Free RFP template
Performance management software RFP template and vendor evaluation scorecard
Fifty RFP questions across ten categories, each with what a strong answer looks like, a weighted scorecard for comparing three vendors, worked scoring examples and a seven-step method for getting from a long list to a decision.
By the CLEAR Talent team
A request for proposal is the point at which a software evaluation stops being a series of sales conversations and becomes a comparison. Every vendor answers the same questions, in writing, by the same date, and the answers are scored the same way by the same people. Done well, it saves weeks of demos with vendors who were never going to fit.
Done badly, it is a 300-row spreadsheet that every vendor answers “yes” to. The template below is deliberately shorter: fifty questions in ten categories — reviews, goals and OKRs, 360° feedback, calibration, analytics, AI governance, security and GDPR, integrations, implementation and support, and pricing — each with a note on what separates a strong answer from a hollow one. Delete the questions that do not matter to you, mark the ones that are must-haves, and score the answers on the weighted scorecard further down.
Before you send it
How to use this RFP template
Most performance management RFPs fail in one of two ways. Either they ask so many questions that the answers cannot be read properly, or they ask questions every vendor can answer “yes” to, so the scores come out level and the decision is made on the demo anyway. The questions below are written to avoid both: each asks how something works, not whether it exists, and each comes with what a strong answer shows, so the panel knows what it is scoring before the responses arrive.
Two decisions come before you send anything. First, mark your must-haves — the handful of requirements a vendor is out without, such as a data processing agreement before contract or a sign-in method your IT policy requires. Must-haves are pass or fail and sit outside the weighted score, so a strong showing elsewhere cannot carry a vendor over a requirement it does not meet. Second, agree the weights on the scorecard. Weights set after the responses arrive tend to follow the panel’s favourite.
Ask for it in writing
Facts you will rely on later
Hosting region, the data processing agreement, the subprocessor list, any security certifications and their scope, and how a subject access request is handled. These belong in the contract file, not in a demo recording.
Anything with a price attached
What counts as a billable user, which modules are in the quote, setup fees, minimum term and the renewal terms. A written number is the only one worth comparing.
What exists today
Named integrations and features that ship now, kept separate from the roadmap. A roadmap item is a reason to revisit later, not a reason to buy now.
Leave it for the demo
How the review feels to complete
A written “yes, reviews are configurable” tells you nothing about whether managers will finish them. Watch a manager’s review being written, ideally on a phone.
The calibration meeting
Ask to see a calibration session with your own rating scale: how ratings are compared across managers, and what is recorded when one changes.
Reports a manager would actually open
Completion, rating distribution and goal progress for one team, generated live rather than shown as a screenshot.
If you have not built a long list yet, start from independent comparisons and the vendors your peers use, then screen out anyone who fails a must-have on information they already publish — pricing, hosting, integrations. Sending the RFP to more than five vendors usually means scoring answers you will never act on.
Build the long list from the buyer’s comparison of performance management software
The questions
Performance management software RFP questions
Fifty questions in ten categories. Send the ones that apply to you, keep the numbering so answers can be matched to the scorecard, and ask vendors to answer in the response column rather than in a separate document. A one-word “yes” scores as a partial answer at best.
The note under each question is for your panel, not for the vendor: it says what separates a strong answer from a hollow one. Remove the notes before you send the file if you would rather vendors did not see them.
Performance reviews
How the review cycle runs, from the form to the final sign-off.
Which review types can we run — annual, mid-year, probation, project — and can each have its own form, rating scale and participants?
A strong answer: HR configures each review type without vendor services, and the vendor shows it in the demo using your own form rather than a sample.
How are deadlines and reminders handled for each stage: self-assessment, manager review, calibration and sign-off?
A strong answer: Each stage has its own deadline, and reminders go automatically to whoever has not yet submitted, without anyone exporting a list and chasing by email.
When a manager writes a review, can they see the employee’s goals, feedback and previous reviews on the same screen?
