SharePoint & Microsoft Resources

What Warner Bros. Discovery taught us about scaling low-code and AI without losing control

by Tim Wallis, Chairman of Silicon Reef

What Warner Bros. Discovery taught us about scaling low-code and AI without losing control
Early in 2025, Warner Bros. Discovery placed a big bet. In one move they bought 30,000 Microsoft Power Apps licences and set out to let tens of thousands of employees build their own tools. For a global business of 35,000 people behind HBO, CNN, Discovery and TNT Sports, that’s a serious statement about where work is heading. The interesting question was never whether the technology would work. It was whether they could scale it without losing control.

Anyone who has spent time around enterprise technology knows why that question matters. You give 30,000 people the ability to build, and within a year you can have apps in every corner of the business, no one sure who owns what, data sitting in places it shouldn’t, and a security team that’s lost sight of the whole estate. The technology usually works beautifully. Governance is the part that tends to fall over.

So the real question wasn’t whether WBD’s people could build. They obviously could. It was how to let thousands of them build and that the IT team could still sleep at night.

That’s where the team at Silicon Reef came in.

Working alongside Microsoft, Silicon Reef built WBD a Power Platform Centre of Excellence. I prefer to call it a Centre of Enablement, because the point isn’t to police people, it’s to let them move quickly inside sensible boundaries. We put the environment strategy, the data loss prevention policies and the security roles in place before any serious building started, with central monitoring across the whole estate. That means every new app now lands compliant with WBD’s security and DLP rules across the thousands in play. This isn’t aspirational to capture most of them, it’s all of them.

With that foundation set, the building could happen at proper speed.

More than 200 people across the business are now active makers, building and sharing apps with each other, now, increasingly with Copilot in Power Apps helping them go from a plain-English description to a working app. The team supported the build of nine production applications – which are now live – and they went from idea to in-use in weeks rather than months. Around 80% of the manual workflows we looked at have been automated out. And the savings on external development add up to roughly $1.5 million over four years.

None of these are toy projects. There’s a broadcast dashboard running in TNT Sports operations. There’s a hair and makeup booking app handling between 5,000 and 7,000 appointments a month across productions. And there’s the close-out process every production runs when a shoot wraps: a hundred-item checklist used across HBO and Warner Bros. studios that used to take around 200 minutes and now takes about 10. Nothing about the standard changed, and the same people are still responsible for it. It just stopped taking up most of an afternoon every time.

If I were pulling out one lesson for any executives reading, it would be this. None of this came from a dramatic AI announcement or from chasing the newest model. It came from getting the unglamorous things right first. Clean structure, clear governance, security built in at the start rather than bolted on at the end. Once that was true, every app built on top of it stood on solid ground. The organisations that get the most from AI over the next year or two, whether that’s Microsoft Copilot, agents built in Copilot Studio, or whatever comes next, will be the ones doing exactly this now, even though the temptation is to skip the ‘boring bit’.

It’s worth being clear about what this wasn’t. It wasn’t about replacing people with technology. If anything it was the reverse. Those 200 makers are WBD employees who can now solve their own problems instead of waiting in a queue for IT. The technology took on the repetitive lifting so that talented people could spend their time on the work that genuinely needs a human. That’s the approach we believe in at Silicon Reef, and it’s been good to see it hold up at this scale.

If you lead a media or entertainment business, there is a clear lesson here. Your industry runs on a thousand operational processes that have never been written down, living in spreadsheets, inboxes and people’s heads. Every one of them is a candidate for this kind of treatment. But only if you build the Centre of Excellence (or Enablement) first. Do it the other way round and you’ll spend next year firefighting instead of innovating.

For WBD, this is a starting point rather than a finish line. The next phase brings in richer analytics with Power BI and extends the same governed model to AI agents built in Copilot Studio. The foundation they invested in is what makes all of that straightforward rather than risky.

I’m proud of what the team built here, and grateful to Microsoft for backing it from the start. Kathleen Mitford, Microsoft’s CVP of Global Industry Marketing, was generous enough to publicly endorse the work, and that means a great deal coming from someone who sees partner delivery across the whole industry.

If you’re weighing up how to let people build with Copilot Studio, the Power Platform, or AI more broadly without losing control of your environment, this is the kind of problem we solve. We help organisations scale safely, with the right structure in place from the start.

Original post can be found on LinkedIn.
Read the article here.

 

Case Study

Warner Bros. Discovery

See how Warner Bros. Discovery translated their process improvement know-how by building it into the technology. Resulting in a scalable, governed Power Platform environment with best practices, security and integration baked in.

Warner Brothers Discovery

More from Silicon Reef

AI Readiness in Universities: Research vs Reality

AI Readiness in Universities: Research vs Reality

Key Takeaways The Times Higher Education AI and Digital Maturity Index (AI-DMI) frames university AI readiness across four pillars - Strategy, People, Utilisation and Technology - and finds that technology investment is consistently outpacing actual use, with people...

What AI Community Agents Mean for Viva Engage

What AI Community Agents Mean for Viva Engage

Key Takeaways AI community agents in Viva Engage are context-aware assistants that suggest verified answers, summarise threads, surface existing content to prevent duplication, and flag knowledge gaps - turning communities from reactive message boards into proactive...