When you sit down for Q3 planning, you’re usually trying to put a few hard questions to rest in one go. What’s still growing? What is being quietly commoditized? Where are the margins taking a hit? And what is the client going to put on your table next, ready or not.
Gartner has put out some new marketing predictions that cut through the noise: the pressure you feel isn’t from one thing. It’s a whole set of things colliding. You have AI redefining how brands talk to people, search becoming more of a meandering process, and trust now being something you can put a number on. Even so, some of the AI commerce tools everyone is making a fuss about won’t be moving the revenue needle any time soon.
For an agency owner, this is the crux of it. Q3 is where you make the call on what to bundle, what to let go of, and what to put a different price tag on. You also decide what your clients will get from you before they can get it from anyone else.
Why you should pay attention
Gartner’s research is worth a look because it doesn’t just give you a forecast; it gives you operating signals. CMOs are putting money into AI, but for the most part they aren’t in a position to scale it yet. The consumer is using AI in their search, but they don’t see an AI answer as the end of the road. That leaves a very particular gap for agencies to fill: your clients want to modernize, but they don’t want to overreact and waste cash.
So don’t read these and think “AI is the world now.” Be more grounded. Figure out which of these calls for a move this quarter and which ones you should just keep an eye on.
1. From running campaigns to orchestrating the journey
Gartner is clear: brands are turning to agentic AI for one-on-one work. Put simply, they are done with one-off campaign stunts. They want AI-driven interactions that last and flow from marketing to sales to support.
The problem is a lot of agencies are still set up by channel. You have a paid media desk, an SEO corner, email over here, and creative in their own lane. Fine when the campaign was king. Not so much when the client is asking if the transition from discovery to retention is seamless.
Your job in Q3 is to start selling orchestration, not just deliverables. A client doesn’t just need more of the same. They need the rules, the data, the QA, the workflows. If you are still pitching “we do campaigns,” you are at risk. Pitch “we build and oversee AI-enabled customer journeys” and you are much stickier.
Don’t go out and buy every shiny new platform. Audit your work. See where you can put in some human oversight and where you can be the layer between the strategy and the machine. My advice: put together a single, no-nonsense offer for lifecycle orchestration. With some journey mapping and performance reporting in it, it’s a lot easier to close than a nebulous “AI transformation” pitch.
2. Authenticity is now a cost of doing business
You will see more of the influencer and creator budget go toward proving authenticity. It’s no longer a soft brand value; it’s a line item. With AI-made content everywhere and social having a say in search, a brand has to show its stuff is for real.
Many agencies are a step behind on this. They treat it like a matter of voice. Clients are starting to demand it be a process.
If you are churning out UGC, social video, or thought leadership, you can’t be casual about it. Vetting needs to be tighter. Provenance has to be checked. One piece of bad content has a way of causing trouble well beyond the post itself, eroding trust in your reviews and your search presence.
Make trust part of your standard operation in Q3. Give your client a framework they can use. Lay down the law on how you label AI, when you verify a creator, and how you handle image manipulation. Put it in the SOW.
And there is an upside: you can put a price on it. Having a set of trust standards is a good reason to put in place a strategy retainer or some brand risk consulting. When it comes to an AI-saturated content landscape, the agencies that can put forward a plan for protecting credibility will have the edge over those still in the weeds of some abstract debate on whether AI is good or bad.
3. Composable, AI-driven teams will put bloated delivery to shame
You might hear Gartner’s take on composable, AI-reliant marketing and think it’s a bit high-level. But if you’re an agency owner, you should be looking at it as a matter of margins.
We can no longer make a case for the way we used to do things: put more heads in the room to get more done. Gartner’s CMO survey has the numbers to back this up—budgets are flat while the demand for AI-fueled growth only goes up. The client wants speed, insight and efficiency, all at once. They aren’t going to write a check for another layer of manual process to make it happen.
That doesn’t mean you go on a hiring freeze. It’s about having a better-designed team. Think in modules: strategy, technical SEO, paid, CRO, analytics, creative, QA. You put them together when you need to. And let AI handle the tedium of the first draft, the tagging, the reporting, the versioning.
For Q3, my advice is to stop letting inefficiency fester in your service delivery. Go in and see where projects are stalling, where an approval is in limbo, or where an account manager is masquerading as a producer. The ones who will be in the best shape in the second half of the year won’t be the largest; they’ll be the ones with the most uncluttered operations.
