AI in Marketing: 86% Use It, Few Actually Win

In 2023, the question was whether to use AI in marketing at all. By 2026, that question no longer exists. According to the HubSpot State of Marketing 2026 — a global survey of more than 1,500 marketers — 86.4% of marketing teams use AI for at least some of their work. Those who don't use it and have no plans to: 1.7%. A year ago, 47% of marketers said they knew how to use AI in marketing; today that figure is 68.2%. This is the fastest skill shift the industry has seen in a decade.

The logical next question: if everyone is using it, who is actually winning? The HubSpot report and real-world cases from 2026 give a surprising answer. The gap has moved. Before, the edge came from using AI at all. Now, the edge comes from how you use it — and from what you do by hand, because AI can't do it.

What the HubSpot Report Actually Shows

The headline number everyone cites is 86.4% adoption. But the more interesting story starts after that.

Understanding has grown — but unevenly. 68.2% of marketers now say they understand how to use AI in marketing (up from 47% a year ago). 67.5% say they know how to measure the impact of AI (up from 48%). Basic skills are catching up fast. What isn't catching up is top-level application: hyper-personalization — delivering targeted messages based on real-time customer behavior — is used by only 12.6% of brands. The gap between "drafting copy in ChatGPT in Notion" and "building a system on AI" is where the real divide lives.

Productivity is up. 26.5% of marketers say AI has significantly increased their productivity; 66.2% say it has helped somewhat or moderately. Around a third of teams are saving 10–14 hours per week; another third is saving more than 15. These are hours, not minutes.

But the quality of output is in question. 56% of marketers surveyed by HubSpot say the internet is now flooded with AI-generated content. 65% report that consumers are getting better at recognizing and ignoring it. 62.7% of marketers believe competing effectively now requires more unique, human-centered content. In other words, the industry is simultaneously generating enormous volumes of AI content and acknowledging that this content is losing on quality.

This is the maturity gap. AI adoption is no longer a differentiator. What differentiates is the separation between a team that uses AI as an accelerator and a team that has built a mediocre-quality production line on top of it.

How to End Up on the Right Side of the Gap

The report contains two statements that sum up the principle clearly.

Johann Wrede, CMO at UserTesting: "I look for people who aren't afraid of AI — people with a curious mindset and the mentality of a coach, not a doer. They coach AI, because AI is the doer." This is a role reversal. A marketer in 2026 is not writing copy — they are managing the process in which AI writes, while the human sets direction, reviews output, discards what doesn't work, and finishes what matters by hand.

Amy Kenly, VP Marketing at The Launch Box: "From a marketing perspective, this means an increased focus on real thought leadership, authentic stories with personality — not the bland, generic content now flooding the world, generated by AI." The advantage goes to those who amplify the human element, not mask it with AI.

From this follows a practical rule: use AI where speed and volume matter, and don't use it where reputation, voice, and point of view matter.

Where AI Works (and Where It Doesn't)

The HubSpot report includes data on specific AI applications. Marketers using AI "heavily or extensively":

• Content creation — 42.5% (plus 38% occasionally)
• Media creation (images, video) — 37.2% (plus 37.7%)
• Automating routine tasks — 35.6% (plus 40.5%)
• Ad automation and optimization — 34.1% (plus 36.5%)
• Strategic planning and forecasting — 33.4% (plus 37.5%)

This is the production conveyor. AI performs well here because the task is well-defined, volume is high, and the cost of a single mistake is low.

Where AI underperforms is where authorial identity, original perspective, and trust are essential. Content marketing for lead generation in a service business sits in the middle ground. Some of that work — drafts, headline variants, images, creative briefs — is handled well by AI. The rest — expert judgment, personal story, a contrarian opinion, a breakdown of a real client case — can only come from a person.

What Is Actually Delivering Results in 2026

The HubSpot report gives direct channel-level data on what works.

Short-form video leads on ROI. 48.6% of marketers name short-form video as the highest-returning format. That is meaningfully ahead of the next options: long-form video (28.6%), live streaming (25.1%), and UGC (24%). For service businesses in Kazakhstan, this is a clear signal: Reels, Shorts, TikTok — these are not "worth trying sometime," they are the primary communication channel in 2026. Phone-camera quality is already sufficient; what matters is consistency and the format of a "30-second answer."

Instagram has overtaken Facebook for the first time. 70% of brands globally use Instagram; 69.6% use Facebook. By ROI, Instagram already leads: 48% of brands rank it in their top 3, versus 42.7% for Facebook. In Kazakhstan, the pattern mirrors the global trend exactly.

