banners are annoying, but is there really a
measurable "banner fatigue" phenomenon among our readers? For example,
can we point out a distinct point of diminishing returns, beyond which
the slopes of one or both graphs significantly steepens? If anyone has
this data for the current or past campaigns, please forward it to me.
I'll try some different visualizations that get past the dollars signs
to the true cost of prolonged panhandling.
Alternatively, we could pivot to a street performance model by getting
the article on Thomas Jefferson to juggle fire batons and spray
painting the article on Popping silver. After Jimmy finishes his
extended plastic-bucket drum solo and we've warmed them up with a few
mediocre jokes, we could pass around the banner for donations. It
would probably only work on the tourists, tho.
This is $10.6
million more than the $20 million fundraising goal indicated
in the blog post. (At any rate, that's the sum I get; I'd welcome anyone
double-checking my math.)
There is no scenario I can come up with where this is actually a good
result. Sure, an extra $10.6 million might be nice in the bank, but it
massively exceeds budget. The fundraiser met its goal, with plenty to
spare, on December 17. And yet we put our readers and our users through
another two weeks of fundraising. Given that we were already really
pushing the goodwill of the broad Wikimedia community (that includes the
users of our products)....well, as I say, this is not a good result.
People were putting Wikipedia on Adblock because of those banners, and they
were doing it long after the goal had been reached.
I'd say I was speechless, but actually I am working extremely hard to hold
my tongue here, awaiting an explanation for this. And yes, I think the
Wikimedia community deserves to know why this happened.