Why the gap is showing up now
AI tools have arrived faster than training. In Microsoft and LinkedIn's 2024 Work Trend Index, a survey of about 31,000 knowledge workers in 31 markets including the UK, 75% said they used generative AI at work. Of those users, 78% were bringing their own tools rather than using ones their employer provided, and only 39% said their company had given them any AI training (Microsoft and LinkedIn, 2024). In other words, adoption is being led by individuals, not by a plan.
UK business data tells a similar story from the employer side. The Office for National Statistics found that around 35% of UK businesses with 10 or more employees used at least one AI technology in June 2026, almost three times the 12% recorded in autumn 2023. Yet only 11% of those businesses said more than half of their workforce had received AI-related training (ONS, 2026). Use is spreading much faster than capability.
The gap is rarely about intelligence or willingness. It is about time, standards and practice. The 2025 Work Trend Index found that 80% of workers and leaders said they lacked the time or energy to do their work (Microsoft, 2025). Asking an overloaded team to teach itself a new way of working, with no agreed standard, produces exactly the uneven results most managers now see.
Why marketing teams feel it first
Marketing was one of the first functions to adopt generative AI, because so much of the work is writing, editing, research and reporting. That early start has not been matched by training. The Marketing AI Institute's 2025 survey of more than 1,800 marketers, mostly in the US, found that 60% of marketing teams were piloting or scaling AI, but only 32% of firms gave their marketing teams AI-specific training. More than half had no policy guiding how staff should use AI (Raconteur, 2025). That is a US-weighted sample, but the pattern matches the UK figures above.
Marketers know the skill matters. In the American Marketing Association's 2025 skills report, based on 1,279 practitioners, generative AI was the top-rated future skill, with 43% expecting it to become more important over the next five years (AMA, 2025). The issue is not awareness. It is turning that awareness into a shared method.
AI also exposes gaps in the fundamentals. An Ipsos study run with Professor Mark Ritson across the UK, US, Canada and Australia found that only 35% of marketing practitioners could correctly define core foundational concepts (Decision Marketing, 2026; the sample size was not published in that coverage). A tool can draft a positioning statement in seconds, but someone still has to know whether it is right. Teams with weak fundamentals tend to accept weak AI output, because they cannot see what is missing.
What the gap looks like in practice
The skills gap rarely announces itself. It shows up as small, persistent problems that are easy to blame on the tools. Look for these signs before deciding what to train:
- Work takes as long as it did before, even though licences are in place and people say they use AI daily.
- Content, reports and proposals vary noticeably in quality and tone between team members.
- Nobody can say which AI outputs were checked, by whom, or against which source.
- Staff use personal accounts for client or company data because there is no approved tool or policy.
- Leaders cannot see where AI is saving time and where it is quietly adding rework further down the line.
- Prompts and workflows live in individual notebooks, so good practice leaves when the person does.
A five-step method to close the gap
The most common mistake is to start with the software. A better order is to start with the work, agree a standard, then train against real tasks. The steps below suit a team of five to fifty people and can be run in about six to eight weeks.
For example, a manufacturer with a three-person marketing team might find that product page copy, monthly reporting and trade show follow-up emails take most of their week. Those three tasks, not a general AI course, become the training plan.
- Audit the tasks, not the tools. List the five or six things each role does every week. Mark where AI could help, and where human judgement must stay. Ask each person to share one recent piece of AI-assisted work that went well and one that did not.
- Agree a quality standard. Write down what good output looks like for each priority task, including tone, accuracy, sources and who signs it off. This is the step most teams skip.
- Build a tested workflow per task. For each of the two or three priority tasks, document the brief, the prompts, the sources the tool may use and the review step. Test it on real work before rolling it out.
- Train in short, practical sessions. Teach the workflow on live work, in small groups, with time to practise. Pair stronger users with those less confident.
- Review monthly and adjust. Share what worked and what did not. Update the workflows, retire what is not helping, and add the next task once the first ones are stable.
