A chain rollout looks like a discount job: one design, twenty copies. It never is. Twenty sites mean twenty surveys, and no two shopfronts on a chain come back the same length. What has to stay identical is the brand: the same fascia system, the same colours, the same letters, the same look after dark. The rollout part of my AxisMod fascia demo exists to show that gap being closed: one set of brand rules on one side, twenty different buildings on the other.
Where a rollout leaks time
The usual way of running a chain is one project file per site. The first one is designed properly. The second is the first one saved under a new name and stretched to the new length. By the tenth, nobody can say with confidence which file holds the current version of the brand. I have watched a multi-site rollout of a single brand come apart exactly like that, and it never happens in one dramatic step.
Three things go wrong, and they tend to go wrong in that order.
- Letters get scaled by eye. A shorter frontage gets slightly smaller letters because they looked too tight. Now site nine reads differently from site three, and it shows the moment two photographs land side by side in the client's report.
- The bill of materials drifts away from the drawing. Quantities are retyped into a spreadsheet per site. One correction on an elevation and the sheet is stale, but nobody goes back to re-check site fourteen.
- The chain total is a guess. The client wants a price for the rollout, not for a site. Adding up twenty spreadsheets written on twenty different days gives a number nobody wants to defend in a meeting.
None of this is incompetence. It is what happens when the only thing connecting twenty jobs is a folder structure.
One rule set, one row per site
In the demo module a chain is a table. A row is a site: its name, the dimensions that came back from survey, whether it has returns, the letter size chosen for it. The brand rules sit above the table and belong to the chain rather than to any row: the fascia system, the field colour and the accent band, the letter type, whether the light comes from the letters or from the fascia itself.
Change the letter type and it changes for every site at once. Change one surveyed length and only that site is redrawn. That is the whole idea, and it is why a rollout stops being twenty projects that happen to share a logo.
Each row is still a complete configuration, not a summary line. Open a site and you get its own elevation and plan, its own section, its own 3D view in daylight and at dusk with the letters lit. The client can look at the awkward site, the narrow one on the corner, and approve it on its own merits before anything is ordered.
What a rollout module has to produce
This is what the demo generates today, and it is the shortlist I would hold any rollout module to. Per site, six production documents: elevation with the plan, the panel and frame schedule, the letter specification, the electrical schedule, the bill of materials, and the rollout summary. Cutting files for the letter faces, returns and backs, with the illumination and fixing layout already on them. Then, for the chain: a merged bill of materials and a chain estimate built from the same geometry as every drawing in the set.
The practical effect is on the margin. The twenty-first site is a row in a table and a survey, not a new project with a new set of hours attached to it. And once the quantities come out of the drawings rather than out of a spreadsheet, the rule I would apply is to order once for the whole chain, so the material that arrives matches the panels that were actually drawn.
The part that was hard
Merging is easy. Holding the rules is not.
A merged bill of materials is trivial to produce if you are willing to let it be wrong. The moment those totals come from anywhere other than the geometry that produced the drawings, the chain estimate becomes a second source of truth, and two sources of truth always drift. Everything in the demo points back at one configuration for that reason, and the totals are a view of it, never a parallel document.
The other hard part is that a shared rule has to survive sites that disagree with it. The rule that keeps a panel joint out of a letter has to hold on the shortest frontage in the chain and on the longest. The wordmark has to sit inside the clear field of the fascia and keep its proportions everywhere, which means it is scaled by the rule and never by hand. Every time I relaxed one of those to make an awkward site behave, the chain stopped looking like a chain.
So the check I run across a whole chain is blunt: every panel has to fit the sheet it is cut from, every wordmark has to sit inside its clear field, and the chain totals have to equal the sum of the site totals. If any of the three fails, it is the configuration that is wrong, not the drawing.
What this changes for a rollout
For a client with a network of shopfronts, a module built this way comes down to three things. The brand is decided once and cannot quietly change between site three and site nine. The price for the whole rollout is available in the same session as the design, not a week later. And the sites that always cause trouble, the short one, the one with a return, the one measured wrong the first time, are corrected as one row without disturbing the other nineteen.
The demo runs on NORVIK, a fictional forecourt chain of three stations, which is enough to show the mechanics. The real value only appears at twenty, where the cost of doing it by hand is the entire point.
I should be straight about what AxisMod is at this stage. It is a demonstration of what I can build, not a rollout product waiting on a shelf for a licence fee. A chain has its own fascia systems, its own brand rules and its own document set, so a module for a real chain is written directly for that chain and for nobody else. The demo exists so you can watch the mechanics work on a made-up brand first, and then decide whether it is worth having one built around your own.
More about the platform: AxisMod.
