2026 AI Predictions & Advice for Modern Marketers From Modern Marketers
What will change in 2026 with AI, and how marketers should respond.
Happy New Year! Hope everyone had a great holiday and is ready to hit the ground running in 2026.
This issue of the newsletter will be a bit different. I’m a sucker for predictions and advice posts, and I have a wealth of knowledge that I can tap into… my previous guests! So that’s what this issue of Stack & Scale is about – we’re going to stack 2026 predictions and advice for marketers.
I asked a handful of my previous guests, who are some of the smartest AI-forward marketers and operators, two simple questions:
– How will marketing change in 2026 with AI?
– What is your advice for marketers using AI in 2026?
It’s still the wild west, and there’s no doubt that we’ll look back one day and laugh at the things we tried and what the AI tools looked like. As we look ahead, the future looks bright.
Dan Sanchez
AI Marketing Consultant & Creator, host of The AI-Driven Marketer Podcast
2026 Prediction
My prediction is that marketers are going to be in trouble, and it’s going to start in 2026 (maybe it’s 2027, but I’m pretty sure it starts in 2026). The big shift is that agents are actually going to be good. Right now, they’re slow, unreliable, and need to be handheld.
What’s been keeping AI from taking a bunch of entry-level marketing jobs is what I call cross-tab/cross-application work. A lot of entry-level marketing work isn’t that complicated (AI could do it easily), but it’s struggled to move across all the tools marketers live in. As agents get better at logging into the browser and doing what you and I would do (copy/paste, recreating newsletters, uploading the right images, working across apps), we’re going to start seeing entry-level marketing positions disappear.
Advice for Marketers
First: marketers have to become AI-driven. Learn how to use the tools. But second, you have to learn how to leverage the things AI will never be able to do: your values, your convictions, your taste, your discernment. If you have years in the game, that starts to matter a lot more.
Also, you have to learn how to tell stories, and how to identify them and tell them well. This is a long game. The smart marketers are building assets that create a competitive moat around their personal brand (and, hopefully, their company brand): unique data, personal brands (even a company with a collection of personal brands), IP frameworks, and other things AI can’t replicate easily. Start building assets like books and frameworks because it’s easier than ever to do so, and they’re becoming increasingly strategic for both companies and individuals.
Jeff Ignacio
Fractional VP and Consultant, Author of the RevOps Impact Newsletter
2026 Prediction
1. The tradeoff of quantity and quality of AI will come decline. Marketers will be able to test out new ideas much more quickly and develop gold content more quickly
2. AI Video is going to unlock the video format for ads in a massive way. The agency model and internal content marketers are going to be able to take on video in addition to text and image
3. Incorporating first-party and third-party data as real-time signals will accelerate marketing campaigns to meet the buyer earlier in their evaluation cycles.
Advice for Marketers
1. Go beyond just chatting with AI. Get comfortable with workflows to 4x your output
2. Test out all of the models. Mix and match them for different purposes.
Trinity Nguyen
CMO of UserGems
2026 Prediction
Many Marketing teams will run on a central intelligence and decisioning layer that unifies data to drive decisions and next-best actions.
This hub will do 3 things at once:
Unify fragmented data & signals from Companies, Buyers (including Buying Groups), and Opportunities.
Turn those signals into prioritization decisions (i.e Account & Contact Scoring) that both Sales and Marketing can agree on - transparent and modifiable.
Orchestrate actions across sales and marketing. Otherwise, insights without actions are just noise.
What was once a black box and dark funnel leading to attribution fights, Marketers in 2026 will have significantly more visibility to make data-informed decisions about “What should we do next, right now, to influence revenue (or increase more in-market buyers)?
Advice for Marketers
By 2026, “AI-powered” will mean nothing on its own. Every platform will claim it. Every workflow will include it.
What will matter is whether AI is directly accountable for a business outcome, especially revenue.
Good AI will:
Reduce noise, not generate more activity.
Explain why a signal matters, not just surface it.
Help teams act on real buying situations tied to real opportunities.
AI for AI’s sake will fade. AI that improves revenue decisions, buyer orchestration, and deal progression will become the de facto infrastructure.
Gray Mitchell
Founder, Igriss
I think that by EOY 2026 photos will be close to impossible to determine if it is AI. Video will be much closer, likely hard to determine.
I think people will get more blind to marketing as a whole, “considering the whole thing fake”
It will become much more focused on resinating with the end user/solving a real problem. Seems like the days of flash and sizzle are slowly fading, or at least raising the bar.
Justin Norris
Sr. Director, Marketing and Web Operations at 360Learning, and Author of AI Builders Blog
2026 Predictions
By 2026, AI won’t feel like a “tool” marketers use. It’ll simply be woven into how work gets done. The real shift is how this raises the quality ceiling. The best marketers will be the ones who use AI to pressure-test their thinking, surface blind spots, and consistently show up sharper than they could on their own. Average work gets automated, and judgment becomes the differentiator.
