Darwinbox People Analytics

Darwinbox People Analytics reads its own HR suite and bills AI by use

Plain-language questions, forecasts and connections to your own AI tool

Six things work in Darwinbox’s favour. The first four help the people who use the tool. The last two help you judge the supplier.

Large workforces spread across regions, where local leaders need their own view

Darwinbox’s customer stories share a pattern. Most of the companies are large, often with essential or contract workers spread across regions or business units. Local leaders there need their own view of the numbers. If that sounds like your company, these stories are worth reading closely. These are Darwinbox’s own stories about its HR platform, and each result belongs to the customer named. If your workforce looks different, ask Darwinbox for a customer like yours. Choose one who uses the analytics, not only the HR suite.

Outside data, forecast scope and contract terms to settle first

Settle these six points before you sign. Each one changes what you pay or what risk you carry.

One limit sits with you, not with Darwinbox. A forecast of who may leave is a judgement about people. GDPR defines profiling as automated processing of personal data to judge things such as performance at work. It also gives people a right not to face a decision made only by automated processing. This applies where the decision has legal or similarly significant effects. If the forecast picks out individuals, ask your data protection lead to check both.

Under the EU AI Act, AI that decides promotion or termination is high-risk, though an exemption can apply. The same list covers AI that monitors how workers perform and behave. The exemption in Article 6(3) never covers a system that profiles people. Whether Darwinbox’s forecast falls under this depends on how you use it, so ask your legal team. Regulation (EU) 2026/1744 applies the duties for these systems from 2 December 2027. In Germany, a works council has a co-determination right over technical devices built to monitor behaviour or performance.

The bill has three parts, and leaving early can cost you

Your bill has three parts. They are the base price, AI usage and set-up work. Leaving has its own cost.

The base price comes by quote. Darwinbox asks you to choose the product areas you need and returns a customised quote.

Your AI bill depends on how often people ask questions and how many employees are scored. Darwinbox’s AI features are available to all clients and billed on actual usage in line with your billing cycle. In AI Pack V2, a plain-language analytics query uses 2 usage units. Attrition prediction uses 1 per employee per quarter. If every employee is scored each quarter, 1,000 employees use 4,000 units a year. The Analytics Agent uses its own Sense Credits. A simple query uses 10 and a complex one uses 15. These figures are indicative, and Darwinbox charges on actual actions. A newer document for the AI Accelerator Pack counts usage per action the agent executes, not per message. The documents count usage in different ways. Ask which one governs your contract and what a unit costs. Darwinbox calls its AI pricing a simple tiered model. It will model the AI cost for your headcount before commitment.

Set-up has its own Implementation Fee, and some later requests are charged as well. Bugs and break-fixes on Darwinbox’s side are fixed without limits on days of work. Other requests are charged by the day of work, at the commercial terms you agreed. During implementation, out-of-pocket expenses are capped at 20% of the Implementation Fee. Once implementation is over, expenses are billed at actual cost. Darwinbox does the set-up, builds the integrations, loads the data and supports testing. You make the business decisions, extract and prepare your data, test and drive user adoption. You must also finish the Darwinbox Academy certification before the project starts. Darwinbox’s own team delivers 47% of implementations, so ask who will deliver yours. You may take application management through an authorised partner under a separate contract.

The Order Form sets the term. The agreement ends on its End Date unless you agreed auto-renewal there. Any renewal must be agreed in writing at least 30 days before. The agreement may end before the Minimum Commitment Term is over. Unless Darwinbox is in breach, you then owe the pro-rated Subscription Fee for the rest of that term. Darwinbox calculates it as if your discount had not applied. After that term, either side may end the agreement on 90 days’ written notice. When the agreement ends, Darwinbox makes a final backup of your data and gives you the backup media. Each side returns or, on request, destroys copies of the data within 30 working days. Each side certifies this in writing.

Ten questions to put to Darwinbox before you sign

These ten questions ask for answers or contract changes before you sign. Each comes with the answer you want.

1. Does the attrition forecast name individual employees, or only teams and groups? Good answer: A written answer in the order form, and a setting for who sees flagged people.
2. Which of your analytics products does our quote cover, and what does each one include? Good answer: A named list of products, with the features of each.
3. What does one AI usage unit cost, and which rate document applies to our contract? Good answer: A price per unit, and the document named in the order form.
4. Will you model our AI cost for our headcount in writing before we sign? Good answer: A worked figure for our headcount, with a point where we can review it.
5. Which of our other systems feed the analytics, and what does each link bring across? Good answer: Named connectors, the fields they carry and how often they refresh.
6. Can our data be used to train AI models under any of your terms, and how do we opt out? Good answer: A written exclusion of training on our data, in the order form.
7. Over what period is the 98% uptime measured, and what do we get if you miss it? Good answer: A stated period and a stated remedy.
8. How does the pro-rated fee work if we leave before the Minimum Commitment Term ends? Good answer: A worked example with our own figures, with the discount treatment spelled out.
9. How do our dashboards, reports and data come out when the contract ends? Good answer: A stated file format and a delivery date in the contract.
10. Which Cortex and Microsoft features are generally available today? Good answer: A dated list, with each feature marked as available or in early access.

Alternatives to Darwinbox

Darwinbox’s alternatives fall in three groups. One group is the analytics inside other HR systems, such as Workday, SAP SuccessFactors, Oracle, UKG, Dayforce, Zoho and Keka. Another is tools that read many systems, such as Visier, One Model, Crunchr, Orgvue and ChartHop. The third is buying nothing.

Start with the analytics in your own HR system if it is not Darwinbox. Workday People Analytics reads Workday HCM and talent data. Oracle Fusion HCM Analytics is made for Oracle Fusion Cloud HCM. UKG pulls its HR, pay and workforce management data into one place. Dayforce offers a library of pre-built dashboards inside Dayforce. Zoho People gathers reports for every HR process in one place. Keka gives a unified view of growth, retention, attendance and sentiment. SAP SuccessFactors sells People Intelligence. If you already run one of these, price its analytics first.

Look at a separate tool when your people data sits in several systems. One Model connects headcount from Workday or SAP SuccessFactors. It also connects labour costs from finance systems and recruiting data from Greenhouse or iCIMS. Crunchr brings workforce data from your existing systems into one view with analytics and governance built in. Orgvue connects HR, finance and operational data. ChartHop connects people and business data into a model of the organisation. Visier maps the systems you run into one workforce model. The trade is a second contract and a second set of connections to look after.

Buying nothing is a real choice. Darwinbox marks its report builder, dashboards and smart search as included. A Darwinbox customer can check whether those cover their questions before adding metered AI usage. Name the decision the analytics will change before you pay for it. Our people analytics guide explains the market. The comparison page sets the researched products side by side.