Can AI help small businesses compete with big marketing teams?

Much of the discussion around AI and employment focuses on whether jobs will disappear. The reality is more complicated.

Economists Daron Acemoglu and Pascual Restrepo describe automation through a task-based model. Technology can create a displacement effect, where machines take over tasks previously performed by people, but productivity improvements and the creation of new tasks can also increase demand for human labour elsewhere. [1, 2]

For marketing, that distinction matters. AI can already assist with tasks such as copy production, research, analysis and creative execution. But reducing the cost and time involved in these activities can also allow organisations to produce more marketing without increasing headcount at the same rate.

For smaller businesses, that second effect is particularly interesting. The question shifts.

Consider a small business with one person responsible for marketing. They might need to plan campaigns, write website content, manage social media, produce emails, research competitors, analyse performance and brief creative work.

The problem isn't necessarily a lack of ideas. It's capacity. There are only so many hours in the working week.

Research by Shakked Noy and Whitney Zhang provides an indication of what generative AI could mean for this kind of knowledge work. In an experiment involving 453 college-educated professionals completing writing tasks, access to ChatGPT reduced average completion time by 40% while increasing output quality by 18%. [3]

That doesn't mean every marketing task becomes 40% quicker. The research examined a specific group of professional writing tasks, so applying the figure across an entire marketing role would overstate the findings.

But it demonstrates the potential. If AI reduces the time required to complete some routine work, a marketer can potentially spend more of their working week on the activities that require human input:

The research becomes even more interesting when we look at who benefits from AI. Noy and Zhang found that the benefits were greatest among participants who had initially performed less well, suggesting generative AI can reduce productivity inequality between workers. [3]

Other research has found similar effects. A study of more than 5,000 customer-support agents found that access to a generative-AI assistant increased productivity by 14% on average, with the largest gains among less experienced and lower-skilled workers. [4]

Historically, gaining access to a wide range of marketing capabilities meant employing experienced specialists, outsourcing to agencies or accepting that certain activities simply weren't possible.

AI can provide a capable marketer with assistance across a much wider range of tasks: the early stages of research, generating alternative copy, summarising information, analysing data, developing ideas and accelerating content production.

That doesn't make the marketer an expert in every discipline. It does, however, potentially increase the range and volume of work a smaller marketing function can undertake.

Not quite.

AI can narrow some capability gaps, but there is an important warning against assuming it makes expertise irrelevant. Research involving 758 consultants at Boston Consulting Group examined what researchers called the “jagged technological frontier” of AI. [5]

It can be tempting to look at generative AI and conclude that a business which previously couldn't afford a marketing team doesn't need one anymore. Why employ someone to write content if AI can write it? Why pay someone to research competitors when AI can summarise them?

But the evidence points towards a more nuanced future. AI is particularly capable of taking on or accelerating individual tasks. That does not mean it can assume the complete role of a marketer, particularly where the work requires organisational knowledge, judgement, strategy or an understanding of customers.

AI could make less-experienced marketers more productive at precisely the time businesses are questioning whether they need junior marketers at all.

There is a risk to this model. If junior marketers simply ask AI to do their work for them, they may become more productive without becoming better marketers.

Research into AI and learning suggests an important distinction between using AI to augment learning and using it to automate the work itself. In one experiment, students using generative AI initially performed better. When AI access was later removed, the benefit persisted for students who had used AI as a tutor — and disappeared among those who had used it to generate answers directly. [9]

A developing marketer might use AI to interrogate data, challenge their thinking, explore alternative ideas or accelerate parts of a workflow. That's very different from asking AI to produce a marketing strategy and accepting whatever comes back.

AI isn't going to remove every advantage enjoyed by large businesses. Budgets still matter. Brand recognition matters. Experience matters. Data matters. A business with a team of excellent marketers using AI effectively still has more resources than an SME with one excellent marketer using the same technology.

But AI could reduce the importance of headcount as a source of competitive advantage. Some work that previously required significant human time can now be accelerated or supported. Less-experienced workers can see substantial productivity improvements. And small teams can potentially produce and analyse more without proportionally increasing their workforce.

For SMEs, that's a significant opportunity — provided the question is the right one.