A strong answer: The evidence sits beside the review form in the demo, not in a separate module the manager has to remember to open.
What happens when a rating is changed after it has been submitted — who can change it, and is the change recorded?
A strong answer: Every change is logged with who made it, what changed and when, and a closed review period cannot be quietly reopened.
Can employees and managers complete their parts of a review on a phone?
A strong answer: Shown on a real phone in the demo, with the same review form, rather than described or shown as a shrunk desktop screen.
Goals and OKRs
How goals are set, aligned, tracked and carried into the review.
Can goals cascade from company objectives to teams and individuals, and can an employee see how their goal connects to the level above?
A strong answer: Alignment is visible to the employee, not only in an HR report.
Can OKRs, KPIs and development goals sit in the same plan, or does each need a separate module?
A strong answer: One plan holds the goal types you use, and the answer says plainly whether any type is a separately priced module.
How is progress recorded — check-ins, percentage complete, key-result values — and how often can it be updated?
A strong answer: Progress is updated as part of regular check-ins rather than reconstructed at review time.
If a goal changes mid-cycle, is the original version kept?
A strong answer: Both versions and the date of the change are on the record, so the review judges the goal that applied.
How do goal results feed the performance review?
A strong answer: The answer says exactly what happens — scored automatically, shown as reference or not linked at all — rather than “seamlessly integrated”.
360° feedback
Who gives feedback, how it is protected and what the employee receives.
Who chooses the raters, and can a manager or HR approve or change the list?
A strong answer: Nomination and approval are both configurable, and the approval step is shown, not described.
How is rater anonymity protected? Is there a minimum number of responses per group before results are shown?
A strong answer: A minimum-response threshold is enforced by the system, and the vendor explains what a manager sees when a group falls below it.
Are 360° questions tied to the same competencies used in the performance review, or are they a separate survey?
A strong answer: The same competencies and scale, so 360° results can be read alongside the review.
Can a 360° run on its own, outside the review cycle — for a development programme, for example?
A strong answer: Yes, with its own timing and participants, without opening a review cycle.
What does the employee receive at the end, and does it lead to a development plan?
A strong answer: A report the vendor shows you, with a clear route from the results to development actions rather than a list of comments.
Calibration
How ratings are compared across managers, and what is recorded when they move.
How does a calibration session work in the product? What does the room see?
A strong answer: Rating distributions by manager and team, with outliers visible, shown live using a rating scale like yours.
Can the panel open the evidence behind a rating — goals, competency ratings, feedback — during the session?
A strong answer: The evidence is one click from the rating, not in a separate export prepared beforehand.
When calibration changes a rating, is a reason recorded, and who can see it afterwards?
A strong answer: A reason is captured at the moment of change and kept on the record, so it can be explained months later.
Is forced distribution available, and can it be switched off?
A strong answer: Optional and configurable, rather than built into the process by default.
Can calibration run across departments or locations that share a rating scale?
A strong answer: Yes, with the scope of each session set by HR rather than fixed to the reporting line.
Analytics and reporting
What you can see, who can see it and how it leaves the system.
Which reports are standard — completion, rating distribution, goal progress, competency gaps — and at which levels: individual, team, department, organisation?
A strong answer: A list of the standard reports, with the levels each one covers, shown in the demo.
Can we export raw data and formatted reports — CSV, Excel, PDF — on the plan you are quoting?
A strong answer: Exports are included in the plan quoted, not a paid add-on, and the formats are listed.
How is report access scoped? Can a manager see only their own team’s data?
A strong answer: Reports respect the same role-based access as the rest of the system, shown by signing in as a manager.
Can we track the same measure across cycles — for example, a competency rated on the same scale year on year?
A strong answer: Trends across cycles without rebuilding the data in a spreadsheet each year.
Can HR build or filter reports without vendor services?
A strong answer: Yes, or a clear statement of what custom reporting costs and how long it takes.
AI governance
Where AI is used on employee data, and who stays in control of it.