Train, hire or bring in support
Most gaps close faster by training people who already know the business. The ONS found that training or retraining existing staff is the most common way UK businesses bring in AI skills, reported by around 40% of medium and large firms. Among businesses that named a lack of AI expertise as a barrier, around 62% were training or retraining staff (ONS, 2026). Lack of expertise was also one of the barriers most often cited for delaying adoption, at around 18% among businesses with 100 to 249 employees.
Hiring still has a place. In the 2024 Work Trend Index, 66% of leaders said they would not hire someone without AI skills (Microsoft and LinkedIn, 2024), and in the 2025 edition 35% of managers said they were considering hiring AI trainers to guide adoption over the next 12 to 18 months (Microsoft, 2025). Those are global survey figures from a technology vendor, so treat them as a signal of direction rather than a UK benchmark.
Public support is growing too. The UK government aims to upskill 10 million workers in AI skills by 2030, up from an earlier target of 7.5 million (GOV.UK, 2026). Free courses help with general literacy, but they will not teach your team your workflows or your standards. That part has to be built in house, or with a partner who works on your real tasks.
How to measure progress
If you cannot measure the gap, you cannot tell whether training closed it. Pick a small set of measures and track them monthly from the first audit. Keep them tied to the tasks you chose, not to general AI usage, because licence logins say little about quality.
- Time to complete each priority task, measured before and after the new workflow.
- Rework rate: how often AI-assisted work is sent back for correction, and why.
- Quality score against the agreed standard, from a sample reviewed by a senior person.
- Share of the team trained and confident on each workflow, checked by a short practical test.
- Share of AI use that runs through approved tools and documented workflows.
Why this matters for AI visibility
Marketing and sales depend on clear, accurate content, and AI assistants now read that content and summarise it for buyers. A team that cannot brief, check and publish well with AI will struggle to be named accurately in AI answers. Thin, unchecked or contradictory pages give those systems little to work with.
Our AI visibility work and content service both start from the same discipline: clear facts, checked sources and a repeatable process. For the wider context, read our post on AI visibility tools for B2B brands, and our view on how AI is reshaping jobs and skills.
When you talk to a training provider or agency, ask them to show you a workflow they built around a real task, how they define quality, how they measure rework and time saved, and what your team will be able to do without them after three months. Vague answers about tools and prompts are a warning sign.
Put this into practice
Book a discovery call and we will review your channels and results first, then agree what to test.
Talk about your team's AI readinessQuestions and answers
What is the AI skills gap?
It is the difference between how much a business expects its people to use AI and how well they have been trained to use it. In the UK, around 35% of businesses with 10 or more staff used AI by mid-2026, but only 11% said more than half their workforce had received AI training, according to the ONS.
Do we need to hire AI specialists?
Not always. Most gaps close faster by training people who already know the work, and training existing staff is the most common route UK businesses take. A specialist helps set standards and build workflows, and is most useful once you have a clear list of tasks to improve.
How long does it take to close the gap?
A first round of task audit, standards and training usually takes six to eight weeks for a small team. The gap closes over months, as people build habits and the team reviews time saved, rework and quality each month.
What should AI training for a marketing team cover?
It should cover the team's own priority tasks, such as product copy, reporting or campaign briefs. Each needs a written quality standard, a tested workflow with prompts and approved sources, a review step and a named owner. General tool tours add little on their own.
Is free government AI training enough?
It is a useful start for basic literacy. The UK government aims to upskill 10 million workers by 2030 through free courses. But general courses cannot teach your workflows, quality standards or client rules, so plan practical training on your own work alongside them.
References
- Microsoft and LinkedIn, "2024 Work Trend Index Annual Report: AI at work is here. Now comes the hard part", 2024
- Microsoft, "2025 Work Trend Index: The year the Frontier Firm is born", 2025
- ONS, "Artificial intelligence in UK businesses: 2023 to 2026", 2026
- Raconteur, "Marketers: upskill in AI or risk redundancy" (reporting the Marketing AI Institute 2025 State of Marketing AI Report), 2025
- American Marketing Association, "2025 Marketing Skills Report", 2025
- Decision Marketing, "Marketers feel the heat as AI exposes skills shortage" (reporting Ipsos, The Marketing Anchors), 2026
- GOV.UK, "AI Skills Boost explainer", 2026


