Home / Services / Websites & Brand / Multi-Site System

Multi-Site System

One design system. Every site on it. Your team can run it.

From R220,000 Scoped per project · delivered in 8–14 weeks from R22,000/moover 12 months, hosting included · R264,000 totalDelivery is still 8–14 weeks — this only spreads the payment, it doesn't stretch the project.

Show price as

Is this you?

Every change gets made four times, and one version is always wrong

You're the ops or marketing lead across a group. Three brands, or one brand in four regions, with sites built by different suppliers at different times. The logos are slightly different sizes. The footers disagree about the company's own address. One site still shows last year's range and nobody is sure who has the login to fix it.

Every change is made once per site, by hand, by different people. Updating a legal disclaimer means briefing three suppliers, chasing two of them, and discovering a month later that one site never got it. You don't have four websites, you have four versions of the truth, and you're the one who gets asked why they don't match.

You've priced rebuilding each site separately and the total was absurd. Worse, it wouldn't fix anything, because nothing would keep them consistent afterwards. Meanwhile your own team could handle most of the day-to-day publishing, if publishing didn't require a developer and a prayer.

What you
actually get

Every line below is a deliverable, not an intention. If it's on this list it's in the proposal, and if it's not on this list it isn't included.

One design-token library, shared by every property

Colour, type and spacing defined once and referenced by every site. Change the brand blue in one place and every property follows. Drift stops being possible rather than being policed.

Two or more sites, or 25-plus pages, on one component system

Each brand distinct where it should be, identical where it must be. New pages and new sites reuse the system instead of commissioning a new supplier.

Region and language variants without copy-paste forks

Variants built into the structure, with hreflang and regional metadata done properly, so a second market doesn't mean a second codebase quietly diverging.

A documented deploy pipeline

Changes go out the same way every time, reviewable and reversible. This is what makes it safe for your team to publish without breaking the design.

Migration and a redirect map, per property

Each property's content moved and its URLs redirected one to one, so no site pays for the upgrade with its rankings.

A WCAG accessibility pass

Checked and fixed across every property. At your scale, accessibility is a compliance question, not a nicety.

Training and governance SOPs, so your team runs it

Who may change what, how, and what needs review, written for humans and taught in person. The system has to survive staff changes, supplier changes, and me.

How it runs

So you know what's happening and when, and so nobody has to ask for a status update.

Week 0

Scoping

This package is scoped per project, so we start with real numbers: how many sites, which languages, what's migrating. You get a fixed figure in writing before anything begins.

Weeks 1–2

Estate audit

Every property crawled and inventoried. What earns traffic, what's dead, where the drift is, and who holds each login.

Weeks 3–4

The system

Design tokens and the component library designed and signed off, plus the governance model, who may change what.

Weeks 5–9

Build, site by site

The first site built fully on the system to prove it, then the rest follow. Each launches when ready, no big-bang weekend.

Weeks 9–11

Variants, migration, redirects

Language and region variants built, content moved, every old URL on every property redirected.

Weeks 11–12

Accessibility and pipeline

The WCAG pass, then the deploy pipeline documented and tested by your team, not just by me.

Weeks 12–14

Training and handover

Your team trained, SOPs handed over, governance agreed with the people who'll live with it.

+90 days

Support

Fixes and tuning included, plus answers to the questions that only surface once real people are publishing.

What I need from you

The projects that run late almost always run late for one of these reasons, so they're worth reading properly before we start.

One internal owner with authority across all the brands. If every brand head can veto the system, there is no system
A full inventory of domains, hosting, analytics and suppliers, including the awkward ones nobody has logins for. Chasing those in week one is fine, discovering them in week nine is not
A decision on which site goes first. Usually the most important one, because it proves the system
Your team's actual time in weeks 12 to 14. Training only works on people who turn up
Willingness to retire pages and properties that no longer earn their keep. Migrating rubbish consistently just gives you consistent rubbish

What's not included

On the public page rather than buried in the proposal, so there's no argument in week five.

Bespoke web application development (see Business Systems)
Ongoing content production
Translation and localisation copy

Your other options

What happens if you don't do this

Do nothing. The drift compounds. Each site keeps its own supplier, its own invoice and its own version of the brand, and you stay the unpaid integration layer between them. The cost is real, just spread thin enough to be invisible on any one invoice.

Rebuild each site separately, as each becomes unbearable. Over a few years that costs more than the system would have, and buys nothing structural: with no shared foundation underneath, the drift starts again the day each rebuild launches.

Go smaller. If you're one brand with one site, this is the wrong package and I'll say so on the call. Rebuild & Replatform at R85,000 – R145,000 gives you the same component-and-token foundation on a single property, and if a fleet arrives later, that foundation carries across.

Straight
answers

You're a small studio. Can you actually support this at enterprise scale?

Fair question, so here's the direct answer. On delivery: the work is phased, you own every phase as it completes, and I don't run a second project of this size in parallel. On support: the system itself is the answer, not my diary. Your team does the day-to-day publishing, the pipeline and SOPs are deliverables rather than promises, and the stack is standard front-end code any competent developer can maintain. You get 90 days of support, then Care if you want it. The whole build is designed to make me removable.

Will each site keep its rankings through the move?

Each property gets its own one-to-one redirect map and its own before-and-after crawl. Because sites launch one at a time rather than in a single weekend, any wobble is isolated to one property and investigated while the others carry on untouched. That staging is deliberate, it's the lowest-risk way to move a fleet.

Why not WordPress multisite? Our team could edit everything themselves.

Multisite couples your properties at the worst layer: one compromised plugin, one botched update, and every brand is down together. What your team actually needs is controlled publishing, and the deploy pipeline gives them that without the plugin surface or the licence stack. If specific people genuinely need daily self-serve editing, say so at scoping and I'll build an editable layer for exactly that content, rather than making the whole estate soft to protect one news page.

What moves the price above R220,000?

Number of sites, number of languages, how messy the migrations are, and how many people need training. The estate audit turns 'from R220,000' into a fixed figure, agreed in writing before the build starts. After that it behaves like every price on this site: the number you agree is the number you pay.

Who owns all of it afterwards?

You do. Every domain, account, source file and the token library itself, in your organisation's name from day one, with documentation to match. I've rescued too many businesses from suppliers who kept the keys. On exit everything comes with you, and any developer can pick it up.

Can our internal team really run it without you?

Yes, that's the point of buying a system instead of a website. Content changes, new pages from existing components and deploys are all your team's after training. What still needs a developer is genuinely new components or integrations, and for that you can use me, on Care or per project, or anyone else, because nothing in the build is proprietary to me.

Want one supplier keeping the whole estate running afterwards?

The Whole Backbone, From R28,000

Your outsourced digital department. On call, on retainer.

See The Whole Backbone →

Sound like
your problem?

One call, no deck. If this isn't the right package I'll tell you which is.

Email instead