Which features use AI, and can each be switched off — for a cycle, a team or a module?
A strong answer: A complete list of AI features, each with its own off switch rather than one setting for everything.
Is AI-generated content labelled, and can a person edit or reject it before it is saved or shared?
A strong answer: Every AI output is labelled and editable, and nothing reaches an employee without a person approving it.
Can AI set a rating or make a decision about an employee without a person approving it?
A strong answer: A plain no, with an explanation of where the final decision is made and by whom.
Is our employees’ data used to train any model — for us, for other customers or by the model provider?
A strong answer: A written answer that covers all three cases separately, matched by the terms in the contract or data processing agreement.
Which AI model providers process our data, and where? Are they on your subprocessor list?
A strong answer: Named providers that appear on the subprocessor list you receive with the DPA.
Security and GDPR
The questions your DPO and IT security team will ask, answered before the contract.
Where will our data be hosted, and will you confirm the region in writing in the contract or DPA?
A strong answer: A named hosting provider and region, confirmed in the contract rather than taken from a web page.
Will you provide your data processing agreement and current subprocessor list before we sign?
A strong answer: Yes, early enough for your DPO to review them alongside the RFP responses.
How do you handle a subject access request and a right-to-erasure request for performance data?
A strong answer: Built-in export and erasure workflows that cover reviews, ratings and feedback, shown or documented, rather than a support ticket.
Which independent security certifications or attestations do you hold, and can we see the current report?
A strong answer: The certificate, its scope and its date — or a plain statement that there is none, with the controls described instead.
How is access controlled — roles, ownership, single sign-on, multi-factor authentication — and what is logged?
A strong answer: The sign-in methods supported today, including whether SAML single sign-on and SCIM provisioning are available, and what the audit log records.
Integrations and data
How people data gets in, stays current and gets out again.
Which HRIS do you connect to with a prebuilt connector today? What data syncs, in which direction and how often?
A strong answer: Named connectors that ship now, kept separate from anything on the roadmap.
If there is no connector for our HRIS, how do people and organisation structure get in, and how do they stay current?
A strong answer: A described process — bulk import, API or both — with who does the work and how often.
Is there an API, what does it cover and is it included in the plan you are quoting?
A strong answer: Documentation you can share with your IT team, and a clear statement of any extra cost.
Can we bring in historical goals and ratings, and in what format?
A strong answer: A supported import format and a view on how much history is worth migrating.
If we leave, how do we get all our data out, in what format and at what cost?
A strong answer: A full export in standard formats, at no extra charge, stated before you sign rather than at renewal.
Implementation and support
What happens between signing and the first cycle, and after it.
What does implementation involve, who does what, and how long has it taken for organisations of our size?
A strong answer: A plan with tasks owned by both sides, scoped to your organisation rather than a standard brochure timeline.
Is there a setup or onboarding fee, and what does it include?
A strong answer: A figure or a basis for one, and a list of what it covers.
Which support channels and response times come with the plan you are quoting, and are they in the contract?
A strong answer: Support terms written into the agreement, not only described on a website.
In which languages are the employee-facing screens and support available?
A strong answer: A list covering every language your workforce needs, with any differences between the product and support stated.
Can we speak to a customer of a similar size or sector?
A strong answer: A reference call arranged with a current customer, not only a case study.
Pricing and commercial terms
The figures you need to compare vendors on the same basis.
What is the total annual cost for our headcount, including every module in this RFP, setup and support?
A strong answer: One total with line items, so it can be compared directly with the other responses.
What counts as a billable user — every employee, only people in a review, or administrators only?
A strong answer: A definition you can apply to your own headcount without a follow-up call.
What are the minimum term, the notice period and any limit on price increases at renewal?
A strong answer: All three stated in writing.
Which features shown in your demo are in the plan you are quoting, and which are add-ons or higher tiers?
A strong answer: A feature-by-feature answer that matches what the panel saw.