Advice for Marketers
I think all knowledge workers need to be exploring on two fronts.
First, build or adopt systems that let you delegate your “paper cut” tasks to AI so you can focus on higher-leverage work. Think of this as AI acting like a swarm of personal assistants.
Second, lean into AI for more strategic deliverables to make you better, not just quicker. Treat it like a thinking partner: ask it to challenge your assumptions, surface counter-examples, and tighten your reasoning before work leaves your desk. As AI makes it ever easier to produce slop at scale, your differentiator becomes the things it can’t scale: taste, judgment, and accountability. Knowing how to make the work that matters more intentional is what will set you apart.
Liza Adams
AI Advisor & GTM Strategist and Author of The Practical AI in to-To-Market newsletter.
2026 Prediction
Most companies are still stuck on “faster.” And focusing on productivity alone is a trap. Based on what I’ve seen with GTM teams, the majority of AI use today sits there: 80–85% is “do this task faster,” 10–15% is “do this better,” and only 3–5% is “do it differently.” The teams that pushed past “faster” into “different” showed what’s possible when you reimagine work, doing work that wasn’t possible before.
The risk in 2026 is stopping at speed. If we only teach AI to do our jobs faster, we risk becoming unnecessary. But when we push into quality and innovation, humans become essential because reimagined work requires human judgment, creativity, relationships, and strategic thinking.
Advice for Markers
Real investment is people first, AI forward. The human side matters: the fear is real, and leaders have a responsibility to upskill and reskill employees, not as a nice-to-have, but because the needs have changed and it’s your job to help people compete now.
Scaling also requires infrastructure (governance, quality, and enablement) because not every AI teammate or workflow earns its place, and cross-platform workflows often need specialized skills (often Ops working with IT and legal) to make them reliable and responsible.
Practically, it looks like democratized creation, centralized integration: let everyone build AI teammates where creativity and experimentation happen, but have a dedicated team step in when workflows need to connect to core systems so you scale without chaos.
Liza wrote a whole post that dives deeper into learning, observations, predictions and advice you should check out: People First, AI Forward: What We Learned in 2025 and Where We’re Headed in 2026
Jeff Garwood
AVP, Data and AI Service Line, North America at OSF Digital
2026 Predictions and Advice
A potential future is that as people interact with LLMs, they’ll actually want to see marketing content that aligns with their interests and research in a subtle, integrated way. That means marketers won’t just be pushing ads. they’ll be weaving relevant information into the flow of a chat-like experience, if users opt in. It’s a shift from just presenting ads to informing people right when they’re curious, and it also means LLM providers will need to manage that space carefully. In the end, it’s all about making relevance part of the conversation, but in a way that’s more interactive and user-guided
The idea is that marketing will be something the users want weaved into their ai experiences
It will also mean that marketers need to dramatically change how they’re presenting their value prop and differentiators in an AI flow of discussion.
Sam Kuehnle
VP of Marketing at Loxo, Author of Marketing Meditation with Sam Kuehnle
2026 Predictions
The click is dying as a leading indicator.
Since “”digital marketing”“ began, we used clicks as a proxy for visibility. But AI changes that dynamic completely - it answers questions without requiring the click.
Visibility no longer requires clicks to matter. We’re seeing it clearly at Loxo. Unbranded impressions up triple-digits. Clicks flat. Pipeline from organic up 2x. The click indicator broke, but the correlation between impressions + pipeline didn’t.
This is the same exact lesson Dark Social taught us in 2020: awareness and direct action aren’t the same thing. In 2026, that principle extends to search, aka “Dark SEO”.
Advice for Markers
Stop measuring by channel. Start viewing things as a GTM ecosystem.
Your best SEO work might never generate a click. Your best content might be invisible in your attribution (i.e. your founder’s linkedin post). But it’s building the mental real estate that drives demos six weeks later when someone searches your name directly.
Audit what’s actually in your buyer’s mind, not what your dashboard tells you. Use AI to be more valuable without requiring the click. And most importantly, get ahead of this with your leadership team so they understand before they look at dashboards measuring the “”old”“ way and think that the marketing strategy isn’t working anymore. The teams that survive 2026 won’t be the ones who understand AI first. They’ll be the ones who operationalize it when their CEO still wants clicks.
Sam Calhoun
Founder of GrowthPilot
2026 Predictions
AI will have a barbell effect for the marketing team’s output. Volume-based teams will create a flood of low-quality content. Winning marketing teams will use AI not only to enhance their existing content strategies but also to build foundational systems that free up human time for high-impact creativity. 2026 will be about AI operations, not just AI generation.
Advice for Markers
Become T-shaped in your AI knowledge. Surface-level familiarity with ChatGPT is now the baseline requirement. To become indispensable, marketers must know about a variety of AI applications & tools, and go deep into the implementation of at least one of those use cases.
