HOW TO SELECT AN ASSESSMENT CENTRE PROVIDER: A CHECKLIST FOR DECISION MAKERS

In most organisations the assessment centre decision is taken in two stages: first on the method, then on the provider. The first decision has become increasingly standard; the contribution of behaviour-based assessment to selection and promotion decisions is now widely accepted. The difficult decision is the second one. Among services carrying the same name in the market there are quality differences that directly affect the validity of results yet are not easily visible at the proposal stage. A wrong provider choice does not only produce budget loss; it creates layered consequences appointment decisions based on invalid measurement, a perception of unfair elimination, candidate experience issues that reflect on the employer brand, and a method that cannot be put back on the agenda internally for years. This article addresses the factors that make provider selection decisive, the evidence of quality that can be requested, the questions to ask during the proposal and interview process, and the set-up of the working relationship after selection connecting each to the decision practice of HR teams. The aim is to move the selection decision away from impression-based preference and towards an evaluation process run against defined criteria.

Why Provider Selection Is Decisive, and Defining the Need

The assessment centre is a service with a standard name but a non-standardised content. The same term can describe a mature practice run with simulations designed specifically for the organisation, a trained and supervised assessor team, and evidence-based integration sessions; it can equally describe a superficial application run with off-the-shelf exercise sets, observers assigned at short notice, and scoring based on averaged impressions. From the participant's perspective the two look similar: the venue is the same, the exercise flow is comparable, the report format is orderly. The difference lies in the design and scoring layer, and the validity of the outcome is determined precisely in that layer.

This visibility problem is the fundamental difficulty in selection. In many services quality can be tested during use; in an assessment centre the real indicator of quality is the accuracy of decisions emerging over the long term. This delay produces two practical consequences for the procurement process: first, quality must be evaluated at the proposal stage through indirect indicators; second, a selection decision based only on price and a reference list does not cover the risk. Our field observation is that in organisations which did not obtain the expected benefit from an assessment centre, the problem usually lies not in the method but in the quality of the provider.

The first step of the selection decision is defining the need. A scalable centre for a recruitment funnel and an in-depth application for senior promotion decisions require different expertise profiles; a development centre calls for a different design and a different feedback capability. Our article examining the fundamental principles of the method in detail multiple exercises, multiple observers, behaviour-based measurement sets out the methodological ground on which this article rests.

The second component of defining the need is the scope decision: which layer of the service will be sourced externally and which will remain in-house? At one end of the spectrum is the turnkey model, where design, delivery, assessment and reporting all sit with the external provider. At the other end is the capability transfer model, where the provider establishes the methodology, trains internal assessors, and the organisation becomes able to run its own centre over time. Hybrid models are used between the two ends: external independence for critical senior-level applications, trained internal assessors for high-volume recruitment centres. The right model varies with application volume, the sensitivity of the decisions and the HR maturity of the organisation. Clarifying this before provider meetings makes it possible to compare proposals on the same basis.

The common assumption about provider size should also be questioned. Large-scale firms offer a broad assessor pool, multi-location delivery capacity and a defined process standard; in return, whether the project will actually be run by the senior team should be confirmed separately. Boutique and specialised structures offer work in which senior experts are directly in the room, with high depth of customisation; the point to evaluate in this profile is scaling capacity in high-volume programmes. The selection criterion is not the provider label but the match between the project and the structure; whichever structure's natural strength corresponds to the volume, sensitivity and customisation needs of the project is the priority candidate.

Visible Evidence of Quality: The Checklist

The way to make quality differences visible is to request evidence rather than statements. The main headings of the checklist for decision makers are as follows:

•     Methodological transparency: The provider should be able to explain its process clearly, from competency definition to exercise design and from scoring logic to the integration session, and should be able to demonstrate its alignment with international assessment centre principles.

•     Organisation-specific customisation capacity: The depth to which simulations are adapted to the sector, the organisational culture and the reality of the target role should be questioned; applying the same case set to every client measures general performance rather than role-specific behaviour.

•     Quality of the assessor team: The training background, supervision arrangements and depth of experience of observers, and the names who will actually work on the project, should be clarified.