Can we run a free trial or pilot with our own managers before signing?
A strong answer: A trial or pilot long enough to run real scenarios with your managers, with its terms stated.
Nothing here is gated. The spreadsheet holds all fifty questions with the notes on strong answers, a must-have column and a response column for each of three vendors. To use it in Google Sheets, open a blank sheet and choose File, then Import, then Upload; Excel and Numbers open the file directly.
The vendor evaluation scorecard template
Template
Vendor evaluation scorecard
Fill in the evaluation details once, then one row per RFP category. Each vendor gets a score from 0 to 5 per category; multiply by the weight and divide by 100, and each vendor’s weighted column adds up to a score out of 5. Must-haves are checked separately: a vendor that fails one is out, whatever it scores.
| Field | What to write |
|---|---|
| Evaluation details | |
| Organisation and scope | Headcount, countries and which populations the new system will cover in its first cycle. |
| Evaluation owner | The person who runs the process and keeps the scorecard. |
| Scoring panel | Who scores: usually HR, someone from IT or security, someone from finance and at least one line manager who will use the system. |
| Scoring scale | 0 not offered; 1 only by workaround; 2 partly meets the need; 3 meets it; 4 meets it well, with evidence; 5 meets it well and was shown working with your own scenario. |
| Must-have requirements | The pass-or-fail requirements, kept to a handful. List them here and record each vendor’s pass or fail before scoring anything else. |
| Vendors scored | The three vendors on the shortlist. |
| Decision date | When the panel will decide, so the demos and reference calls fit before it. |
| For each category (up to 10) | |
| Category | One row per RFP category. Use the ten from the question bank, or merge the ones that matter little to you. |
| Weight (%) | How much the category matters to this decision. The weights add up to 100. Agree them before anyone reads a response. |
| Vendor A score (0–5) | The panel’s agreed score, from the written answers and the demo. |
| Vendor A weighted | Score × weight ÷ 100. The column adds up to a score out of 5. |
| Vendor B score (0–5) | As for vendor A. |
| Vendor B weighted | As for vendor A. |
| Vendor C score (0–5) | As for vendor A. |
| Vendor C weighted | As for vendor A. |
| Evidence behind the scores | Which answers, demo moments or reference calls the scores rest on. A score with no evidence is an opinion. |
As a spreadsheet
One row per category. Copy the grid below and paste it into the first cell of a spreadsheet, or download the CSV.
| # | Category | Weight (%) | Vendor A score (0–5) | Vendor A weighted | Vendor B score (0–5) | Vendor B weighted | Vendor C score (0–5) | Vendor C weighted | Evidence behind the scores |
|---|---|---|---|---|---|---|---|---|---|
| 1 | |||||||||
| 2 | |||||||||
| 3 | |||||||||
| 4 | |||||||||
| 5 | |||||||||
| 6 | |||||||||
| 7 | |||||||||
| 8 | |||||||||
| 9 | |||||||||
| 10 |
Show the plain-text template
VENDOR EVALUATION SCORECARD Organisation and scope: Evaluation owner: Scoring panel: Scoring scale: Must-have requirements: Vendors scored: Decision date: CATEGORY 1 Category: Weight (%): Vendor A score (0–5): Vendor A weighted: Vendor B score (0–5): Vendor B weighted: Vendor C score (0–5): Vendor C weighted: Evidence behind the scores: CATEGORY 2 Category: Weight (%): Vendor A score (0–5): Vendor A weighted: Vendor B score (0–5): Vendor B weighted: Vendor C score (0–5): Vendor C weighted: Evidence behind the scores: CATEGORY 3 Category: Weight (%): Vendor A score (0–5): Vendor A weighted: Vendor B score (0–5): Vendor B weighted: Vendor C score (0–5): Vendor C weighted: Evidence behind the scores: CATEGORY 4 Category: Weight (%): Vendor A score (0–5): Vendor A weighted: Vendor B score (0–5): Vendor B weighted: Vendor C score (0–5): Vendor C weighted: Evidence behind the scores: CATEGORY 5 Category: Weight (%): Vendor A score (0–5): Vendor A weighted: Vendor B score (0–5): Vendor B weighted: Vendor C score (0–5): Vendor C weighted: Evidence behind the scores: CATEGORY 6 Category: Weight (%): Vendor A score (0–5): Vendor A weighted: Vendor B