WordPress VIP multisite: how 170 titles run as one system

🇹🇷 Türkçe: Bu yazının Türkçesini oku →

Every large multi-brand publisher I have worked with has asked the same question sooner or later, and rarely in polite language: how do we keep the brands distinct without paying to run a separate stack behind each one? The answer is usually a compromise everyone tolerates. One masthead lands on the “reference” platform, the others live with whatever the last agency shipped, and the migration budget is always in the next fiscal year.

News Corp Australia stopped tolerating it a decade ago. They put 170 newspaper and magazine titles and forty-plus digital brands — from The Australian to Vogue Australia to The Daily Telegraph — onto a single WordPress VIP multisite. That project is the one I keep referencing when a multi-market operator asks me how this is meant to work at their scale.

Two things make it worth writing about now, ten years in. The numbers are still improving inside the same architecture, which is the answer to anyone who thinks a unified platform inevitably ossifies. And this is the model the industry quietly copied, but most of the copies stopped short of the two or three decisions that make the original work. I want to say what those decisions were, in language a CMO or a CTO can take into a board conversation without a technology footnote.

Australia’s largest media company sits on one platform

News Corp Australia reaches seven million Australians a month and publishes 170 newspaper and magazine titles, accounting for roughly three-quarters of the country’s daily metropolitan newspaper circulation. That is not a portfolio a content management system is supposed to unify. Each masthead has its own audience, its own visual identity, its own newsroom and its own editorial calendar.

Before 2015 the digital operation was what you would expect of a portfolio grown by acquisition: a mixed platform landscape assembled over the previous decade, no shared publishing rhythm, no shared component library, and a build-out cycle that could take days for a new section. Every routine editorial change had to route through IT.

In late 2014, News Corp Australia commissioned a year-long migration with WordPress VIP and its agency partner XWP. The first production site, PerthNow.com.au, went live on 7 October 2015. The full migration of the fifteen major web properties finished the following year. Ten years on, the estate has grown to 40+ sites, all on the same platform.

The before-state was expensive in a way the invoice did not show

Two costs come out of the fragmented pre-migration arrangement, and both land on the CFO’s desk. The direct one is easy to see on a purchase order: multiple licences, multiple hosting contracts, multiple integration teams. The second cost is what I would call the pace tax, and it never shows up as its own line item. Every simple change a competitor made in a week took a month, because the change had to be reproduced across systems that did not agree on what a section, a template or a byline was.

The News Corp Australia leadership decided the pace tax was the bigger problem. That is the framing most publishers get wrong. The visible line item is the hosting bill, but the invisible one — the deals not done because the site could not carry a campaign fast enough — is usually larger by an order of magnitude. Weston Ruter, CTO at XWP at the time and the migration’s project architect, put the goals plainly:

“Some of our goals with the migration were to improve workflow efficiencies, empower site producers to make changes with a full preview of their impact, and cut the time it takes to build out a section of the site from days to hours.”

Weston Ruter, CTO at XWP

Notice what is not in that sentence. Not a mention of the technology stack. Not a benchmark number. The whole framing is operational: what a site producer can do without asking an engineer for help. That is the frame a buyer should hold onto when a vendor arrives with a slide deck. The platform is the enabler. The change under negotiation is the operating model.

What they built on top of the platform is the interesting part

The obvious answer — “they moved to WordPress VIP” — is not what makes this case study worth studying. The interesting part is what News Corp Australia and XWP constructed on top of it: a shared foundation they called the Site Production Platform, one system across every brand rather than a separate WordPress installation per masthead. A vetted core plugin set and a shared base theme sit under every brand’s site. Distinct visual identity, section structure and audience configuration lives on top, per brand, edited by producers through a “Site Build” surface rather than through code.

The design decision underneath that structure is the one every multi-brand publisher I have talked to has to get right, or the model falls apart within eighteen months. Give each brand its own bespoke platform and the shared efficiency evaporates as customisations diverge, one small exception at a time. Force every brand into a rigid template, and the brand teams route around IT and rebuild their differentiation with third-party embed codes and marketing widgets that break analytics and page performance. News Corp Australia’s answer is a fixed shared foundation with a curated set of levers the producers actually control. That balance is the real engineering work in a project like this, and it is far more valuable than the technology choice itself.

Everything else follows from that decision. Producers build a new section in a working session rather than waiting for a release. Editorial teams preview a change in the context of the real brand before it goes live. The engineering team stops being the queue for every small update and gets to work on the platform itself, which is where the compound gains eventually come from.

