Why an IT department is not enough for a large company
The IT department makes sure systems run without failures. AI changes how sales, purchasing, accounting, production and people work. Only the top manager can lead such changes, together with a team whose main job this is.
According to BCG, about 70% of the difficulties in AI adoption come from people and processes, 20% from technology and only 10% from algorithms. The IT department is strong in technology, but most of the difficulties lie outside it.
The IT department is still needed: it connects AI to the CRM, the accounting system (in Russia often 1C) and telephony, and it manages access. But it does not decide what to change and why.
What exactly the IT department lacks and how to split the roles is covered separately.
Goal
IT departmentKeep systems running without failures
AI department led by the CEOHelp the company earn more
Business
IT departmentDoes not analyze sales, purchasing or production
AI department led by the CEOStarts with analysis: where money is lost
Product
IT departmentDoes not build products, handles requests
AI department led by the CEOChooses a solution and takes it to results
Money
IT departmentSuccess means “we launched it”
AI department led by the CEOEffect in money set before the start and in the contract
Speed
IT departmentA queue of requests and months of approvals
AI department led by the CEOA pilot on one process in 14 days
Own AI use
IT departmentUses AI now and then
AI department led by the CEOWorks with AI daily and teaches departments
Motive
IT departmentProtects its area and budget
AI department led by the CEOAnswers for the company’s results
AI due diligence of the whole company
In a large company we check every unit. AI due diligence, a full check of the company before decisions about AI, shows where AI will bring money and where it will bring risk.
We check the processes of each department, the data and its quality, the systems, the cost of people and contractors, what has already been done with AI, and the contracts with suppliers.
We look separately at scattered pilots: in a large company they often run in several units at once, with different contractors and no shared metric.
We also check the choice of supplier. According to a survey by a major Russian telecom’s cloud unit (December 2025), 91% of Russian companies work with one AI supplier and have nothing to compare its price and quality with.
As a result, management gets one picture of the whole company: where the money is, where the risks are and which unit to start with.
AI transformation program
For a large company, the roadmap becomes a one-year AI transformation program: which solutions we introduce, in what order, with what budget and metric.
Without a program, AI stays a set of pilots. In a survey of 700 companies by the same telecom’s cloud unit, 74% of Russian companies had no AI development strategy.
The program includes both ready-made solutions and development. In a survey of 150 CTOs at large companies by a major Russian tech company and the consultancy Yakov and Partners, 78% use ready-made AI solutions from outside partners, and 71% fine-tune ready models instead of training from scratch.
We put the program into practice in three formats: a pilot from $1,499 in 14 days, an AI department of 3–5 AI agents from $4,999 in 4–6 weeks, and a turnkey business with several departments and a management dashboard from $18,000 in 2–3 months.
The stages of the program and how it differs from digital transformation are described on the page on AI transformation.
AI department
AI works where departments meet, so a large company needs an AI department: a separate unit that links departments into one process and is responsible for the result in money. The roadmap includes its design.
AI director: responsible for the roadmap and the result, reports to the top manager.
Process owners: department heads, each with their own number.
IT department: connections to systems and access. Lawyer and security: data, 152-FZ (Russia’s personal data law), contracts with suppliers.
Training: who in the company is responsible for preparing employees and managers.
Until you have an AI director on staff, we take on the role: from $999 a month, 5 hours a week.
Training for managers and employees
Employees who fear being replaced do not use AI. So for each solution the roadmap records whom to train and in what before launch.
For managers we run an intensive course: from $240 per participant. A department head must understand how to count the effect and how to accept AI’s work, or nobody will check the metric.
For employees, the roadmap sets whom to train on which solution: how to work together with AI, what to check after it, and where to send disputed cases.
Training stands in the plan on a par with rollout: without trained people, a solution’s metric does not grow.
Support until the result
The roadmap does not end with a document. We support rollout until the result written in the metric.
Support for launched solutions: from $480 a month, cancel in any month. It includes metric control, changes and a report.
From the reports we recount the effect on real numbers and decide what to scale and what to switch off. If it does not work, we switch it off.
If several AI contractors already work in the company, support brings their reports down to one roadmap metric.
The AI director links support to the top manager: the result is reported in money, not in project statuses.