score (0–5): Vendor B weighted: Vendor C score (0–5): Vendor C weighted: Evidence behind the scores: CATEGORY 7 Category: Weight (%): Vendor A score (0–5): Vendor A weighted: Vendor B score (0–5): Vendor B weighted: Vendor C score (0–5): Vendor C weighted: Evidence behind the scores: CATEGORY 8 Category: Weight (%): Vendor A score (0–5): Vendor A weighted: Vendor B score (0–5): Vendor B weighted: Vendor C score (0–5): Vendor C weighted: Evidence behind the scores: CATEGORY 9 Category: Weight (%): Vendor A score (0–5): Vendor A weighted: Vendor B score (0–5): Vendor B weighted: Vendor C score (0–5): Vendor C weighted: Evidence behind the scores: CATEGORY 10 Category: Weight (%): Vendor A score (0–5): Vendor A weighted: Vendor B score (0–5): Vendor B weighted: Vendor C score (0–5): Vendor C weighted: Evidence behind the scores:
Nothing here is gated: the spreadsheet and the text file hold exactly the fields shown above, with ten blank category rows. Add a total row that sums each weighted column. The CSV opens in Excel or Numbers, or in Google Sheets through File, then Import.
How to shortlist three performance management vendors
Seven steps from a long list to a decision. The order matters: settle the must-haves and the weights before any vendor has answered, and keep the demos for the three vendors whose written answers earned one.
Write down the problem and the must-haves
Before talking to any vendor, agree in a paragraph what the new system has to fix — the chasing, the calibration, the missing evidence behind ratings — and list the handful of requirements a vendor is out without. Everything else is scored, not required.
Screen a long list down to five
Start with six to eight vendors from independent comparisons and from peers who run a similar process. Remove anyone who fails a must-have on information they already publish — pricing, hosting, integrations, sign-in — and send the RFP to the five that remain.
Agree the weights before the answers arrive
Set the weight of each category on the scorecard while nobody has a favourite. Weights reflect what matters to this decision: an organisation replacing a spreadsheet weights reviews and goals; a regulated one weights security and calibration.
Send the RFP with a deadline and a format
Give vendors about two weeks, a named contact for questions and the CSV to answer in, one answer per question. Tell them a one-word “yes” scores as partial, so the answers you get back explain how rather than whether.
Score independently, then reconcile
Each panel member scores the written answers alone, on the same 0–5 scale. Then meet, discuss only the scores that differ by more than a point, and agree one score per vendor and category. Record the evidence for each in the scorecard.
Cut to three and run scripted demos
Invite the top three and send them the same script: your rating scale, a review in which a rating changes after submission, a calibration session, a manager completing a review on a phone, and a subject access request. Rescore the categories the demo tested.
Check references and terms, then decide
Take a reference call with each finalist and get the contract terms and total cost in writing. If the top two finish within a fraction of a point, the scorecard has not decided; settle it on the highest-weighted category where they differ, or a short pilot with your own managers.
Vendor scorecard examples: three evaluations
Three scorecards for three different buyers, each comparing three vendors on the 0–5 scale. The first has a clear winner, the second shows why must-haves sit outside the score, and the third shows what to do when the scores are too close to decide.
These organisations, vendors and scores were written for this page to show how the scorecard works. Vendors A, B and C are not real products, and the examples describe no real organisation or any CLEAR Talent customer.
Replacing spreadsheets
Example 1: a 180-person software company buying its first system
Reviews have run in a shared spreadsheet for four years, and the company has just introduced OKRs. The panel weights reviews and goals most heavily and puts price ahead of integrations, because the HR team is one person. Vendor A wins clearly and keeps its lead through the trial.
