Whatever happened to checking data before a press release gets written?

So yesterday the Centre for Cities released its monthly High Street Tracker. It got what the Centre presumably wanted--a load of press coverage and little analysis. A few people made sure I got a link to it, because they're aware of my general disdain for the Centre and for other London-based thinktanks that write about regional cities without ever, seemingly, having been there.

I'm not going to do a full fisking on the data. This is more a short post to say 'what the actual ****? Does nobody read the press releases they put out and wonder if it all makes sense?'

This is the important page--you can search for your own city. It uses data from Beauclair for spend info (which I have no issue with; we use them and have cross-checked with our footfall data and with individual shop retail spend) and Locomiser, who use their own proprietary software estimations based on mobile phones--a system that I can;t find explained anywhere.

Now, we have both multiple cameras from Springboard across the city centre AND sensors from Movement Strategies as well as really detailed sales data from Beauclair. We're pretty data-d up. So I *know* that the Centre's footfall data is just wrong. What I don't understand is why it didn't occur to anyone at the Centre that it is OBVIOUSLY wrong.

So what's the problem?

Well, the Centre for Cities' headline data covering July for Manchester suggests that only 17% of office workers are back in the city centre. Total footfall is at 47% of pre-covid levels. And spend is at 94% of pre-covid levels.

So they're saying that on less-than-half footfall we have only 6% less than 'normal' spend. That's without tourists, football fans, overseas students (huge spenders in Manchester), event-goers, etc etc. 94%. This is just illogical. Either there's a tiny number of people spending a huge amount (and that isn't really the case from the spend data) or one of the bits of data is wrong.

Likewise, if only 17% of office workers are back and we're at 94% of total spend, either office workers have really no input into the local economy OR the data is questionable. Otherwise, getting the office worker level up to 70 or 80% of 'previous' will presumably see us FLYING past previous sales levels. Which rather puts the lie to Paul Swinney, Research Director at the Centre, saying

[a] "reluctance to head back to the office in our largest and most economically important cities means that people in the so-called 'sandwich economy' that caters to city centre office workers are facing an uncertain future as we get ever closer to the end of the furlough scheme in September".

It's either one or the other--either office workers have a huge impact on the city centre economy as he says or they don't as the current iteration of his data demonstrates.

And when it gets into the detail it's even stranger. Like this--

Before lockdown, 6% of footfall was from the city centre. Now it's 5%. That's a reduction. Despite the latter total footfall being at 47% of the former. Somehow all that footfall from the tens of thousands of city centre residents has massively reduced. Clearly not likely. And pretty directly contradicted by the accurate Beauclair data which shows a near doubling in the proportion of spend coming from the city centre (yes, I know there isn't a direct correlation between footfall and spend, but it should make people pause)

Why do I think this happened? To be honest, I'm not sure. Partially the definition the Centre for Cities uses for 'Manchester city centre' might differ from ours'. Partially, I think it's a question of indexing. I'm sure the Centre and Locomiser haven't made this very simple mistake, but the baseline figure they're using (ie 'pre pandemic') appears to be from Jan-Feb 2020. Why do tourist resorts do well in the current survey? Because resorts have no visitors and low footfall in January and lots of visitors in July and August. And city centres are the opposite--fullish in January with students, football fans, gig goers, etc, and pretty empty over the summer. It's why we do direct comparisons with the same period in previous years (or in 2019 at the moment).

But mostly I think it's not stopping to think whether this makes logical sense. Or cross-checking the info you have before issuing a press release. And there's a lesson for us all in that.