Skip to content
RE:ORDER

AI strategy and transformation

We start with a diagnosis: where artificial intelligence (AI) will deliver a measurable effect and which risks need to be managed. On that basis, we build a strategy and rules for using AI, then implement them in the tools your team already uses. We measure the effect against the situation before implementation.

Let's talk

Why many AI implementations fail to deliver

According to a 2025 MIT NANDA report, only about 5% of AI tools built for companies make it from pilot to everyday use, and most organisations see no measurable impact on results. The cause is rarely the model itself. The tools don't fit the processes, don't learn from company data, and nobody measures the effect. Meanwhile, employees at over 90% of the companies surveyed use personal AI tools, although only 40% of companies have bought official licences. That is why we start with the process, the rules and how success will be measured, and choose the tool last.

01

Ways to work with us

  • Getting started with AI

    You know you want to use AI but not where to begin. We review your processes, data and tools, select use cases with clear value and set the order of implementations, with a budget and clear ownership.

    • Process review
    • Priorities
    • Plan and budget
    • Measure of effect

    Outcome: A strategy and an implementation plan in which every use case has a defined measure of success.

  • Bringing order to AI

    AI is already in the company, often on employees' personal accounts. We set the rules: which tools, for what data and tasks, who is responsible and who approves new uses. We assess risk, set up a register of AI uses and build the rules into the policies and procedures you already have. We take into account GDPR, the EU Artificial Intelligence Act (AI Act) and ISO/IEC 42001.

    • AI policy
    • AI use rules
    • Risk assessment
    • Register of AI uses
    • AI Act and ISO 42001

    Outcome: Rules your staff know and follow, and a register of AI uses.

  • AI in a process

    We pick one process and measure it before and after implementation. We implement, for example, a company assistant, an assistant that answers from your documents, process automation, document capture and workflow, automatic meeting notes, a knowledge base or a costing model.

    • Company assistant
    • Document assistant
    • Automation
    • Document workflow
    • Costing models

    Outcome: The project ends when we can show a measured difference.

  • Your team and AI

    A tool works when people know how to use it. We run workshops built around your processes and tools, with exercises on your team's real tasks rather than textbook examples.

    • Tailored workshops
    • Exercises on real tasks
    • Rules for working with AI

  • Post-implementation support

    Implementation does not end on launch day. We support users, update the solution and extend it to further processes.

    • User support
    • Updates
    • Further development

You don't need to change your tools to use AI. We work in: Google Workspace · Microsoft 365 · Claude · ChatGPT · Gemini · Copilot · your own infrastructure. When data cannot leave your organisation, we implement the solution in your infrastructure.

02

Measurable result

We start with one process and its baseline: how long it takes, how much it costs and how many errors occur.

Lead time
Time from receiving a document to its approval
Workload
Team hours spent on the task per month
Unit cost
Cost of handling one case or one document
Error rate
Share of documents that need correcting

AI must not create new security gaps. We design every implementation around information security requirements. Our team holds ISO/IEC 42001 Lead Auditor (AI management systems) and AIGP (AI governance) qualifications, has experience of audits under Poland's KSC Act, and brings academic work on AI in management, digital transformation and technology ethics.

03

Selected work

  • A voice assistant for pharmacy patients: training speech recognition and synthesis models, a device prototype, testing and user research.
  • Process automation in public administration and software for handling processes.
  • Advisory and training on using AI in management and sales, and digital transformation plans for businesses.

04

Contact

Describe your situation. Within two working days we will tell you where to start, what scope of work we suggest and how long it will take.

Let's talkMeet the team

Related areas