The measured result, ten years in

The published numbers stand up:

• Seven million Australians read News Corp Australia’s digital brands each month, across 40+ sites on the same platform.
• 40x improvement in time to publish — the metric that hurt the most before the project is now the metric the operation leads with.
• 50% reduction in build time on new site sections, reported by XWP after the platform matured.
• 50+ custom plugins integrated on the shared foundation, all vetted through the VIP code review gate, so a new plugin does not become tomorrow’s outage.

Two of those numbers deserve a second reading. The 40x figure is not “we shipped forty times more code.” It is “the round trip between a producer wanting to change something and that change being live is forty times shorter.” That is the number that shows up in the CFO’s revenue conversation as a deal that ships in a week rather than a quarter, and it is the number a buyer should ask about first when a vendor presents a case study.

The plugin count is the other one. Fifty custom plugins on a shared foundation is not a warning sign in this context — it is proof of governance. Fifty custom plugins on an unmanaged WordPress install is what a support engineer opens a ticket on. The difference is the code review gate every VIP customer runs through, which means the fifty are audited, versioned and known to the platform team. That is what the operating budget on this pattern actually buys.

The unresolved part is not the platform, it is the next surface

Every good case study has an unresolved part, and this one has one worth naming. The platform decision the News Corp Australia leadership made in 2014 is not the one their successors have to make again — the numbers still improve, and replacing that foundation is not on anyone’s roadmap I have heard of. But the audience has shifted underneath it. Some readers now discover News Corp Australia’s brands through AI-mediated search surfaces rather than through a masthead homepage, and the machine-readable inventory that serves those readers is a different engineering conversation than the one that gets a producer’s section live in a working session.

So the choice a multi-brand publisher faces today is not “unified platform, yes or no” — News Corp Australia settled that argument for the industry. It is “on top of a unified platform, what is the next surface you build for, when the next surface is not another human browser.” That is where the next few years of investment on this pattern will go, and it is the question I would put to a leadership team commissioning a similar consolidation right now, before the platform contract is signed.

What we would build, and what it costs

This is where a strategy blog stops and the invoice starts. What does an operation like this take to stand up, and what does a mid-market operator — not one with 170 titles, one with five or ten — copy from it?

Five items we sequence on a real WordPress VIP multi-brand build, in the order they earn their place on the plan:

1. A shared theme foundation with per-brand configuration on top. Not per-brand themes. One core theme with a documented extension surface — colour, typography, layout regions, header and footer configuration — that a brand team can edit through the admin without touching template code. Budget: six to eight weeks for a design systems team, plus two weeks of front-end scaffolding.
2. A curated shared plugin set, code-reviewed on ingestion. Vetted before it enters the foundation, versioned once the whole estate is on the same release, retired centrally when it breaks. The rule that keeps fifty plugins workable at News Corp Australia is the rule that comes off the wall in most agencies: no plugin enters the estate without a review and an owner. Budget: half a senior engineer’s time on an ongoing basis, plus the review cost per plugin.
3. A producer surface that maps to the real editorial roles. Every newsroom has its own hierarchy — section leads, editors, producers, freelancers — and the admin should match. Anything else pushes work back onto engineering. Budget: three to four weeks per brand for the initial mapping, then a rolling investment as roles change.
4. Preview parity: what the producer sees is what publishes. The News Corp Australia framing on this — “full preview of their impact” — is not decorative. Preview parity is what removes the review-cycle bottleneck. Budget it as a first-class engineering concern, not a nice-to-have, because the first missing preview kills producer trust and every subsequent change routes back through engineering.
5. An audit trail the compliance team can read. Multi-brand publishing at scale means multiple regulatory environments, sometimes multiple markets. The VIP audit surface is what keeps the “who changed what and when” conversation short. Budget: a week of governance workshop, then a quarterly review.

The build cost for a mid-market operator running five to ten sites on this pattern is a two-quarter engagement with two senior engineers, a design systems lead and a producer-experience specialist. That is not the small number. It is smaller, in almost every case we have run through the arithmetic on, than what the current fragmented estate is quietly costing to keep on life support. That is the reason the case still lands ten years after News Corp Australia proved it out.

If you are evaluating this pattern for a multi-market or multi-brand operation, three questions are worth taking into the first conversation with any partner, ours included. What is the honest inventory of your current digital estate, brand by brand? Which of your producers is willing to be the reference site for a shared foundation? And what would you stop asking engineering to do first, on the day the new platform is stable? The answers shape a plan that costs a fraction of a “big bang” migration and lands in half the time. Start with our WordPress VIP pillar, the multi-country and multi-language management page, or come straight to the WordPress VIP agency page if you want that conversation now.