Digital signage
managed from your phone

The reason is simple: the change happens between two customers, standing up, with one hand. Anything that needs a moment of concentration gets postponed, and postponed twice means it is not happening at all.

• Count them once, honestly. Open the app, find the screen, find the item, edit, confirm. Five is workable. Nine means the person at the counter will stop after the second week.
• Structure decides the count. A layout with the price as its own editable field takes two taps; the same price baked into an uploaded image takes a computer and a design tool. That choice is made at setup and it is the one that matters.
• One playlist beats four. Every additional place a price could live is another place to look for it.

Which turns into a setup rule worth following even when it feels over-cautious: build the first version so that the things you will change weekly are the easiest things to reach, and accept that everything else can be awkward.

The honest test is not a demo: hand your phone to whoever works the counter and ask them to change one price. Whatever they hesitate over is what will not be updated in March.

Three habits make that path work, and all three are about the shot, not the software:

• Shoot landscape. The screen is horizontal and the phone defaults to vertical. Turning the phone sideways before pressing the button saves every crop decision afterwards.
• Leave room at the edges. Whatever fills your viewfinder edge to edge will lose its edges. Framing slightly wide costs nothing and survives any screen shape.
• Shoot it where it is. A tray on the counter under the shop's own light reads as real. The same tray moved to a windowsill for a better photo reads as a stock image, which is precisely the thing people ignore.

There is one technical detail worth knowing because it is the usual cause of a slow screen: a modern phone photo is several megabytes and far larger than any screen can display. That resolution buys nothing on the wall and costs upload time on a shop connection. Anything that resizes before sending is doing real work, even though it is invisible.

The larger effect shows up after a few weeks. When the camera is the pipeline, the content stops being a project and becomes a record of what actually happened this week — today's tray, the new supplier, the thing that sold out. That is harder to plan and much harder to ignore than anything designed in advance.

Take the photo at the moment the thing looks best, not at the moment you sit down to update the screen. Those two moments are hours apart, and only the first one is worth photographing.

The most common culprit is the one that looks most convenient at installation: the guest wifi. Guest networks in bars, gyms and waiting rooms usually sit behind a captive portal — the page that asks you to tick a box or accept terms before you get through. A person taps it without noticing. A screen on the wall cannot tick anything, so the first time that portal expires, typically after twenty-four hours or after a power cut, the device is quietly off the internet. It keeps showing content, which is why the problem goes unnoticed for days: nothing looks broken, the updates just stop arriving.

The fix costs nothing at setup and is expensive to retrofit: put the screen on the same network as the till, not the one you hand to customers. Better still, run a cable. Nobody enjoys hearing that in a room that was decorated last month, but a wired screen never loses its connection, never asks anyone for a password, and comes back on its own after a power cut. Where the sockets and the network point are decides more about how a screen behaves over two years than which model you buy.

The behaviour when the link does drop is worth knowing, because it looks alarming and is actually correct: the screen keeps showing the last thing it successfully received. Yesterday's menu stays up. That is deliberate — the alternative is an error message, or a black rectangle, in front of a queue. A display that needs a live connection in order to show anything at all is a display that will eventually be dark at the worst possible moment, and no venue should accept that trade.

All of which turns into one small habit that is worth more than any amount of reading: make a trivial change and watch it land. Once, on the day of installation, from where a customer stands. Change a word, count how long it takes to appear, and now you know what normal looks like in your room. The whole value of that five minutes is that it gives you a baseline — so the day something genuinely does not update, you can tell the difference between slow and stuck without guessing, and you already know which of the two things to check first.

And if the venue has no usable network at all, that is a real answer too, not a dead end. A screen can run from a local schedule and pick up changes whenever it does get a connection — someone's hotspot for a minute, an hour of guest wifi in the morning. It is not ideal and it is honest about what it is: the content that is already on the device shows regardless, and the updates catch up when they can.

To be straight about it: we do not manage your network, and we cannot see it from here. What we can promise is the part that is ours — content already delivered keeps playing without a connection, and a change made on a phone is queued until the screen can take it, rather than lost because it was made at the wrong moment.