It’s time to look at your pricing, too. If you’re using AI to work faster but billing by the hour, you’re not just short-changing yourself, you’re training the client to put a premium on time rather than results. Put a package around what you solve: a qualified pipeline, some cost savings, a lift in conversions.
4. New ways to be found, well beyond the funnel
Gartner is right that ambient devices are going to be a bigger part of how a brand is experienced. But don’t fixate on the “smart” in smart device. What matters is that it’s happening outside the normal funnel.
A discovery moment isn’t just a search bar or a click on an ad anymore. It could be a voice command, a nudge from a wearable, or an assistant making a call for the user.
Some agencies are still of the mind that a page view is where intent begins. That’s a limited view. Your clients’ data and content have to be in order so they can make it to any surface. You need product info that is unambiguous, service descriptions that are plain to see, and local signals that are hard to miss. Make it so a machine can read your reviews, FAQs and support material without any trouble.
The question for your Q3 planning isn’t if you need to put together a voice strategy for every client (you don’t). It’s whether they are set up to show up right when someone is being discovered off-page.
So focus on the quality of the content, not the quantity. A well-put-together service page, some good comparison pieces, solid schema, and a FAQ that is actually useful. An agency that makes its client easier for a machine to parse is worth more than one that churns out blog posts and prays for reach.
5. GenAI will be more of a research tool than a revenue driver for now
This is perhaps the most down-to-earth of Gartner’s calls: we don’t expect GenAI to be behind more than 10% of ecommerce in the near future. It’s a good reality check on where the action is versus where the money is.
There is a lot to watch here. Yes, AI shopping is a thing. But the consumer side of the story is different. People are still doing their due diligence. They may see an AI Overview, but they don’t always end there. In fact, some will dig in and look at more options.
And that is where your counsel to the client needs to be. You don’t have to let go of the old guard—traditional search, retail media, PDPs, reviews and the like—just because AI has come on the scene. The point is to have both working in tandem. On one hand, you have buyers who are using AI to put a finer point on their questions and make comparisons. On the other, they are still turning to the tried-and-true to do their due diligence and make a call.
So for Q3, there are two traps agencies should avoid. One is to put too much money into some flashy AI commerce test that doesn’t actually fit with how your buyers operate. The other is to act as if the world hasn’t changed. A wiser approach is to put more behind what drives consideration: put together better FAQ pages, get your product and service info in order, make for starker comparisons and put some life back into your proof points. And for heaven’s sake, let your high-intent pages answer a question in plain English, not just a string of keywords.
Put simply, AI is going to be a bigger part of the research side before it makes inroads on the revenue side. Agencies that see the difference will be in a position to give sounder counsel and keep budgets where they belong.
We don’t need to upend our Q3 plans. We just need to be more discerning about them. You can read Gartner’s take and come away with this: stop the talk on whether AI is overhyped and figure out where it alters the way you work or what you put in a package. The ones who come out ahead in the second half of the year will be the ones that make three things happen: they put some governance on AI, they put a number on trust, and they tidy up how they deliver services. It’s not the most exciting stuff, but it has a way of paying off.
FAQ
What’s the one Gartner forecast an agency owner should be minding?
If you’re looking for a place to start, it’s the shift to agentic AI and the kind of one-on-one it enables. It ripples through service design by changing the client’s view of marketing. When you’re no longer tied to a channel, you can’t just run a campaign and move on; you have to build a system. That has implications for everything from headcount to how you structure a retainer.
How do you plan for Q3 when a client’s budget isn’t moving?
Don’t make it a sales issue, make it a packaging one. They want to grow, but they have to defend their numbers. So your plan for the quarter should be all about being efficient and clear. Go through any custom work with thin margins and see if you can productize it. Ditch the manual reporting. Make a case for your work based on results, not on how many tasks you’ll do. When money is tight, the agency that is the easiest to say yes to will often win out.
Is AI making SEO obsolete?
Not at all. It’s just redefining what good work is. You can’t live on keyword rankings anymore. You have to make sure your client is easy for every surface to read—be it a search engine, an AI summary, or local results. The technical side is as important as ever, along with your information architecture, data, and the rest of the fundamentals.
