Assessment
AI assessment
Most companies don't know what AI can do for them until someone looks at how they work. That is the assessment. First the assessment, then the technology.
- Process analysis
- Opportunities and risks
- Possible automations and agents
- Priorities and roadmap
The challenge
Most companies don't have a tools problem: they have a judgment problem. There is pressure to apply AI, plenty of offers on the table and little clarity about which processes are worth it and which aren't.
Meanwhile, teams are already using AI on their own: personal accounts, client information in public tools and no written rules about what can be shared.
The solution
We audit how your company works today: processes, tools, data and the actual use already being made of AI. From that come the use cases worth pursuing, ordered by impact and effort, and the ones better left aside. The conclusion may be to automate, integrate, use AI, implement an agent, change the process or do nothing for now.
We deliver a roadmap with what needs to be done, in what order and with what safeguards: security, data handling, who decides what and what training the team needs. You can execute it with us or with whoever you prefer.
What we analyze
An assessment of how your company works, not a talk about AI
We look at the whole company from the point of view of the work done by hand and the systems in between:
Processes
What is done, who does it, how long it takes and where it gets stuck.
Current tools
ERP, CRM, email, spreadsheets, internal software and what can be connected.
Data
Where it is, in what state, how it moves and who touches it.
Risks
Access, sensitive data and the use of AI already happening without control.
Opportunities
Where the most time is lost and where an automation changes the day-to-day.
Possible automations
Which processes are solved with rules and integration, with no need for AI.
Possible agents
Which multi-step processes justify an agent and with what limits.
Recommended tools
Which technology fits your sector, your data and your team.
Priorities
Order by impact and effort, including what is best left aside.
Roadmap
Phases, dependencies and how each one is measured.
What you take away
- What to automate
The processes with real return, described step by step.
- What not to automate
What isn't worth it or isn't prudent, and why.
- With which technology
Rules, integration, AI or an agent for each case.
- In what order
Priorities by impact and effort.
- What impact it can have
The hours, errors and cycle times you can expect to recover.
- What risks exist
And how they are contained: permissions, validations, oversight.
The document is yours. You can execute the roadmap with us or with whoever you prefer.
With judgment or without
Scattered tools and invisible risk
Everyone uses whatever they have heard of, with their personal account, putting client information where it shouldn't be. Plenty of noise and little measurable result.
A few cases, well chosen
Two or three processes with real impact, tools suited to the sector, clear rules on which data can be used and a team that knows what it is doing.
How we do it
- 1
Analysis
Processes, systems, data and current use of AI in the company. What is done by hand and what is being done without control.
- 2
Prioritized use cases
What to automate, what to integrate, where AI adds value and what is better left as it is. With estimated impact and effort.
- 3
Security and governance
Which data can be used and with which tools, who approves what and what gets logged. A written protocol.
- 4
Roadmap and training
Order of execution, measurement criteria and team training on the tools they will actually use.
Frequently asked questions
What do I take away from the assessment?
A document with the map of your processes, the use cases prioritized by impact and effort, the risks detected, the data-use protocol and a roadmap with the order of execution. It's yours: you can execute it with us or with whoever you want.
Does the assessment have a cost?
Yes. The first meeting is free: it's for getting to know each other and seeing whether there is room for improvement. The full assessment is analysis work with your team plus a written report, and it is quoted according to the size of the company and the processes to be reviewed.
Is this ChatGPT training?
No. Training is one part, not the goal. The work is deciding which of your company's processes benefit from AI, which don't, with what safeguards and in what order. Training the team without deciding that first is of little use.
Is AI safe for my company?
The risk isn't in AI, but in how it is used: personal accounts, client data in public tools or models that train on your information. With the right tools and clear rules, the risk can be contained.
What data should I not put into an AI tool?
As a general rule: personally identifiable data, information under professional secrecy, unpublished contracts and intellectual property, and credentials. For those cases we use private tools or anonymize the data beforehand. Exactly what applies depends on your sector and on the requirements of each project.
Do I need to be a large company?
No. We mostly work with small and mid-sized businesses, which tend to have very clear and very manual processes: exactly where a short assessment delivers quick results.
And what if the conclusion is that I don't need AI?
We'll tell you. The assessment may conclude that the right move is to automate with rules, connect two systems, change the process or do nothing for the moment. That is a legitimate conclusion and we write it up like any other.