For business teams

Without verified metrics, everyday work is flying blind.

That is why oneAgent answers your questions in the chat with metrics your BI team has verified and approved. And it alerts you automatically to deviations from your target.

Example: oneAgent alerts purchasing that delivery reliability in Q3 2026 is 91 percent, 4 percentage points below the target of 95 percent. The alert is triggered at more than 2 percentage points below target. As a next step, oneAgent suggests viewing delivery reliability by supplier.

How does work improve in each business team?

You have reliable metrics at hand in the chat right away and are alerted automatically to deviations from your target. This lets you base your everyday decisions on verified metrics. The following examples show what this can look like in different business teams.

Well prepared

Knowing the figures sooner means negotiating from a stronger position.

If a metric deviates from its target, oneAgent tells you before anyone asks. In purchasing, this means you see at once which supplier is not keeping its delivery commitments. You take corrective action before a shortage hits day-to-day business and go into your next supplier meeting with hard figures.

If you receive the right figures too late, or go into a meeting with the wrong ones, your chances of success are slim. The supplier, on the other hand, knows its figures precisely.

In purchasing, you decide when a deviation matters, for example more than 2 percentage points below the delivery reliability target. If a value crosses the threshold, oneAgent alerts you with the verified metric and suggests viewing delivery reliability by supplier.

Example: Purchasing

Example: Without oneAgent, purchasing negotiates in the annual review over figures the supplier knows better. With oneAgent, an alert with the verified metric arrives beforehand: delivery reliability in Q3 2026 is 91 percent, four percentage points below the target of 95 percent. By supplier: supplier A 86 percent, supplier B 95 percent, supplier C 98 percent. The total is weighted by deliveries.

Asked on the spot

You get the figures when you need them, not when the next report comes out.

Standard reports come out on a fixed schedule and only answer the questions they were built for. But if you need to react quickly to a deviation, waiting for the weekly report gets expensive. Ask oneAgent your question when it comes up. You get an immediate answer as a current, verified metric. In production, this means that in the morning meeting you see which line is producing more scrap than planned and can take countermeasures the same day.

Only the weekly report on Friday shows that scrap has risen and flags it. Until then, production keeps running and every day costs material.

You ask in the chat in your own words, for example for yesterday's scrap rate by line. oneAgent works out with you which metric you need exactly. The result is an approved metric with its source and filters. If needed, you dig deeper and ask follow-up questions.

Example: Production

Example: Without oneAgent, scrap rises on Monday, October 12, and the weekly report only shows and flags it on Friday, October 16, without naming the cause. With oneAgent, production asks the question in the chat on Tuesday, October 13. oneAgent clarifies the day and answers with the verified metric for Monday: line 1 2.1 percent, line 2 4.6 percent, line 3 2.6 percent. The total of 3.0 percent is weighted by units produced.

Nothing overlooked

You set the targets, oneAgent monitors them and reports every deviation.

You set a target value for a metric and oneAgent keeps an eye on it. When a relevant deviation occurs, you get a message straight away and simply create a matching task with an owner and a deadline. oneAgent follows up on the deadline until the task is done. In logistics, this means that if the return rate rises above the defined upper limit, you are told automatically and do not have to watch the figure yourself.

Without monitoring, you only learn about a deviation if you happen to look at the right report or someone points it out to you.

You set an upper limit, for example a return rate of 10 percent. If it is exceeded, oneAgent reports it. If needed, you create the matching task right from there. You decide who is responsible and by when. oneAgent then follows up on the deadline until the task is done.

Example: Logistics

Example: Without oneAgent, the return rate has been above the upper limit since October 1 and nobody is informed. It is only noticed if someone happens to look at the report. With oneAgent, oneAgent reports that the return rate in September 2026 is 11.8 percent, 1.8 percentage points above the defined upper limit of 10.0 percent. A task has been created to analyze the reasons for the rise and set measures to reduce it: owner, the Head of Logistics; due October 30; status in progress.

What does this look like in practice?

Imagine you run the online shop and have set a revenue target. This is how you get from the alert to the task.

  1. Alert

    oneAgent reports: the online shop is below its target.

    Online shop revenue, Q3€5.50MVerified metricTarget €6.00M, i.e. €0.50M or 8.3% below
    Suggestions from oneAgent
    • by product group
  2. By region

    Online shop by region

    Deviation from targetVerified metricDACH−€0.50MBenelux+€0.02MNordics−€0.02M
    Suggestions from oneAgent
    • Create task
  3. DACH by product group

    DACH by product group

    Deviation from target, online shop DACHVerified metricGarden furniture−€0.38MHousehold appliances−€0.09MSporting goods−€0.03M
    The gap comes from garden furniture.
    Suggestions from oneAgent
    • View marketing campaigns
  4. Task

    Task

    Identify the cause of the garden furniture gap and define measures for Q4

    Trigger: online shop DACH, garden furniture, €0.38M below target
    Responsibility: oneAgent suggests, you decide
    • Owner, garden furniture product group (selected)
    • Owner, DACH online shop
    Due October 15

From the alert through your follow-up questions to the task, the entire analysis rests on verified metrics.

How quickly can you get started with oneAgent?

In most cases, you can get started quickly and see a tangible benefit. Your BI department and finance team have usually already defined many of the metrics you need. That is your starting point. You can refine and extend later, once oneAgent is up and running.

First: adopt what already exists

The BI team makes the start. It transfers into oneAgent what has already been defined. Many metrics are then available right away, which makes many ad hoc requests unnecessary.

Then: get going and refine

With the company-wide metrics, you can already work productively with oneAgent. You decide afterwards which additional metrics your team needs. This way you increase the benefit of oneAgent step by step.

How your colleagues benefit from oneAgent

BI & ControllingFrom the hundredth report to root-cause analysisFor everyone who defines and approves metrics
ManagementDecide with confidence instead of first clarifying the numbersFor everyone accountable for decisions

Frequently asked questions from business teams

Do I need to know about data or queries to use oneAgent?

No. oneAgent's chat is supported by AI that makes sure oneAgent understands your question, even when you ask it in your own words. In the conversation, oneAgent works out with you exactly what you want to know and answers with the verified metric.

How far can I rely on oneAgent's answers?

You can rely on verified metrics, because the figures are not generated by AI. oneAgent uses AI only to understand your request. The figure comes from the definition your company has approved and is shown with its source and filters.

What happens if the metric I need does not exist yet?

oneAgent tells you so. If you explicitly ask, you get an analysis that is marked as not verified. To have the metric verified in the future, it is best to raise your need with the responsible BI team.

Does oneAgent only answer questions?

No. As an AI assistant, oneAgent takes on work that it carries out on its own: you set targets, oneAgent monitors them and alerts you to relevant deviations. You can then create a task, give it a deadline and assign it to the person responsible, and oneAgent follows it up until it is done.

How do targets and deviation alerts work?

Either plan and target values are already in place, or you set your own. If a deviation exceeds a value you have defined, oneAgent notifies you. You can then, for example, create a task, give it a deadline and assign it to the person responsible.

How does oneAgent differ from ChatGPT?

Both understand questions asked in your own words, but the figure comes from different sources. With oneAgent, it comes from the definition your company has approved. With ChatGPT, it comes from a language model that does not know your definitions. That is why ChatGPT may return wrong figures, even when the data comes from your own company.

How do you get started?

Tell us which figure you need most often in your day-to-day work. In the demo, you see how you could retrieve it instantly and keep an eye on it in the future.

Book a demo
Your contactThomas IngenhorstCo-founder and Managing Director of oneLake GmbH, responsible for oneAgentView on LinkedIn
Book a demoContact