•     Measurement and validity practice: The provider should be asked how it monitors the accuracy of its own applications — inter-rater consistency, the relationship between results and subsequent performance.

•     Reporting and feedback standard: Sample reports should contain behavioural evidence; the duration, structure and ownership of participant feedback sessions should be defined.

•     Data security and ethical framework: The storage, access to and deletion of participant data, and compliance with personal data protection legislation, should be set out in writing.

•     Verifiable references: References should be requested for work carried out at similar scale, sector and purpose; interviews should cover not only outcomes but also how the collaboration functioned.

None of these headings is disqualifying on its own; read together, they clarify the provider's maturity profile. The basic principle is that evidence outweighs assertion. Stating that the team is experienced is not information; sharing the assessor training curriculum and supervision arrangements is information. Claiming that designs are organisation-specific is not information; showing an anonymised example of a comparable customisation is information.

Reference calls also carry low informational value when they are not structured. Instead of a general satisfaction question, the questions should address how the process worked: how the organisation's feedback during customisation was reflected in the design, whether the assessor team committed to in the proposal remained the same in delivery, whether reports were understandable to business unit managers, whether feedback sessions were completed within the planned timeframe, and how post-application objections or queries were managed. These questions surface the delivery quality that does not appear in the contract text.

The component where quality differences are most concrete is the integration session, and it should be addressed separately in interviews. In superficial applications this session is reduced to a short procedure in which scores are averaged. In mature applications it is the most labour-intensive stage of the process, in which observation evidence is discussed participant by participant, the observations of different assessors are compared, and the final profile is written through joint evaluation. The time allocated per participant in this session is among the plainest indicators of the difference between the two levels of practice and should be asked openly at the proposal stage. A comparable indicator is feedback practice: who delivers results to the participant, at what depth and over what duration directly reflects the provider's delivery standard.

Candidate experience is a further heading to evaluate in recruitment-purposed centres. The assessment day is the point of most intensive contact between the candidate and the organisation, and the provider's communication style in the room is perceived as the organisation's employer brand. The provider's candidate information templates, the design of waiting and transition moments, and its practice for communicating negative outcomes are distinguishing topics to raise in interviews.

The technology component has also become a standard part of proposals and requires its own evaluation criteria. Digital platforms, online simulations, game-based tools and AI-supported analyses provide scale, standardisation and candidate experience when used correctly. The questions the decision maker should ask are these: how has the link between the behaviour the tool measures and the competency the target role requires been established; can the evidence underlying automatically generated scores be explained; is the technology positioned instead of or alongside the trained assessor? In an appropriate configuration, technology enriches data collection and scales the process; final decision responsibility remains with the assessor who reads evidence in context.

Assessment Center: Proposal Evaluation and Interview Questions

The instrument that turns the checklist into a workable evaluation is the set of interview questions. The first set tests design depth: which competencies are defined for this role and through what analysis, in how many exercises is a single competency observed, how are exercises adapted to our sector reality? The concreteness of the answers demonstrates the provider's design practice quickly. The second set tests delivery discipline: how many participants per assessor, how does the integration session run and how long does it take, by what method is inter-rater consistency ensured? The third set tests what happens afterwards: can a sample report be shared, who delivers participant feedback and in what structure, what support is provided for integrating results into our HR systems?

The warning signs in the interview process should also be defined. A proposal that quotes a price without listening to the scope is working from a ready-made product rather than the need. A design proposing to measure every competency in a single exercise does not meet the multiple-observation principle of the method. A practice that produces results by averaging scores instead of running an integration session reduces the assessment centre to a test application. An approach that produces no content in response to candidate experience and feedback questions creates risk in recruitment-purposed programmes. Answers that acknowledge no limitation of the method should also be evaluated carefully; mature practitioners communicate the limits of the method openly.

The productivity of interviews depends on structuring the request for proposal correctly. In practice a workable RFP document contains the following: the definition of the target role and level, the estimated number of participants and the delivery calendar, the purpose of the application (recruitment, promotion, development), the existing competency model if any, the delivery location and expectations regarding online components, the scope of reporting and feedback, and data processing conditions. Proposals obtained without sharing this information are standard package prices and carry no comparative value; a price given before scope is clear will be revised at a later stage. The same document should define the outputs requested from providers: a sample report, an introduction to exercise types, CVs of the assessors who will work on the project, a reference list, and the price broken down into unit items.

