A few years ago, a great media buyer was the person who knew every setting. Bid adjustments by device, hour and location. Tightly themed ad groups. Interest stacks layered just so. That skill set built careers, including mine.
In 2026 the algorithm does most of that. Google folded broad match and automatically created assets into AI Max for Search and now serves ads inside AI Mode. Meta's Andromeda system decides which creative reaches whom. TikTok's Smart+ automates targeting, bidding and creative selection. Microsoft runs Performance Max across Bing and Copilot. The machine buys the media now.
That does not make media buyers obsolete. It moves the job upstream. The leverage is no longer in the levers inside the platform. It is in what you feed the platform and how you judge what comes back.
Signals Are the New Targeting
When every advertiser has access to the same automation, the difference is the data you give it. Clean conversion tracking, server-side events, offline revenue uploads and value-based bidding tell the algorithm who your best customers actually are.
I think of it this way: targeting used to be telling the platform who to find. Now it is teaching the platform what a good customer looks like, and letting it find them. An account optimizing to qualified pipeline will beat an account optimizing to form fills, every time, with the same budget.
Creative Is the New Targeting, Too
On Meta and TikTok in particular, the creative itself decides who sees the ad. A video that speaks to first-time parents finds first-time parents. A testimonial from a business owner finds business owners. That means creative strategy is now audience strategy.
The practical shift: fewer campaigns and ad sets, more genuinely different concepts. Not ten versions of the same ad with different headlines, but different angles, formats, people and problems. Creative volume and diversity are the input that compounds.
Structure Is About Giving the Machine Room
Fragmented accounts starve algorithms of data. The 2026 default is consolidation: fewer, larger campaigns, broad targeting with clear guardrails (exclusions, brand controls, budget caps) and enough conversion volume per campaign to learn.
The buyer's skill is knowing which controls still matter: negative keywords and brand exclusions, customer acquisition goals, placement reviews, and when to keep a manual campaign as a control group.
Measurement Is the Strategy
Every platform now grades its own homework, and every platform claims credit for the same customer. The media buyer who can say what actually moved revenue is the most valuable person in the room.
That means blended metrics (MER, new-customer CAC, payback period), incrementality tests on the biggest channels, and mix modeling for budget allocation. Platform ROAS is an input, not the verdict.
The Answer Engines Are the New Battleground
Search behaviour is fragmenting into conversations. Google serves ads inside AI Overviews and AI Mode. OpenAI began testing ads in ChatGPT in early 2026 and has been expanding them, including to Canada. Organic AI visibility and paid placements inside answers will increasingly be planned together.
Brands that are clear, consistent and cited across the web will win both the organic answer and the paid one.
What This Means for Teams
The modern performance team looks less like a room of button-pushers and more like a small growth unit: a strategist who owns the economics, a measurement lead who owns the truth, and a creative engine that produces and tests at pace, with automation doing the rest.
The job didn't disappear. It got more strategic. The best media buyers in 2026 are part analyst, part creative director and part economist, and that is a far more interesting job than adjusting bids by the hour.