What Does an AI Consultant Actually Do?

Most people hiring an AI consultant for the first time do not know exactly what they are buying. That is not entirely their fault. The term covers an enormous range of activity, from someone who helps you pick a tool to someone who redesigns how your entire commercial function operates. The difference between those two things is significant, and knowing which you need before you hire matters.

The Core Job, Stated Plainly

An AI consultant helps a business identify where AI can create measurable commercial value, design the right solution, and ensure it actually gets used.

The emphasis on commercial value is deliberate. The job is not to build impressive technology. It is to deliver a defined business outcome: time saved, revenue generated, or headcount growth avoided. Those are different briefs and they produce very different engagements.

A consultant who starts by asking what technology you want to use is working from the wrong end. A consultant who starts by asking what business problem you are trying to solve, and what success looks like in measurable terms, is working from the right one.

The Five Things a Good AI Consultant Does

This is the most important and most commonly skipped step. Before any technology decision is made, a good AI consultant maps the existing workflows, quantifies where time and money are being lost, and identifies which problems are worth solving with AI and which are not. The diagnostic work is where the commercial value is found. The technology is how it is delivered.

A good AI consultant can build the business case for a board or leadership team, not just the technical specification for a developer. That means documenting the current state, quantifying the cost of the problem, modelling the conservative, base, and upside ROI scenarios, and framing the cost of inaction. If your consultant cannot do this, you will have to, and most businesses are not equipped to do it well.

The solution design should start with what needs to change in the business and work backwards to the technology required. Not the other way around. A consultant who leads with a specific tool, model, or platform before understanding your commercial problem is optimising for the wrong thing.

Building the AI is half the job. Getting a commercial team to change how they work is the other half, and it is usually harder. Most AI projects underinvest in the change management, training, and enablement that determine whether the team actually uses what has been built. A good consultant owns this, not just the technical delivery.

If the ROI metrics are not agreed before deployment, they will not be measured after it. A good consultant defines exactly how success will be measured, establishes the baseline before anything is built, and tracks the results on an ongoing basis. Without this, you cannot prove the value of the investment and you cannot improve it over time.

What Separates a Commercial AI Consultant From a Technical One

Most AI consultants come from a technical background. They think about problems in terms of what the technology can do. That produces technically correct solutions that do not always map to commercial reality.

A commercially-led consultant thinks about problems in terms of business outcome first. The technology is the instrument. The business result is the point.

The practical difference shows up in the brief. A technically-led consultant will ask: what do you want to build? A commercially-led consultant will ask: what does the business look like in twelve months if this works, and what does it look like if it does not?

That second question is the one that shapes whether an AI project delivers commercial value or sits in a drawer.

What a Good AI Consulting Engagement Actually Looks Like

A well-structured AI consulting engagement has five stages.

Commercial audit. The existing workflows are mapped and assessed. Where is the most time being spent? Where is the most manual, repetitive work happening? Where are the biggest delays and bottlenecks? The audit produces a prioritised list of opportunities ranked by commercial impact, not by ease of automation.

Solution design. For each priority workflow, a solution is designed. This includes defining what triggers the AI, what decisions it makes, what tools and data it needs, and when it hands off to a human. This is a commercial and operational exercise, not a technical one.

Commercial case. The ROI model is built: the baseline, the conservative outcome, the base case, the upside, the cost of implementation, and the payback period. This is what gets the project approved internally.

Build and integration. The solution is built and integrated with existing systems. For most commercial teams this means CRM, email, project management, and reporting platforms.

Enablement and measurement. The team is trained on how to work alongside the new system. The agreed metrics are tracked from day one. ROI is reported on a regular basis.

The Questions That Tell You Whether You Are Talking to the Right Person

Before hiring any AI consultant, ask these:

Can you show me the commercial outcomes of a previous project, not the technology but the business result? A good consultant has numbers from delivered projects. Hours saved, costs reduced, revenue generated.

How do you define success before a project begins? The answer should reference specific metrics and a documented baseline. If it references deliverables or features, that is a signal.

What is your adoption plan? The answer should describe how the team will be trained, how adoption will be measured, and what happens if the team does not change their behaviour.

What happens if the AI underperforms? The answer should describe a specific risk mitigation and how the commercial exposure is managed.

Walk away if the consultant leads with tools, models, or platforms before understanding the business problem. Walk away if they cannot give you real numbers from delivered projects. Walk away if the word adoption does not come up unprompted.

Frequently Asked Questions

What is the difference between an AI consultant and an AI developer? An AI developer builds what they are told to build. An AI consultant decides what is worth building and why. A developer optimises for technical correctness. A consultant optimises for commercial outcome. If you already have a clearly defined problem and a clear brief, you may need a developer. If you are still working out what to build, where the value is, or why previous AI projects have not delivered, you need a consultant first.

When should I hire an AI consultant? The right time is before you have decided what to build. Hire a consultant when you have tried AI projects that did not deliver commercial results, when you are unsure how to prioritise AI investment, when you need to build a credible business case for leadership approval, or when you are planning a significant investment and want an independent commercial view before you commit.

How much does an AI consultant cost? A focused AI audit or commercial assessment typically starts at £3,500. A scoped implementation project with AI agents deployed across one commercial function ranges from £15,000 to £30,000. A full commercial function transformation sits between £30,000 and £60,000 depending on team size and complexity. The right benchmark is the fee relative to the conservative ROI case. A well-scoped project should show a credible path to payback within twelve months.

How do I know if an AI consultant is any good? Ask for the commercial outcome of a previous project, not the technical description of what was built. Good consultants have real numbers from delivered work. They talk about hours saved, costs reduced, and revenue generated, not agents deployed or automations built. They ask about your commercial baseline before they recommend anything.

Do I need a full-time AI consultant or a fractional one? For most businesses the answer is fractional or project-based. A full AI transformation engagement does not require full-time resource, it requires the right expertise at the right points. A fractional commercial AI consultant can design the strategy, oversee the build, and lead the adoption work without the overhead of a full-time hire.

What should an AI consultant deliver at the end of an engagement? A documented commercial outcome against the agreed baseline, a working system that the team is actually using, an ROI calculation against the original model, and a clear handover so the business owns what was built. Not a report. Not a prototype. A live system with measurable results.

Jessica Thomas is a Commercial AI Consultant who diagnoses the business problem before designing the solution. If you want an independent commercial view on an AI project, [book a call](https://calendly.com/jess-jessicathomas/30min).