Price comparison is the stage where errors are most frequent, because proposals often do not describe the same scope. The sound route to comparison is to request the price broken into unit items: design and customisation effort, number of assessor days and seniority of the team, observation intensity per participant, report depth and the number of feedback sessions. Once this breakdown is made, it becomes visible that the price difference between proposals is often not a price advantage but a reduction in scope. The decision maker's question is not which proposal has the lower budget, but which proposal offers evidential strength appropriate to the risk carried by the decision.

A discipline can also be proposed for the selection process itself: the decision should not be left to a single meeting or a single function. Including a manager from the business unit that will use the results, alongside HR, in meetings with shortlisted providers ensures that questions are asked from both a methodological and an operational perspective. Where possible, providers should be asked to share an anonymised case account or the design of a sample integration session; the quality of the service can be evaluated through evidence of this kind.

After Selection: Setting Up the Collaboration and Measurement

Choosing the right provider is the first half of the process; the second half is setting up the collaboration. The first principle is a pilot: rather than moving to a comprehensive programme in a single step, a pilot run with a limited participant group tests the provider's delivery quality in practice, allows organisation-specific adjustments, and produces evidence for internal stakeholders. The second principle is contract clarity: the approval process for the competency set, the scope of exercise customisation, commitment to the assessor team at the level of named individuals, report delivery standards, data retention conditions and intellectual property boundaries should all be in writing. The third principle is internal team alignment: the process owner on the organisation's side should be defined and, where possible, internal observers should be included; this participation both provides quality oversight and begins the transfer of assessment practice into the organisation.

The delivery side of internal team alignment should also be planned. If internal observers are to be assigned, they should have completed assessor training, be paired with a senior external assessor in the first applications, and attend the integration session as observers. The calendar cost of the assignment should be calculated from the outset; an observer's involvement spans more than one working day including preparation, the application day and the integration session. Without this planning, internal participation becomes the first component cancelled during busy periods and the capability transfer objective is not achieved.

The fourth component of the set-up is the measurement framework, which should be defined jointly before the first application. A three-level framework is functional in practice. The delivery quality level covers participant experience feedback, inter-rater consistency data and adherence of the process to plan. The decision contribution level covers the rate at which results are actually used in appointment and development decisions, and decision makers' assessment of report adequacy. The accuracy level covers a joint review, after a reasonable interval, of the relationship between centre results and participants' subsequent performance data. A provider that accepts this framework from the outset has undertaken to make its delivery quality measurable; a candidate that produces no content on measurement has given a signal to be taken into account in the evaluation process.

The long-term structure of the relationship should also be planned at the start. An assessment centre produces its highest value not as a one-off event but as a system component integrated with the competency model, with promotion and succession processes, and with the development architecture. The final question to ask in provider selection is therefore this: does this supplier provide only the application day, or does it also contribute to connecting results to corporate systems, developing internal capacity and calibrating the application over the years? The indicator of a high-quality collaboration is that knowledge is left with the organisation in every application and that the organisation's own assessment maturity rises.

The checklist as a whole can be tied to a single evaluation question: what is purchased when an assessment centre is procured is not a measurement instrument but a structured decision process. An organisation looking through this frame compares proposals not on feature lists and price but on four questions: how disciplined is this team in observing behaviour, how openly does it discuss evidence, how clearly does it communicate the limits of the method, and with what data does it monitor the accuracy of its own decisions? The answers to these questions make evaluable the layer of the service that is not visible at the proposal stage. For the sake of transparency: as an organisation that provides this service, this checklist is also the list of standards to which we hold our own applications, and our recommendation to decision makers is that these questions be asked with equal rigour of all candidates, including us. You can review the details of our approach on our assessment centre application service page.

At Kaan Böke Management Consulting, with 35+ years of corporate experience and 25+ years of C-level leadership perspective, we design and run assessment centre applications with organisation-specific competency design, bespoke simulation development, a trained assessor team and evidence-based integration discipline, delivering strategic and applicable solutions. Contact us to clarify your assessment centre requirement together through this checklist.