To scale consulting with AI without multiplying your hours, you first write your method down as rules you approve, and then let agents and a virtual consultant work from them. You keep what cannot be delegated: approving the rules, deciding about people and answering to the client.
This article is part of Legathum Legacy.
In short
- Scaling a method means taking on more projects without every step going through your diary.
- The first job is writing: your rules, in plain language, in a Book of Criteria.
- Agents and a virtual consultant take the part you repeat the same way with every client.
- Approving rules, deciding about people and closing the agreement stay with you.
- Each client's virtual consultant selling and delivering from start to finish is in development.
What scaling a consulting method with AI means
Scaling a method means being able to take on more projects without every step going through your diary. With AI, that happens once the method stops living only in your head: it is written down as rules, and there are agents and a virtual consultant that work from them.
It is different from hiring more people and from turning the method into a course. In the first case the judgement is spread among whoever joins. In the second it stands still. Here the judgement stays yours, because no rule counts until you approve it.
The virtual consultant is an AI virtual clone: the replica of a real person that works with their judgement and says it is an AI. At Legathum, the platform that puts it to work for a method is called Legathum Legacy, and its definition is short: your method, scaled with AI.
Where your hours go today
A consulting project takes hours at four moments: selling, understanding the company, delivering and following up. In all four there is a part you repeat the same way with every client and a part that needs you there. The table separates the two:
| Phase of the project | What repeats and AI can take on | What still needs you there |
|---|---|---|
| Selling | Explaining the method and answering the usual questions | Negotiating and closing the agreement |
| Understanding the company | Asking the diagnosis questions and putting the answers in order | Reading what does not fit and deciding the approach |
| Delivering | Applying the approved rules and preparing the documents | Deciding about people and approving what goes out |
| Following up | Tracking progress and flagging what drifts | The difficult conversations with management |
The middle column is the one that takes your hours today without needing your judgement in each case. It is also the one that can be written down.
First step: write the method down
An agent cannot work with what is not written. That is why scaling starts with a writing job, which you do once and revisit when the method changes.
Start with what you already repeat the same way with every client: which questions you ask first, which signs you look for and what you recommend depending on the answer. Each rule goes into a Book of Criteria in plain language, with versions, and does not count until you approve it. When you change a rule, it changes in every project at once.
If your method still changes with every project, that is the first job. Legathum Legacy scales the method you already have.
The three pieces that scale a method
Legathum Legacy has three pieces that work on the same method:
- The Book of Criteria: the rules of your method in plain language, with versions and human review. It is what the agents and the virtual consultant work from.
- The virtual consultant: a clone with your avatar and your voice that works with those rules and says it is an artificial intelligence the first time it speaks.
- Hybridising the companies you work with: you automate processes across every area with agentic AI, with human oversight at the critical points and a record of every decision.
The third piece is the same company hybridisation with AI that Legathum applies in its own consultancy. The difference is that you deliver it, to your clients, with your method.
What the virtual consultant does today and what is in development
It helps to separate what exists from what is coming, so you decide on what is there.
What exists today: each twin's Master Brain and Book of Criteria, with reviewed and approved rules; the face and voice from a single recording of about seven minutes; consent by email with a video, the manual review and revocation with one click; introducing itself as an AI the first time it speaks; and the twin on Google Meet, Teams and Zoom, in more than 20 languages.
Before it works, the twin passes the Mirror Test. Real cases the person has already solved are given to it without the answer and compared blind. It only works if it agrees with the person in at least 90% of the cases and crosses no red line.
In development: each client's virtual consultant selling, scheduling, charging for and delivering their service from start to finish without the person stepping in. That is what it is designed to do. Today, at Legathum, the 15-minute meetings are taken by Deibson Silva in person.
What does not scale
Four decisions stay yours, with AI or without it:
- Which rule goes into the Book of Criteria.
- Decisions about the people in the companies you work with: hiring, promoting or moving someone to another role.
- The agreement with the client: negotiating, closing and answering for the result.
- Your consent to the use of your face and your voice, which you can withdraw with one click.
The second matters most to anyone who works with teams. The EU AI Act treats as high risk the systems used to select staff or to take decisions about the employment relationship (Annex III, point 4), with obligations from 2 December 2027, and the GDPR (Article 22) protects people from decisions based solely on automated processing. Anyone talking to a virtual consultant must be told they are dealing with an AI, from 2 August 2026 (Article 50). These references are EU law; other jurisdictions have their own rules.
Where to start
Before talking about platforms, do this exercise with your last project:
- Pick the part of the method you explain the same way to every client.
- Write its first ten rules, each with its source: where it comes from and when it applies.
- Mark what you never delegate.
- Count how many hours went into the middle column of the table.
With that list you know whether your method is ready to scale or whether it has to be written down first. The detail of what the virtual consultant can take on is in AI virtual consultant: what it can deliver of your method and what stays yours.
Legathum Legacy is part of the Organizational Consultant Licence and the Forward Deployed Engineer Licence from SciTechBiz Institute. The terms are on the page for the Organizational Consultant Licence. The same part of the platform, taken by a company for its own operation, is the Legacy plan of the Legathum platform.
Frequently asked questions
Can you scale a consultancy with AI without losing your own judgement?
Yes, if the judgement is written down. The agents and the virtual consultant work with the rules in your Book of Criteria, and none of them counts until you approve it.
Which part of my method can AI take on?
The part you repeat the same way with every client: explaining the method, asking the diagnosis questions, applying approved rules and following up.
What stays mine?
Approving the rules, deciding about people, negotiating and closing the agreement, and answering to the client for the result.
Can the virtual consultant sell and deliver my service from start to finish?
That is what it is designed to do, but each client's virtual consultant selling, scheduling, charging and delivering from start to finish is in development.
How do I get access to Legathum Legacy?
With the Organizational Consultant Licence or the Forward Deployed Engineer Licence from SciTechBiz Institute, which also include iPersonality and the Academy.
Where do I start?
By writing down the rules of the part of your method you repeat most. To see whether it fits Legathum Legacy, the first step is a 15-minute meeting with Deibson Silva.
Sources
- Regulation (EU) 2024/1689 (AI Act), Article 50 (transparency obligations). Consolidated text, European Commission AI Act Service Desk.
- AI Act, Article 113 (entry into force and application). AI Act Service Desk.
- AI Act, Annex III (high-risk AI systems, point 4). AI Act Service Desk.
- Regulation (EU) 2026/1744 (Digital Omnibus on AI), amending Regulation (EU) 2024/1689. Official Journal text in the BOE database, in Spanish.
- European Data Protection Board, Guidelines on automated individual decision-making and profiling (GDPR, Article 22).