Learn something new about AI every week, and pick a lane to have deep knowledge in. For example, follow newsletters to stay up to date across the AI landscape, but pick a technical lane to be hands-on with and master, like workflow automations.
Brandon Redlinger,
Hey, that’s me! I write this newsletter 😊.
And I’ve got a lot to say for this…
2026 Predictions
Prediction 1: Most AI and agent projects will still stall or fail.
The thing about AI is a lot of stuff demos really nicely. But then when you go to implement it, it doesn’t quite work the way they show it on the demo. And culprit for a lot of what I’m seeing is poor data foundations, exception handling, and accountability.
McKinsey’s research keeps pointing to the same culprit: companies struggle to capture value with AI projects when data is messy, fragmented, or hard to access across systems.
The other reason this happens is a lack of change management. You can’t expect a major project with grand expectations to succeed without proper change management.
Prediction 2: The speed at which AI deals get done will be much slower than in 2025.
A lot of early AI buying happened in the AI gold rush. That phase is fading fast. Now buyers want to know what data the system touches, what it logs, what it trains on, and what happens when it’s wrong.
In companies both big and small, I’m seeing how much “trust due diligence” is becoming table stakes, including the importance buyers place on privacy and external validation (not just what marketing puts on the website).
Further, AI security incidents are common enough that companies are tightening controls, which adds steps, stakeholders, and time.
Prediction 3: Executive leadership and boards will demand an AI scorecard.
Directors are already saying they lack adequate metrics to assess AI’s impact. And CMOs are under static AI budgets and escalating scrutiny. I recently read Gartner’s research that hints at the pattern: GenAI ROI gets framed as “time saved.” Useful, but incomplete, and certainly not helpful with fighting for more budget.
This is where marketing leaders must start reporting AI, like how finance reports on spend: consistent definitions, trends over time, and quality checks. A few metrics that I think will show up fast: AI CAC ratio, AI time-to-value, and AI-sourced pipeline quality.
Advice for Marketers
Advice 1: Make AI earn its way into your workflow. Before you automate it, break it (on purpose).
Make it impossible to ship cool demos that can’t survive in your work environment. Use this checklist:
Only automate proven workflows: You must know details and what success looks like. AI is not a silver bullet.
One throat to choke: name a single DRI for the workflow. If ownership is shared, it’s owned by nobody.
Exception library: list the top failure modes and define the fallback (auto-fix, route to human, or hard stop).
Data contracts: source of truth, freshness, format, and what happens when blank.
Kill switch + rate limits: if it starts spamming, hallucinating, or breaking compliance, shut it off instantly.
Observability: track accuracy, drift, and downstream impact, not just whether it “runs.”
Advice 2: Run every AI initiative like an investment decision (because it is!)
In 2026, “we’re experimenting with AI” won’t survive budget scrutiny. Marketing leaders who win will treat every AI initiative like an investment memo.
Write the ROI story before you build anything:
What job gets better?
What business outcome moves because of it?
Why will it work in your environment (your data, your workflow, your buyer)?
Price the three real costs (not just the tool):
Build: tools + integration + implementation
Operate: human review + exceptions + QA
Change: enablement + adoption + process updates
Set a hard 30/60/90 value clock:
Day 30: it works reliably inside the real workflow
Day 60: the team uses it without nagging
Day 90: you can show directional impact or you kill it / narrow scope
Force the CFO comparison:
Is this better than doing nothing, tightening the process, or hiring a person?
If not, it’s not ROI it’s not real
Advice 3: Don’t “learn AI.” Own one revenue workflow end-to-end.
If you haven’t gone deep on AI yet, this is where I’d start. Pick one workflow that touches revenue and take responsibility for the whole system. Just start, you’ll learn a TON along the way.
Pick a revenue-critical workflow: outbound, content-to-pipeline, lifecycle nudges, or event follow-up (hint: I’ve given you some full templated plays to run in previous newsletter issues). Start where impact is obvious and stakeholders actually care.
Define the “win”: what does better look like (faster, fewer errors, higher quality, higher conversion, cleaner handoffs)? Pick the one that makes your manager look good.
Own the inputs: list the data you need, where it comes from, and what “good” looks like.
Own the QA: create a checklist for accuracy, voice, compliance, and “no hallucinations.” If it can’t pass QA, it doesn’t ship.
Own adoption: bake it into the team’s existing tools/process (Slack, CRM, docs). If people need a new habit to use it, it won’t stick.
Own reporting: track before/after and keep a running log of wins + misses. The goal is to be able to say: “Here’s what changed, and here’s the proof.”
Commit to a 30-day upgrade: you’re not rebuilding the company. You’re making one workflow meaningfully better in a month, then iterating weekly.
Probably my top tip for this is to use AI!!! If you get stuck anywhere along the way, hop into your favorite LLM chat interface and ask it to guide you.
Trinity & Brandon’s predictions are the most accurate. @Brandon do you have an AI Scorecard resource or post about one?
Great piece, Brandon. Lots of good advice packed in here.