Solutions by typeCustomer experience
Did the campaign actually do anything, or does it just look like it?
You paid for the campaign, and nobody knows if it worked
Monday, half past nine. The meeting to review Saturday’s event: there was a stage, a screen, a balloon arch and two presenters in the atrium. Someone says the place was packed. Someone else says it was a Saturday like any other. The only figure on the table is the till, and a Saturday till depends on the weather. The atrium cameras recorded all six hours and nobody has opened them.
3 cameras across 2 sectors
This is how you ask for it
«Compare how many people crossed the atrium on the Saturday of the event with the Saturday before, at the same hours.»
With no number, the loudest voice wins the argument. The campaign that worked gets repeated by luck, and so does the one that did not, because nobody can show the difference. Next year’s budget is split exactly like last year’s, not because that is best but because there is nothing to argue with. And the brands that paid to be in that atrium ask for a report that ends up written in adjectives.
You decide once
You decide this once and never set it up again. You draw on the image where the counting happens — the mouth of the atrium, the stretch of pavement in front of the shop window, the bike lane — and you say how often you want the number: hourly, daily, per Saturday. IRIS counts with the same line and the same criteria every time, on the day of the event and on the Saturday before. That is why the two numbers can be subtracted: they were not counted by a different person in a different mood.
A campaign is paid for before and judged afterwards. The stage, the screen, the crew and the permits are paid in January; the question of whether it was worth it arrives in March, when nobody is left who remembers what the atrium looked like at five in the afternoon. And this is not only a shopping centre problem. A campaign to get people wearing helmets in the bike lane has exactly the same one: you know what the posters cost, you do not know how many people rode past without a helmet before you put them up.
It is not that information is missing. It is that it was recorded and nobody counted it. The cameras in the atrium, at the shop window and on the boulevard have watched everyone go past for months. Counting that by hand is impossible, so somebody counted one loose afternoon with a clipboard, and that loose afternoon became the figure of the year.
IRIS counts from the same spot every single day. You draw a line on the image and from then on there is a number every hour: how many people crossed the atrium, how many walked past the window, how many stopped to look at it. When the Saturday of the event comes there is nothing to set up, because the Saturday before is already counted, and so is the one from last year. The campaign stops being judged on impressions and starts being judged by subtracting two numbers measured the same way.
And there is one more number, the one that really settles a shop window: how many people walk past and how many stop. Eighteen hundred on the pavement with two hundred stopped at the glass is not the same shop as eighteen hundred with fifty. The window gets changed every three weeks, and until now it got changed on taste, because nobody knew which one had worked. Now you compare the week of the new window with the week before, at the same hours, and the number says whether it stopped anyone or was merely pretty.
What changes
Before, the event report was written from memory, with the till figure beside it and a photo of a full atrium. Now you open the comparison between that Saturday and the one before, hour by hour, and you see where people moved and when they stopped moving. The conversation goes from «I think it went well» to «at six the atrium flow doubled and the upper floor felt nothing».
Where we have seen it
CAM-41Boulevard: cars, bikes and people
«Also tell me how many people ride past without a helmet, so I know whether we need a road-safety campaign.»
CAM-537Before, during and after: what the campaign actually brought in
«Compare the people who crossed the atrium on the event Saturday with the Saturday before, at the same hour.»
CAM-538Eighteen hundred walk past and two hundred stop
«Tell me every day how many people walked past the window and how many stopped to look at it.»
What it does not do
IRIS counts people. It does not know who they are, it does not recognise faces and it cannot tell one shopper from another. Nor does it know whether anyone bought anything: your till says that, not the camera.
And one number is only worth something if the other was counted the same way. Move the camera, change the framing or draw the line somewhere else and the comparison breaks. Comparing Saturdays means nobody touches the line.
And it does not count what does not cross the line. If somebody comes in from the car park and goes straight upstairs, they never crossed the atrium and for that number they do not exist. That is why the lines go where the flow is actually decided, and why an atrium figure cannot be presented as the visitor figure for the whole centre.
Questions
- How do I know the rise came from the campaign and not from the weather?
- Not from the camera alone, and it is worth saying so. The camera gives you the number, not the cause. What you can do is compare the same weekday, at the same hours, with the same line, and look at several weeks in a row rather than one. If atrium footfall rises on the Saturday of the event while car park entries do not, that already tells you something. IRIS supplies the clean figure; you supply the explanation.
- Can I compare with a Saturday from a year ago?
- Yes, as long as that Saturday was already counted with the same line. Counting is not switched on for the event: it runs every day, so by the time you want to compare, the history is already there. That is why counting is the first thing to go live on a retail street, even with no campaign in sight. The value does not show up on day one: it shows up the day you have something to compare with.
- Can it count bikes and scooters, not just people?
- Yes. The same line can count cars, buses, bikes, scooters and pedestrians separately, each with its own number and its own time slot. That is what lets you judge a safety campaign on a bike lane: how many people rode past before and how many ride past now, and at what hours. A lane that fills at eight in the morning and empties by eleven is not fixed the same way as one that only fills on Sundays.
- Do I need new cameras to count?
- Not necessarily. If the camera already sees the whole passage and the image is stable, it can count. What matters is not the brand or its age but the framing: a camera set up to watch a door at night may be useless for counting a walkway by day. Sometimes one has to be turned or raised a little, and from then on it counts every day without anyone touching it again.
Demo · IRIS at work
We won't just tell you. We'll show you.
It is a summary of the real screen: it plays by itself, step by step, and shows only the path of one case. The actual tool has far more than fits in here.
- The alert fires
- It lands in the tray
- You open it
- You confirm or dismiss
Not seeing yours here?
These words are the ones we have built, not the ones IRIS understands. The list falls short on purpose: you ask for it in a sentence, and sentences do not run out. Tell us what is in front of your cameras and we will tell you whether it understands it, or not yet.
And the best part
You can add more cameras;you cannot add more hours to anyone’s day.
Cameras can be multiplied; a person’s hours cannot. IRIS looks at everything coming in and leaves them the one thing no machine does better: deciding what to do.
Tell us your problem and we will say whether IRIS understands it, or not yet.
We reply the same working day.