WPP Open Pro: When Self-Serve AI Can Replace an Agency

In October 2025, WPP launched WPP Open Pro, a self-serve version of its AI marketing platform that lets a brand plan, create and publish campaigns without hiring an agency. Within a week, Google released Pomelli, a free tool that generates on-brand social campaigns for small businesses, and Canva launched Canva Grow for designing and running ads on Meta, as Digiday reported.

A year on, WPP has not published a user count or a revenue figure for Open Pro. Its February 2026 strategy update mentions the launch in a single line, in a report showing 2025 revenue less pass-through costs down 5.4% like-for-like. Looking back at Cindy Rose’s first year as CEO, Digiday described Open Pro as a self-service creative toolbox competing in a crowded market alongside Google and Canva.

None of that settles the question the launch raised. It sharpens it. Tools that let a small business run its marketing without an agency exist, several are free, and the companies selling the ad space are building them hardest. So the useful question is not whether WPP’s product wins. It is which parts of agency work you can hand to software, and which you cannot.

Open Pro’s real competitor is the ad platform itself

Open Pro covers four steps: AI-assisted campaign planning, analysis of audiences and competitors, content creation at scale, and publishing straight to ad platforms, with an option to hand execution to WPP’s own media arm. It is built on WPP Open, the AI platform WPP’s agencies use for large clients. According to The Current, pricing is usage-based with no minimum company size, and WPP aimed it at two groups: local teams of global brands that lack resources for full campaigns, and growing companies with small but experienced marketing teams.

Rose was candid about the commercial logic at launch. WPP does not largely serve small and mid-sized businesses today, she said, so it expects Open Pro revenue to be incremental. She declined to discuss pricing.

So Open Pro is less a threat to WPP’s agencies than a way into a market they never served, with a path upward to WPP’s paid media services. For most small businesses, though, the real alternative to an agency is not WPP. It is the AI already built into the places they spend money. Meta aims to let a business supply a product image and a budget and have AI create the ad, choose the audience and adjust it by location, with full automation targeted for the end of 2026, according to a Wall Street Journal report carried by Reuters. Google launched Pomelli as a free tool. Canva Grow sits inside a design tool many small businesses already use.

These tools are good at what they do, and they are getting better quickly. But each one is built by a company whose revenue rises when you spend more on its platform. A platform’s AI will optimize your campaign inside that platform. It will not tell you to move budget to a channel it does not own, or to spend less because the offer is weak. WPP’s tool has its own pull toward WPP’s services. None of that is sinister. It is the main thing to keep in mind when software gives you advice.

Where self-serve AI really can replace an agency

An honest answer starts with the cases where an agency is the wrong purchase. WPP itself positions Open Pro for campaign implementation and lower-funnel activation, which is a fair description of what self-serve tools do best.

If you sell a proven product to a customer you understand, on one or two channels, most of the remaining work is execution: producing variations, launching campaigns, keeping budgets and audiences tidy. Self-serve AI does that well and cheaply. On a small budget, an agency fee can take a share of total spend that would do more good as media.

It also works when someone in your business has good judgment and a few hours a week. The tools need a person who can brief them clearly, recognize weak output and read results without wishful thinking. With that person, the tools are a production team. Without that person, they are a fast way to spend money.

And it replaces agencies that sell only execution. If what you pay for is resizing, posting and button-pushing, software will do it for less, and it should. This is the one-to-many work that, as the argument in the commercial reality of positionless marketing runs, no longer needs a dedicated specialist waiting in a queue.

Where it cannot, and how to tell which case you are in

The tools fail where the problem sits upstream of execution. If the offer is weak, the price is wrong or nobody can say why a customer should choose you over the alternative, AI will produce competent ads for a weak proposition, quickly and in volume. That is the gap explored in why human copy wins back ground once everyone generates the same AI copy: when every competitor can generate the average, the average stops selling.

They also fail at grading their own work. A platform’s dashboard reports results the platform has an interest in, including conversions it claims credit for that would have happened anyway. Knowing whether your advertising actually caused sales takes someone who does not profit from the answer, which is the job of performance marketing that measures what ads actually caused.

And they struggle when channels depend on each other. A search campaign, an email sequence and a content program feed each other in ways a single-platform tool cannot see. Planning that connects them, so one idea carries across formats, is the work of content marketing built around one idea across channels.

To find out which case you are in, list what you pay your agency for, item by item, and mark each line as execution or judgment. Execution is producing, posting, resizing and launching. Judgment is deciding what to say, where to spend, what to stop and what the results mean.

If most of the list is execution, try the self-serve tools first and keep the fee. If most of it is judgment and you are not getting it, the problem is the agency, not the absence of AI. The fuller version of this decision, including when building in-house beats both, is in the honest case for hiring an agency and when it doesn’t hold.

Self-serve AI will take a large share of the work agencies used to bill for. What it leaves is the part that was always worth paying for.

Frequently Asked Questions

What is WPP Open Pro?

WPP Open Pro is a self-serve version of WPP Open, the AI marketing platform WPP’s agencies use for their clients. Launched in October 2025, it lets brands plan campaigns, analyze audiences and competitors, generate content and publish to ad platforms without hiring an agency, or hand execution to WPP’s media business. WPP aimed it at local teams of global brands and at growing companies with small marketing teams, and expects its revenue to be incremental.

How much does WPP Open Pro cost?

WPP has not published a price list. At launch, CEO Cindy Rose declined to discuss pricing. According to The Current, Open Pro uses usage-based pricing, so a business pays for what it uses, and there is no formal minimum company size. Compare that with free alternatives such as Google’s Pomelli and the AI tools built into Meta and Google’s own ad platforms before committing, since many small businesses will find those cover much of the same ground.

Can a small business run its marketing without an agency?

Yes, in many cases. A business with a proven offer, a clear customer and one or two main channels can use self-serve AI tools to produce ads, launch campaigns and manage budgets. What the tools do not provide is judgment: deciding what to say, spotting a weak offer, measuring whether ads truly caused sales and planning across channels. If someone in the business can supply that judgment, an agency may not be needed.

Are AI ad tools from Meta and Google better than WPP Open Pro for small businesses?

For a business that spends mostly on one platform, the platform’s own AI tools are often the simpler choice, since they sit where the money is spent and are included with the platform. Meta is working toward letting advertisers create and target ads from a product image and a budget. The trade-off is that each platform optimizes within its own walls. A tool like Open Pro can plan across platforms, but adds another layer to learn and pay for.