What BA knows about us when we enter a lounge

[Marc]

Maybe that’s just me being a nerd, but I made some (sort of) interesting observations when reverse engineering the code of the BA lounge occupancy page.

It’s this one here http://ba-uk-occupancy.livereachmedia.com and it looks like it drives the displays in T5 to show lounge loads.

First of all, clearly this URL was not meant to be published publicly but thanks to a kind soul here on the forum it was shared.
Secondly, everything below are my observations and interpretation of the code and should only be seen as hypotheses.

When they installed the displays in T5, I first assumed they’d use the data from lounge check-ins and (as you don’t check-out of the lounge) just keep each entry active for an average stay time of an hour or so. But as the page is hosted on livereachmedia.com it looks like the system is layered onto their CCTV footage using AI to determine loads.

A look into the code, where quite a few bits of script are visible, reveal interesting things:
– There’s a total of 12 zones (not sure if these might consist of more than one camera each) used to gather data
– They track data for ALL BA lounges at T5 – i.e. including CCR and Arrivals
– They only show loads for First and the Club lounges on the page for the terminal displays
– Quite a few further information are gathered to analyse whose visiting (more on that later)
– In fact, I’d say the terminal displays showing the loads are just a nice by-product

Let’s start with the the names of all zones.
There are 6 primary zones:
– Arrivals Lounge
– Concorde Room
– First Lounge
– Club Lounge South
– Club Lounge North
– Club Lounge B Gates

And 6 additional zones that are linked to the primary zones above – it looks like these are set up for lounges with multiple entrances:
– £m Door Entry (linked to the Concorde Room zone. I also love that they literally call it that in their code (for reference: https://www.headforpoints.com/2015/04/18/why-british-airways-paid-1-million-to-build-a-door/)
– Atrium Reception (linked to the Concorde Room)
– Atrium Reception (linked to First lounge)
– First Wing Entry (linked to First lounge)
– Escalator Entrance (linked to Club Lounge South)
– Elevator Entrance (linked to Club Lounge South)

Let’s take a closer look at the code.
I’ll not go into detail about some of the technicalities but rather break down a few key pieces of information.

– The primary zones were set up on 2022-05-03, the additional ones 2022-06-13

– The absolute maximum capacities set for each lounge are:
Arrivals: 126
CCR: 188
First: 418
Club South: 627
Club North: 305
B Gates: 232

– Each lounge has a ‘comfort factor’ that’s set to:
80% for Arrivals (=101) and First (=334)
60% for CCR (=113)
85% for Club South (=533), North (=259), B Gates (=192)

– And finally, there are some imho pretty interesting attributes I found about what’s getting tracked:
“analytics_repeat_visitors”:true,
“audience_engagement”:true,
“audience_gender”:true,
“audience_age_group”:true,
“safe_distancing_measure”:true,
“safe_distancing_mask”:true,

So while they track basic demographics like
– gender
– age-group

they also track
– if you are a repeat visitors
– your ‘engagement’ – not sure about the factors for how they measure this
– distancing (assuming this related to social distancing)
– if people wear a mask – well, only maybe as ‘mask’ could be used in a different technical context here

That’s it. Of course, nothing is ground breaking here. And maybe this isn’t really interesting for anyone else anyway, but as a tech and data nerd, I enjoyed it 😀

[ChrisBCN]

Great work! Take a gold star.

[Froggee]

You’ll probably find that half the people on here are nerds but most don’t have anything like that level of tech skills. I found it fascinating thank you!

[Marc]

Thanks a lot! 🙂

I’ve also tried to see if it’s possible to show CCR and Arrivals data on the page, but unfortunately those datasets are not available.

You can get actual numbers for the last 10 minutes for First & Club lounges though.
This is from this morning when I went through the code:

Timestamp, Entries, Exits, Current occupancy
[“2022-08-16T08:07:00Z”,3,5,180],
[“2022-08-16T08:08:00Z”,0,3,177],
[“2022-08-16T08:09:00Z”,2,6,173],
[“2022-08-16T08:10:00Z”,1,2,172],
[“2022-08-16T08:11:00Z”,3,7,168],
[“2022-08-16T08:12:00Z”,5,4,169],
[“2022-08-16T08:13:00Z”,1,2,168],
[“2022-08-16T08:14:00Z”,5,5,168],
[“2022-08-16T08:15:00Z”,4,1,171],
[“2022-08-16T08:16:00Z”,2,4,169]

[jjoohhnn]

Good work! I wonder if they have these for any other airports..

[can]

Very well done!

I’m sure now they’ll hide the scripts 🙂

[WaynedP]

Intriguing, thanks.
Great thread title too. Am tempted to start the obvious next one titled “What BA IT knows about what we know” for documenting if anything suddenly gets changed on this “never intended for public consumption” website and coding.
That should interestingly reveal that someone in BA IT follows this forum and selectively decides which of the known open IT stable doors gets bolted shut, and which remain wide open.

[yonasl]

This is very cool. Thank you for sharing!

[zio]

@Marc in awe of your IT skills…I think BA need you! Thank you for sharing- very interesting.

[marshy11]

I have no idea what I just read but it was interesting reading!

[Lady London]

Hum. scary on personal data gathering. Hearing that ‘Twilight Zone’ jangly music

[JDB]

Lady London wrote:

Hum. scary on personal data gathering. Hearing that ‘Twilight Zone’ jangly music

If you genuinely have privacy concerns, you should be much more worried about people’s home CCTV/doorbells and their illegal data gathering. See Fairhurst v Woodard [2021] 10 WLUK 151

[Andrew J]

Bets on the number of days before BA now shut down that site?

[ChrisBCN]

Lady London wrote:

Hum. scary on personal data gathering. Hearing that ‘Twilight Zone’ jangly music

This is something that is so minor and so small in the world of personal data tracking, I feel you would be horrified if you knew what else is being collected all around you!

[kevind]

Wow. You are my idol! I wish I could do stuff like this. I’m a data tech geek myself, but Marc if you are Ronaldo, I’m Eric Prunier!