Where to start

Small enough for weeks. Big enough to matter.

We start where your team will notice the difference quickly. Before anything begins, we agree with you on how we will measure success. After six to ten weeks, you will see whether it works, and everything after that builds on it.

Book a call View the process You can start from two directions. Either with a task that keeps getting stuck in operations. For that, this page shows the right use case. Or with the question of which capabilities you will need in the future. Then we start with strategy or AI transformation. Planning your own AI solution? We develop it on BRIDGE

How we choose the entry point

Three questions before we start.

Before we suggest an entry point, we go through three questions with you. We only start once the answer to all three is yes.
  • Not too small

    Will your business notice the difference?

    It has to be about something that slows you down every day, such as long lead times, recurring errors or knowledge that only a few people have. Otherwise even a successful pilot changes little.

  • Not too big

    Will we know within a few weeks whether it works?

    We agree before the start on how we will recognize success, and the data we need is accessible. After six to ten weeks, you know whether it pays off. If it doesn’t, we stop, as agreed up front.

  • Connected

    Can the next step build on it?

    What the pilot creates stays: the knowledge in BRAIN and the capability in BRIDGE ONE. So the next step doesn’t start from zero again.

    •

    Align the selection criteria with Delivery as part of discovery (“Governed AI Workforce” concept: discovery, measurable pilot with a stop rule). Decision by Heiko, 20 Oct: an entry point of the right size instead of “the capability that is missing today”.

    AI use cases

    Where you can start.
  • Technical service Service agent

    Core capability

    Your technicians find the solution in minutes.

    Capability that growsFix faults quickly and reliably.

    Starting point

    Knowledge about the equipment sits in manuals, in tickets and with a few experienced colleagues. New technicians spend a long time searching, and specialists are constantly interrupted.

    • What you get A service agent on BRAIN. Your technicians ask it the way they would ask a colleague, and every answer shows where it comes from: the manual, past faults or documented experience. Runs on BRAIN
    • How we measure success
      • How often the answers are right
      • How long a fault takes to fix
      • How often specialists still have to step in
    • Reference about 96% of the answers at SHW are correct. There, the service agent draws on about 160,000 documents. SHW approval and measurement method
  • Find and keep knowledge Company knowledge

    Enabling capability

    Experience stays in the company, even when people leave.

    Capability that growsKeep knowledge easy to find and pass on experience.

    Starting point

    By 2040, around 13 million people in Germany’s labor force will reach the statutory retirement age. With them goes knowledge that is not written down in any system. And where there is no secure AI inside the company, employees often turn to private tools. Link the source: Destatis, press release N039 of 23 June 2026 (13.3 million by 2040)

    • What you get A secure company AI on BRAIN that knows your documents, wikis and tickets and names the source with every answer. On top of that, BRAIN collects knowledge where it is created anyway, in meetings and solved cases. Nobody has to write anything down just for this, and whoever takes over asks BRAIN instead of spending ages searching. Company AIKnowledge from conversationsCheck status: capture from conversations. Starting format per one-pager: quick assessment, 2–4 weeks, fixed price
    • How we measure success
      • How long onboarding takes
      • How many questions are answered without a follow-up
      • Which knowledge areas are already covered
  • Quotations & sales Respond faster

    Core capability

    Quotations of the same quality, no matter who writes them.

    Capability that growsAnswer inquiries quickly, correctly and consistently.

    Starting point

    Technical inquiries end up with a few specialists. Quotations take time, and their quality depends on who happens to be available.

    • What you get An agent that prepares a draft quotation from the inquiry, specification, customer history and application knowledge. Your sales team reviews it, adds to it and decides. Runs on BRAINCheck status (NW use case, AutoSales concept)
    • How we measure success
      • How long a quotation takes
      • How many turn into orders
      • How much time the specialists get back
  • Quality & evidence Standards and audits

    Enabling capability

    Standards and evidence on demand.

    Capability that growsProve compliance with standards at any time.

    Starting point

    REACH, RoHS, IATF 16949 or MDR: the evidence is scattered across folders and systems. Every audit and every customer inquiry starts with a search.

