AI in Consulting: When One Consultant Works Like a Team (Part 1/2)

A competitive benchmark in the entry-level range of a tool manufacturer: six comparison brands, the two most similar products per article, every price taken as the median of five to ten sources, a fixed dealer panel, no promotional prices. Two years ago, that was a week's work for an analyst. Today, the result is ready after one day – cleanly documented, refined over several iterations, and prepared as a decision paper.

The difference is called AI. More precisely: AI agents that do not merely write texts but work along a defined methodology – with files, data rooms, spreadsheets, and presentations. We have been using these tools in our engagements for months. Our thesis after this experience: AI does not change what good consulting is made of. It changes how much of it one experienced individual can deliver. Four levers make the difference – provided the foundation is right. More on that later.

Lever 1: A predefined methodology, applied identically every time

Quality in consulting does not come from genius but from method: clear rules for what is measured, how it is measured, and what counts as reliable. In the competitive benchmark mentioned above, this is defined precisely – comparison within the relevant sales channel, median instead of mean, a fixed set of competitors, no discount promotions.

An AI agent sticks to these rules – on the first run just as on the tenth. What used to depend on daily form, time pressure, and the diligence of whoever did the work becomes reproducible. The March analysis can be repeated in September with identical methodology; deviations are then market movements, not measurement errors. For steering a company, precisely this comparability over time is the real value.

Lever 2: One person works like a small team

A classic consulting team divides the work: one person structures, one calculates, one builds slides, one documents. AI agents take over the executing roles in parallel – the consultant orchestrates and decides.

An example from an engagement: for a machining company with around 20 employees, a complete production planning tool was built in a few working days – machine master data for 25 machines, absence planning, an order pool, a weekly schedule, and a capacity dashboard in a two-shift model. That is the scope of a multi-week subproject, delivered by one person with AI support. For an SME of this size, a multi-person consulting team would never have been economically viable.

Lever 3: Experience becomes multipliable

The most valuable capital of a consultancy is its body of experience: Which questions expose a problem? Which patterns announce a crisis? How do you recognize that an organization promises more than it delivers? In practice, this knowledge lives in people's heads – and is applied only as far as the experienced consultant's calendar allows.

Cast this experience into standards – question catalogs, evaluation grids, maturity models – and an AI agent can apply it in full. Our Corporate Health Check, for example, works with a catalog of 135 questions across all ten dimensions of corporate management and a weighted maturity scoring. Project pressure no longer causes questions to be dropped “for lack of time”. The experience is no longer tied to the senior's presence – it takes full effect in every engagement. That is Institutional Learning, applied to consulting itself.

Lever 4: Analyses in hours instead of weeks

Whether it is a data-room analysis in due diligence, a market model with scenarios and sensitivities, or the evaluation of order and capacity data from the ERP: analyses that used to tie up weeks are now produced in hours to days. The competitive benchmark mentioned at the beginning went through five versions until the delineation of the product range was exactly right – a level of detail for which classic project budgets simply never had the time.

More important than the time saved is how the character of the work changes. When an analysis costs hours instead of weeks, you can test more hypotheses, calculate more variants, iterate more often. The discussion with the client shifts from data gathering to conclusions – where it belongs.

What this means for the SME sector

When buying consulting, industrial SMEs often faced an uncomfortable choice: either the large firm with a large team and a fee to match – or the solo consultant whose capacity is quickly exhausted. AI resolves this dilemma to a significant degree: the experienced solo consultant delivers the output of a small team without its cost structure. The daily rate no longer buys just one person's hours, but the capacity of a well-rehearsed team – led by someone who personally assesses every intermediate step.

For the SME sector this means: analytical depth that used to be reserved for corporate projects – complete diagnostics, cleanly calculated scenarios, reliable data work – becomes economically accessible. The result is shorter engagements with more substance instead of long engagements with a lot of presence.

Recommendation

When you buy consulting, in future do not ask only for references but for the way of working: Which methodology is used? Which standards are defined? And how is AI deployed – as a gimmick, or as a tool along these standards?

Because as powerful as these levers are, they only work under certain conditions. AI without clean methodology produces arbitrariness at high speed – and the structure must still come from the consultant. Why that is, and how you recognize the difference, is the subject of part two (link).

‍

Further articles:

AI in Consulting: The Structure Still Comes from the Consultant (Part 2/2) (link)

The BASICON Method (link)

Institutional Learning (Part 1/2) (link)

Peak Performance through Transparency (Part 1 / 2) (link)