Why ad selling still runs through a person
Advertising sales agents sell space and time to businesses: a print page, a radio slot, a banner, a sponsored segment. Two tasks drive the job. One is finding and contacting potential clients. The other is negotiating prices and terms until a buyer signs. Software can draft the pitch. It cannot carry the accountability that a media buyer wants when a campaign underdelivers.
The pressure here is real, though. Self-serve ad platforms and automated buying already absorb a large share of small, repeatable orders, and that is the exact work junior reps used to cut their teeth on. The Bureau of Labor Statistics counts about 91,700 advertising sales agents in the United States, with median pay of $64,820 (BLS, 2025), and projects employment falling 7.2% between 2025 and 2035 (BLS projections, 2025). That is erosion at the entry end, not a job disappearing.
What survives is the part buyers argue about: a custom package across formats, a makegood after a bad flight, a renewal with a client who has been burned before. Our coverage score, which asks how much of the task time AI can handle today, reads 41 on the 0 to 100 scale explained on the coverage method page.
What AI runs, what it drafts, and what the rep keeps
AI handles the repeatable paperwork of the role on its own. Preparing promotional plans, sales literature and standard proposals is one example. Maintaining client records and sales activity logs is another. That group accounts for 14% of task time in our breakdown.
A larger slice is shared work, where the tool speeds a rep up but someone checks it. Locating and contacting potential clients now starts with a scored list and a drafted first email. Explaining rates, formats and audience numbers to a buyer starts with a pulled report. Both still need a person to judge what is worth sending and what is wrong. That shared group is 68% of task time.
Then there is the part that stays with the agent: 18% of task time. Negotiating contract terms sits here, because the concessions are relationship decisions, not pricing math. So does recommending a campaign approach after a client’s product, budget or boss changes mid-quarter. None of this needs hardware, which matters: there is no physical task to automate, so no robot has to be built and shipped before the work shifts.
What the evidence does and doesn’t show
There is no published head-to-head test of AI against working advertising sales agents. That is why our quality-parity grade for this job is D, and why this page gives no parity number. A grade like that means not measured, not measured and failed.
What would settle it is specific. A publisher or ad network would have to run matched territories for a full selling season, one worked by reps with AI support and one by an AI-led outreach and proposal system, then publish close rates, average deal size, renewal rates and revenue retained. Campaign-level renewal data would matter more than reply rates, because replying to an email is not selling. Until something like that exists, treat confident claims about AI outselling media reps as marketing. The grading scale is set out on the quality-parity method page, and our full approach is on the methodology page.
When the balance could shift
Most likely between 2035 and 2045 (8 in 10 of our scenarios). The replacement-year method page explains how that window is built and why it is a range rather than a date.
Two things could pull it closer. Cost is the first: running the software side of this work is cheap next to a salaried rep on commission, and buyers who are happy with self-serve rarely go back. The second is scope creep in the tools. Once an assistant can research a prospect, build the package, price it against inventory and follow up without a human in the loop, the entry-level rep role thins out further.
Two things hold it back. Buyers want a named person on the hook when a campaign misses, and contracts are signed between people who expect to call each other. And the money that matters in media sales is concentrated in a few large, negotiated accounts, where terms are bespoke and trust takes years. Those are hard to hand to a system that cannot be held responsible.
What to do: move your week toward negotiated and at-risk accounts, and let the tools take the list-building and first drafts.
How to stay needed in media sales
Lean into the three tasks that are hardest to hand over. Negotiate the custom package yourself, including the trade-offs on placement and makegoods. Own the problem account: the client whose campaign underperformed and who needs a plan, not an apology. And take the strategy conversation, where you tell a buyer which format actually fits their product and budget, even when it means a smaller order this quarter.
Two skills pay. First, campaign data literacy: reading delivery, reach and attribution reports well enough to argue with a buyer’s analyst. Second, using AI tools with verification, so drafts and prospect research save you hours without putting a wrong rate or audience figure in front of a client.
If you are weighing a move, the closest work sits nearby. Sales Representatives of Services is the nearest sibling role, while Sales Managers and Advertising and Promotions Managers are the usual steps up. You can see how the whole group scores on the sales representatives family page, or how the wider media industry looks on the information sector page.
Two more things worth a look: our list of jobs expected to shrink, which uses official projections like the one above, and the side-by-side comparison tool if you want to put this job against the one you are considering next.