- Scoring panel
- HR manager, finance director, head of engineering, two team leads
- Must-have requirements
- Reviews and OKRs in one system; published pricing within budget; reviews completable on a phone
- Decision date
- Eight weeks after the RFP went out
| Category | Weight (%) | Vendor A score (0–5) | Vendor A weighted | Vendor B score (0–5) | Vendor B weighted | Vendor C score (0–5) | Vendor C weighted | Evidence behind the scores |
|---|---|---|---|---|---|---|---|---|
| Performance reviews | 20 | 4 | 0.80 | 3 | 0.60 | 4 | 0.80 | A and C configured the panel’s own form in the demo; B used its sample form. |
| Goals and OKRs | 20 | 4 | 0.80 | 4 | 0.80 | 4 | 0.80 | All three cascade objectives; level on the evidence seen. |
| 360° feedback | 10 | 4 | 0.40 | 3 | 0.30 | 3 | 0.30 | A ties 360° questions to the review competencies; B and C run it as a separate survey. |
| Calibration | 5 | 3 | 0.15 | 2 | 0.10 | 4 | 0.20 | Low weight: two managers, no pay link yet. C showed the strongest calibration view. |
| Analytics and reporting | 10 | 3 | 0.30 | 4 | 0.40 | 4 | 0.40 | A’s standard reports cover the need; B and C let HR build their own. |
| AI governance | 5 | 3 | 0.15 | 2 | 0.10 | 3 | 0.15 | B could not say in writing whether data is used to train models. |
| Security and GDPR | 10 | 4 | 0.40 | 4 | 0.40 | 4 | 0.40 | All three provided a DPA and subprocessor list on request. |
| Integrations and data | 5 | 3 | 0.15 | 4 | 0.20 | 2 | 0.10 | No HRIS connector needed yet; A and B import by CSV, C only through services. |
| Implementation and support | 5 | 4 | 0.20 | 3 | 0.15 | 3 | 0.15 | A’s plan named the work on both sides. |
| Pricing and commercial terms | 10 | 4 | 0.40 | 3 | 0.30 | 2 | 0.20 | A’s total came in under budget; C’s included a setup fee above it. |
| Weighted total (out of 5) | 100 | — | 3.75 | — | 3.35 | — | 3.50 | A leads C by 0.25 and B by 0.40. The panel confirmed A in a one-month trial with two team leads. |
Must-have fails
Example 2: a 600-person healthcare provider where the top scorer is ruled out
Ratings feed pay progression, reviews run in two countries, and the DPO sits on the panel. Security, calibration and AI governance carry most of the weight. Vendor B scores highest, but cannot provide its data processing agreement until after contract signature — a must-have — so it is out, and the decision is between A and C.
