Credit: Nick Grinberg
Next&Co co-founder Nick Grinberg wants clients to pay for outcomes, not staff numbers, as AI takes over the manual work agencies have long charged for.
Grinberg, co-founder and head of strategy at the independent performance agency, made the case in an interview with AdNews about Next&Co's plans to launch an AI-assisted delivery model alongside its traditional service.
The agency built its agents on top of Prometheus, an internal platform that audits where client ad spend is wasted.
"The headcount model... is the laziest way of assessing an agency's capability and scale,” Grinberg told AdNews.
Grinberg said the industry has relied on staff numbers as a proxy for value for too long and that capability should be judged on results instead.
The shift, he said, is driven by Next&Co's own move toward AI-assisted delivery.
"What outcome can the agency deliver for you with what level of certainty and at what cost?" he said.
"Can we just agree that headcount isn't a substitute for capability anymore?"
Grinberg believes the metric that matters isn't awards or creative polish.
"I don't give a toss if my media campaign wins awards, or if my creative is pretty," he said.
Instead, what counts for Grinberg is whether the work drives sales, clicks, leads and downloads for clients.
Media optimisation, reporting and search engine optimisation are increasingly handled by AI agents, Grinberg said, freeing staff for strategy and client relationships.
"Instead of you crawling through five different ad accounts, it fetches all the data and gives you a quick synopsis of what it means against that client's commercial context, so you can just tick it off,” he said.
Humans remain responsible for the result.
"AI doesn't generate the business outcome. It facilitates one, but the human steers it," Grinberg said.
Daily account management, such as checking budget pacing and conversions, still needs to happen.
"Agents help us do that much faster, which frees up the account lead who was previously buried in the account to instead review and approve the optimisations," Grinberg said.
Not every agency AI tool is built the same way, according to Grinberg.
He said broad, ungoverned access to general AI models carries risk, warning that unsupervised open-source tools "doesn't have the guardrails to stop it hallucinating."
Next&Co's agents are trained on the agency's own workflows and years of media-optimisation experience rather than a general-purpose model.
"Our knowledge base and training material are different," Grinberg said.
"They've got our workflow and process baked in, taught by humans with years of media optimisation IP.”
Whether that remains a point of difference is unclear. Most agencies serious about AI are likely to build a similar proprietary layer over the same handful of foundation models.
Grinberg said it was also agency structure, not just pricing, that's under pressure. Next&Co has always run smaller, senior-heavy teams rather than a junior-heavy one, he said.
"Those junior-heavy models are going to get wiped," Grinberg said.
He didn’t say where junior staff would gain the experience agencies have traditionally taught them on the job.
Client appetite for agency AI use also varies, Grinberg said. Some clients have shut it out entirely over data and security concerns, restricting which tools their agencies can use.
"You can't use Claude. You can't use X," Grinberg said.
Next&Co is trialling an AI-assisted delivery model with clients who have signed off on it, Grinberg said, and plans to offer it alongside its traditional, human-only model, letting clients choose between them.
"We're going to present it as a choice," Grinberg said.
He believes clients should not penalise agencies for working faster.
"I genuinely hope the evaluation becomes more about outcomes. If I can get you a better ROI in less time, should I be punished for being efficient?" he said.
Next&Co builds performance incentives into contracts where clients allow it.
Grinberg said it is too early to judge how agencies are handling the shift.
"I don't think it's fair to judge what anyone is or isn't doing yet, because it's still in flux," he said.
Even after 15 years running an agency, the pace of change is hard to predict for Grinberg.
"I've been doing this for 15 years, and honestly, I still sometimes feel like I've got no idea what's around the corner," he said.
Overall, Grinberg rejected the idea that agencies become redundant as AI use grows.
"I firmly reject the notion that agencies, or the service layer of the agency, are going to become redundant. None of this makes the ecosystem less complex, it makes it more complex," he said.
Have something to say on this? Share your views in the comments section below. Or if you have a news story or tip-off, drop us a line at adnews@yaffa.com.au
Sign up to the AdNews newsletter, like us on Facebook or follow us on Twitter for breaking stories and campaigns throughout the day.