    • What you get BRAIN organizes data sheets, standards and test reports and answers questions about them with sources. If a piece of evidence is missing or outdated, BRAIN points it out. Runs on BRAINCheck status
    • How we measure success
      • How long it takes to prepare for an audit
      • How quickly customer inquiries are answered
      • How many gaps are found
  • IT operations BRIDGE OPS

    Enabling capability

    Steer your IT in the chat, within clear limits.

    Capability that growsRun systems reliably, even with only a few specialists.

    Starting point

    IT depends on a few people. Joiners and leavers, access rights, certificates and incidents take up time every day. And for many companies, NIS-2 requires management to take responsibility for secure operations. Check the NIS-2 reference: it only applies to entities in scope

    BRIDGE OPS in detail
    • What you get BRIDGE OPS, our DevOps agent. You write in Teams or Slack what you need, and it gets it done in your Microsoft environment, with exactly the permissions you give it. Before every change, you see what would happen. Data stays in your tenantIn use at BRIDGE, a customer is asking for it (Heiko, 21 Oct). Azure only, business hours only. Log export and second approval are contract items in the customer info; still to be shown as working features
    • How we measure success
      • How long routine tasks take, such as a new joiner
      • How early you hear about incidents
      • How much time your IT specialists get back
    • What it may not do It never acts with more permissions than you give it. An approval catalog defines what runs without asking, and on the first day it can only read.

    • No use case yet?

    Then we find it together, in one day.

    In the use case workshop, we collect the ideas from your business units and check each case for benefit, data and effort. At the end, you know where to start and what comes next.

    Ask about the workshop
    1. Collect ideasYour business units bring their cases, and new ones come up in the conversation.
    2. AssessEach case gets a short profile and is ranked by benefit and feasibility.
    3. Set the orderYou get a roadmap in three waves: what works right away, what needs its own logic and what is better done without an agent.

    Format as in the partner documents (“use case workshop / potential analysis”, 1 day), held at SHW on 8 July. Price open.

    Process

    How an entry point works.

    There are three steps, and after each one you decide whether to continue.

  • Discovery · one to two weeks

    We look at the task, your data and your systems. You then get a proposal for the pilot: what it should achieve, how we will measure it and when we will stop.

  • Pilot · six to ten weeks

    A team works in a clearly defined area with real data and real cases. We measure whether the agreed goal is met. The results belong to you.

  • Operation · ongoing

    What works becomes part of everyday work. We run the solution, keep an eye on it and develop it further, for a fixed monthly price.

    •

    Agree durations with Delivery. Wording follows the “Governed AI Workforce” concept (discovery, measurable pilot with stop rule, results belong to the customer).

    What comes next

    One entry point leads to more.

    We set up every entry point as a capability in BRIDGE ONE. That way, the next steps build on what is already there. A second agent uses the same foundation as the first. That is why it is ready sooner and costs less.
  • Phase 1 · BRAIN

    Strengthen one capability

    One area becomes faster and more reliable and depends less on a few individuals. For heads of functions and business units.

  • Phase 2 · BRIDGE ONE

    Steer capabilities

    With Capability Management, it is clear for every capability who owns it, how good it is today and where it should go. That way, the improvement lasts. For business unit heads and the COO.

  • Phase 3 · BRIDGE ONE

    Connect strategy and capabilities

    With Strategy Execution, you see which capability contributes to which strategic thrust and goal. For executive management and those responsible for transformation.

    •      How we charge. We offer discovery and the pilot at a fixed price. Operation is billed monthly, per entry point. The infrastructure runs in your tenant or with us in the EU.

    Decision: Show prices on the website? Figures from the “Capability Development” proposal: Capability Map from €18,500, Capability Development from €39,500, infrastructure operation from €590 per month. Previous proposals (e.g. wave 1 at SHW) are billed by effort: clarify the “fixed price” statement with Delivery.

    Where would you start?

    Tell us where things get stuck in your day-to-day work. We look at your data and suggest an entry point where we can show whether it works.

    Book a call AI transformation