AI Consultancies in London: Which Kind Fits Your Business? (2026)

AI consultancies in London come in four kinds: enterprise AI and data consultancies, cloud-platform specialists, public-sector specialists, and studios for small and mid-sized businesses. Which kind fits depends on your size, your sector, the platform you already run on and whether you want the same firm to build what it recommends. For a small or mid-sized business, size usually settles most of the question. This guide covers what each kind does, who it suits and who it suits less.

We wrote this guide, so here is the disclosure. CodeLeap is one of the six firms described, and this is not a ranking or an endorsement. We picked a handful of well-known firms to illustrate the kinds, not every firm. Facts about each firm come from its own website as we checked it on 1 November 2026. A firm that wants a detail corrected can contact us and we will fix it. We offer AI consultancy ourselves, so we say plainly where we fit and where we do not.

Key takeaways

• Four kinds of AI consultancy operate in London, and each is built for a different buyer. Your size, sector and platform narrow the choice before any firm’s name matters.
• None of the five other firms here publishes prices on the pages we read, so for them cost is a first-call question.
• A cloud-platform specialist suits a business that has chosen its platform. A studio suits a business that wants a fixed scope and a published starting price.
• Whichever kind you pick, check that the advice stands alone from the build and that you own the result.
• Our own published starting points are £3,000 for a one-week AI readiness assessment and from £5,000 for a first AI agent.

What does an AI consultancy in London actually do?

An AI consultancy helps you decide where AI is worth using in your business, what each use should cost and in what order to do them. The work usually has four parts: finding candidate processes, checking whether your data and systems can support them, ranking them by payback, and writing a roadmap. Some firms stop there. Others carry on and build, and the engagement becomes a software project.

The need is common. In the government’s AI adoption research, 71% of businesses cited not having identified a use for AI as a barrier. Limited AI skills, expertise or knowledge was cited by 60% (DSIT, fieldwork February to May 2025). Choosing a use and a sequence is the consultancy’s job. Our guides to what an AI automation consultant does and to a 90-day AI transformation roadmap cover the work itself.

Which kinds of AI consultancy can you choose from?

AI consulting firms in London fall into four kinds, each built around a different buyer. The table lists every firm in this guide, alphabetically.

Provider	Type	Built for
CodeLeap	Small and mid-sized studio	UK SMEs, fixed scopes
Datatonic	Cloud-platform specialist	Google Cloud data and AI
Faculty	Enterprise AI consultancy	Public services, defence, health, finance
Mesh-AI	Enterprise AI and data consultancy	Regulated enterprises
Methods	Public-sector specialist	Public sector, regulated bodies
PA Consulting	Enterprise AI consultancy	Eight industries, government included

The grouping is ours, based on what each firm says about itself, and some firms sit between two kinds. Every firm here states a London office. That shows where it has a presence and not who will do your work, so ask, as our guide to choosing a software development company in London explains.

The four kinds of AI consultancy, with examples

Each firm below has the same five fields, and firms are alphabetical within each kind. Prices appear only where a firm’s own site shows them.

Enterprise AI and data consultancies

These are firms whose own sites point to public bodies, regulated industries or large organisations. They run strategy, data and AI programmes, and many take the work through to delivery.

Suits: an organisation with several departments to bring along, formal governance or procurement, or a regulated sector where a firm has worked before.

Suits less: a business that wants one process improved and a price before the first call. These three firms describe work for public bodies, regulated sectors and many industries, and none publishes prices on the pages we read.

Faculty

• Based in: London (Old Street). The pages we read list no other office.
• What it offers: AI strategy as a listed core service, described as working out how to use AI for your strategy and then delivering the plan.
• Says it serves: organisations in industries that include public services, defence, national security and law enforcement, health and care, energy, financial services and insurance. Its site states no client size.
• How it works and charges: prices not published.
• Example work: forecasting hospital demand and capacity for the NHS during the pandemic.

Mesh-AI

• Based in: London (Liverpool Street), with a second office in Lisbon. Its site says it is now part of Indicium AI.
• What it offers: a data and AI strategy accelerator, a data and AI readiness assessment, and building from concept to production.
• Says it serves: highly regulated enterprises, with a stated specialism in financial services and in energy and utilities.
• How it works and charges: it describes a transformation framework it calls Orbital. Prices not published.
• Example work: AI content curation at LSEG Risk Intelligence.

PA Consulting

• Based in: London (Bressenden Place), with other UK offices in Belfast, Bristol, Edinburgh and Manchester. Its About page gives its size as about 4,000 people.
• What it offers: AI, analytics and data within a wider offer called Reimagine AI, digital and data, which also covers digital transformation and technology strategy.
• Says it serves: eight industries, including government and public sector, defence and security, energy and utilities, financial services and health. Its site does not describe clients by size.
• How it works and charges: prices not published.
• Example work: an AI colleague for nuclear decommissioning, for Sellafield and the Nuclear Decommissioning Authority.

Cloud-platform specialists

These firms build data and AI work on one cloud provider’s platform. Their own pages name that provider’s products throughout, so the platform comes first. Google Cloud sells its own AI products, and a partner of this kind builds on them.

Suits: a business that already runs on that platform, or has decided to, and wants the data and AI work built there.

Suits less: a business that has not chosen a platform or runs on another one, because the firm’s offer is organised around one provider.