TikTok has overtaken X (formerly Twitter). 54.5% of brands use TikTok as a marketing channel. By ROI, TikTok (32% in top 3) has edged past X (31.3%). For services targeting a younger audience — language schools, fitness studios, beauty salons, barbershops — this is no longer an experiment.

Paid and organic social are the top two channels by impact. 25.9% of marketers name paid social as their most effective channel; 23.5% name organic social. Both are well ahead of every other option.

Personalized content works. 93.2% of marketers say personalized experiences have led to more leads and sales. But only 12.6% of brands are doing hyper-personalization. This is an opportunity gap — most understand the value; few have operationalized it.

What is losing ground. The most commonly cut budget items in 2026: out-of-home advertising (23.3% of brands are reducing spend), direct mail (21.6%), print advertising (21.4%). This doesn't mean they're useless — it means they are disproportionately expensive in most cases.

What This Means for Service Businesses in Kazakhstan

From this data, a clear pattern emerges for businesses with monthly marketing budgets of ₸500,000 to ₸5,000,000.

AI for the production layer. Post drafts, headline variants, language translation, format adaptation, baseline images — delegate these to AI. One person with AI replaces a team of two or three at this layer. This is already the norm; doing it all by hand is a waste of time.

Personal voice for the top layer. Posts written in the owner's voice — about real clinic cases, real mistakes, real client outcomes — get written by a person. AI can help with structure, but the text needs to be human. This is what HubSpot calls "human-centered content," and it is where the competitive edge sits.

Primary focus on short-form video. If there is one direction to prioritize in 2026, it is short-form video in Reels and TikTok. Not because it's trending, but because the survey data shows it is the highest-ROI format available. For a dental clinic — "how we treat a root canal," "what to do after implant surgery," "answers to patient questions." For a salon — before/after. For a language school — "one practical thing from today's lesson." Consistency matters more than production quality.

Instagram and TikTok as the two primary channels. Keep Facebook active, but treat it as secondary. If the budget is limited, prioritize Instagram and TikTok.

Standardize personalization. Hyper-personalization is still out of reach for most small businesses. But basic personalization — segmenting by service type, funnel stage, or location — is achievable, and according to HubSpot's data, it works.

Hiring around a coach mindset. When bringing on a marketer or marketing lead in 2026, the relevant question is no longer "do they use AI" (everyone does), but "can they coach the process." These are different skills. The first is technique. The second is a way of thinking.

What Not to Do

From the same HubSpot data and broader industry observation, the anti-patterns:

• Flooding social channels with uniform AI-generated content. It shows immediately, and conversion drops. 65% of consumers already recognize and ignore this content.
• Using AI for thought leadership. Columns, expert posts, case breakdowns — this is human territory. AI will produce the text, but it will not build the reputation.
• Ignoring short-form video. Long-form posts on Facebook and LinkedIn remain relevant but secondary. Short-form video is the lead channel; everything else is playing catch-up in 2026.
• Overloading the team without adding an AI layer. 83.5% of marketers say they are expected to produce more content than before; 35.7% say significantly more. If expectations are rising and headcount isn't, and AI hasn't been integrated systematically, burnout is guaranteed.

Checklist for the Next Month

1. Split marketing work into two layers: production (AI does well) and reputation (requires a person). Write down what belongs where.
2. Establish a regular short-form video rhythm — 2–3 videos per week in Instagram Reels and/or TikTok. Consistency matters more than production quality.
3. Choose one primary channel (Instagram or TikTok, depending on the target audience) and direct 60–70% of budget toward it.
4. Create a template for "personal owner post" — once every two weeks, built around a specific real case. Write it by hand.
5. Connect AI to routine work: post drafts, topic variants, translation between languages, images. Measure how many hours were saved over the month.
6. Check whether AI is being used where a human voice is required — case write-ups, responses to reviews, expert commentary. If it is, rewrite those by hand.
7. When planning for 2027, budget more for short-form video and Instagram; reduce out-of-home and print.

The Core Point

In 2026, marketing has changed — but not in the way "AI everywhere" suggests. It is a maturity gap. Teams that have learned to use AI for production while strengthening the human layer are winning. Teams that simply flooded their channels with AI-generated content and cut their people are losing quality and trust — and in a few months, it shows up in their numbers.

The HubSpot report is not a prediction. It is a map of where money is already being spent, what is showing returns, and what is being shut down. For service businesses in Kazakhstan, this map is useful not as a ready-made strategy but as a filter against hype: free "AI marketing" courses promise magic; the real data shows where results actually come from.

At Alteora, we help service businesses find this balance: AI as a production accelerator, people as the source of reputation, and focus on the channels that actually deliver results in 2026. No innovation for its own sake, and no pretending nothing has changed since last year.