Datatonic

• Based in: London (Canary Wharf). Its contact page also lists offices elsewhere in Europe and in Montreal.
• What it offers: generative AI, Gemini Enterprise, a unified data and AI platform, cloud data migration, managed services, marketing analytics, Looker and training.
• Says it serves: financial services, gaming, media, retail and telecommunications, on its industry pages. Its site states no client size.
• How it works and charges: its partner page describes a partnership with Google, and its service list is built around Google Cloud products such as Gemini Enterprise and Looker. Prices not published.
• Example work: a generative AI platform for Vodafone on Google Cloud.

Public-sector specialists

These firms say they serve government, health, justice, defence and other public bodies. The government’s AI Playbook has a Buying AI section that sets out how public bodies are advised to buy, including business cases and routes to market such as frameworks.

Suits: a department, local authority or regulated organisation that wants a supplier whose pages are organised around public-sector work.

Suits less: a private business with no public-sector or regulated requirements, since the firm’s pages are written around public bodies.

Methods

• Based in: London. The pages we read list no other office.
• What it offers: data and AI services including data strategy, advanced analytics and business intelligence, with a five-step approach from collecting data to applying AI.
• Says it serves: public-sector departments in defence, health, home affairs, justice, local government and transport, among others. Its site also lists private-sector clients in energy, commodity and utilities, and in financial services and insurance.
• How it works and charges: it offers flexible capability on demand, and says it combines specialist UK expertise with nearshore and offshore delivery. Prices not published.
• Example work: AI-powered guidance for justice decision-making.

Studios for small and mid-sized businesses

These are small studios that work with businesses without a data or AI department. They suit a small or mid-sized company with a named process to fix, or a question about where AI would pay off. They suit less a programme with heavy governance across thousands of staff.

Other small studios also work with SMEs. We describe only ourselves in this group because we can vouch for our own facts.

CodeLeap (us)

• Based in: London. We are led from London, with a remote, global team of around 28.
• What it offers: AI consultancy covering AI audits, data readiness reviews, roadmaps and implementation planning, a fixed-price AI readiness assessment, and builds such as AI agents and automation.
• Says it serves: UK SMEs.
• How it works and charges: scope and price are fixed and agreed before work starts. You pay by milestone, and all code and IP are yours once delivered and paid for. The AI readiness assessment is £3,000 for a one-week programme. A follow-on build starts from £8,000 and a first AI agent from £5,000.
• Example work: Dellar Davies, a specialist events agency in the travel sector. We ran discovery across its functions, assessed tools by category, held a team workshop and delivered a prioritised roadmap.

How do you tell which kind you need?

Answer four questions, and the kind usually follows.

• How large is your organisation, and who must approve the spend? A small or mid-sized business with one decision-maker points to a studio. Several departments, a procurement process or a board paper point to the larger consultancies.
• What sector are you in? Public bodies and regulated industries have specialists written around their rules. If that is you, look there first.
• Which platform do you already run on? If your data and tools sit on one cloud provider, a specialist for it shortens the path. With no platform yet, one is early.
• Do you want the same firm to build? Some firms advise only, some build only, and many do both. Decide whether you want one team or two.

If you cannot yet name a process worth fixing, score yourself first. Our AI readiness checklist takes about 15 minutes.

What does a first engagement look like, and what does it cost?

A first engagement is usually a short piece of advice that ends in a written plan. The plan ranks your processes, says what data and systems each needs, recommends building, buying or leaving each one alone, and prices a first project. Larger firms may run it as a programme of discovery across many functions.

Cost is where the kinds differ most visibly. Datatonic, Faculty, Mesh-AI, Methods and PA Consulting publish no prices on the pages we read, so for them the figure comes from a conversation. We can only give published prices for ourselves.

• AI readiness assessment: £3,000 for a one-week programme. It includes three calls, your data, workflows and team skills mapped, and a proposal to keep. It gives an honest verdict where AI is not worth it, with no obligation to build.
• Follow-on build: from £8,000, taking the top opportunity from assessment to a live system.
• First AI agent: from £5,000, for one scoped job.
• Support: from £400 a month.

Scope and price are fixed before work starts. For a build, you can also cancel within the first 21 days and get your money back in full.

A build is easier to see in a case. SBU, a UK energy broker, had us build an automation layer across Salesforce, an LLM and its email. Within three months its team was generating five times the qualified leads, without adding headcount.

Which questions should you ask in a first call?

Ask every firm on your shortlist the same questions, in writing, and compare the answers.

1. Who does the work, by name, and are they employees?
2. What will I hold at the end, and could I give it to another builder?
3. If you also build, how do you keep the advice independent of what you can sell?
4. Which of my systems do you need access to, and how will you reach them?
5. Where is my data processed and stored, and which AI models does it pass through?
6. Where does a person approve what the AI does?
7. What does it cost, when do I pay and who owns the result?
8. What would make you tell me that AI is not worth it for us?

Question 5 deserves a written answer. The ICO’s guidance on AI and data protection covers accountability, transparency, lawfulness, accuracy, fairness, security and individual rights. The page says it is under review because of the Data (Use and Access) Act, so check the current version.

The ICO also says a data protection impact assessment is required where processing is likely to result in a high risk. Its criteria include innovative technology such as AI (ICO). If a firm will build something, ask how it follows the NCSC’s guidelines for secure AI system development. They cover secure design, development, deployment and operation. This is not legal advice.

Your next step

Write the one-page brief first: the problem, who uses the result, which systems it touches and what a good outcome looks like in hours or pounds. Then send the same brief to two or three firms of the kind that fits you, and compare what comes back in the same shape.

If that kind is a studio, we would be glad to be one of the three. A free 20-minute call will tell you whether our AI consultancy work or a one-week AI readiness assessment is the right first step. If the answer is another kind of firm, we will say so.