- Scoring panel
- Head of HR, DPO, IT security lead, finance business partner, two ward managers
- Must-have requirements
- DPA and subprocessor list before contract; hosting region confirmed in writing; no AI decision without human approval
- Decision date
- Twelve weeks after the RFP went out, to allow DPO review
| Category | Weight (%) | Vendor A score (0–5) | Vendor A weighted | Vendor B score (0–5) | Vendor B weighted | Vendor C score (0–5) | Vendor C weighted | Evidence behind the scores |
|---|---|---|---|---|---|---|---|---|
| Performance reviews | 15 | 4 | 0.60 | 4 | 0.60 | 3 | 0.45 | A and B lock closed periods and log rating changes; C logs changes but allows reopening. |
| Goals and OKRs | 5 | 3 | 0.15 | 4 | 0.20 | 3 | 0.15 | Low weight: goals are set locally and matter less than ratings here. |
| 360° feedback | 10 | 4 | 0.40 | 5 | 0.50 | 3 | 0.30 | B enforces minimum-rater thresholds per group and showed the employee report. |
| Calibration | 15 | 4 | 0.60 | 5 | 0.75 | 3 | 0.45 | B and A showed reasons recorded against every changed rating; C’s reason field is optional. |
| Analytics and reporting | 10 | 3 | 0.30 | 4 | 0.40 | 4 | 0.40 | B and C report across both countries in one view. |
| AI governance | 10 | 4 | 0.40 | 4 | 0.40 | 2 | 0.20 | A and B label AI output and allow it to be switched off per module; C has one global switch. |
| Security and GDPR | 20 | 4 | 0.80 | 3 | 0.60 | 4 | 0.80 | B would share its DPA only after signature, which fails the must-have; its controls scored 3. |
| Integrations and data | 5 | 3 | 0.15 | 4 | 0.20 | 4 | 0.20 | B and C offer an API on every plan; A’s import covers the need. |
| Implementation and support | 5 | 4 | 0.20 | 3 | 0.15 | 4 | 0.20 | A and C named a lead for the rollout; B quoted a standard timeline. |
| Pricing and commercial terms | 5 | 4 | 0.20 | 2 | 0.10 | 3 | 0.15 | B’s quote put calibration in a higher tier. |
| Weighted total (out of 5) | 100 | — | 3.80 | — | 3.90 | — | 3.30 | B scores highest but fails a must-have and is excluded. A is selected, 0.50 ahead of C. |
Too close to call
Example 3: a 350-person professional services firm with two vendors level
The firm runs a development-led process: 360° feedback every year, goals reviewed quarterly and promotions decided by a panel. The weights are spread evenly, and A and B finish 0.05 apart — less than one panel member changing one score. The scorecard narrows the decision; it does not make it.
- Scoring panel
- People director, L&D lead, operations director, two practice managers
- Must-have requirements
- 360° feedback on the firm’s competency framework; export of all data in standard formats on exit
- Decision date
- Ten weeks after the RFP went out, including a second demo
| Category | Weight (%) | Vendor A score (0–5) | Vendor A weighted | Vendor B score (0–5) | Vendor B weighted | Vendor C score (0–5) | Vendor C weighted | Evidence behind the scores |
|---|---|---|---|---|---|---|---|---|
| Performance reviews | 15 | 4 | 0.60 | 4 | 0.60 | 3 | 0.45 | A and B both ran the firm’s promotion-review form in the demo. |
| Goals and OKRs | 15 | 4 | 0.60 | 3 | 0.45 | 4 | 0.60 | A and C carry quarterly goal results into the review; B shows them as reference only. |
| 360° feedback | 15 | 3 | 0.45 | 4 | 0.60 | 3 | 0.45 | B’s 360° report leads straight into a development plan. |
| Calibration | 10 | 4 | 0.40 | 4 | 0.40 | 2 | 0.20 | C has no calibration view; the promotion panel would work from exports. |
| Analytics and reporting | 10 | 3 | 0.30 | 4 | 0.40 | 3 | 0.30 | B tracks competency ratings across cycles on the same scale. |
| AI governance | 5 | 3 | 0.15 | 3 | 0.15 | 4 | 0.20 | All label AI output; C documented its model providers most clearly. |
| Security and GDPR | 10 | 4 | 0.40 | 4 | 0.40 | 3 | 0.30 | C’s subprocessor list arrived incomplete. |
| Integrations and data | 10 | 3 | 0.30 | 3 | 0.30 | 4 | 0.40 | Only C has a connector for the firm’s HRIS; A and B import by CSV or API. |
| Implementation and support | 5 | 4 | 0.20 | 3 | 0.15 | 3 | 0.15 | A offered a named lead through the first cycle. |
| Pricing and commercial terms | 5 | 3 | 0.15 | 3 | 0.15 | 4 | 0.20 | Totals within 10% of each other. |
| Weighted total (out of 5) | 100 | — | 3.55 | — | 3.60 | — | 3.25 | A and B are 0.05 apart. The panel ran a second scripted demo on goals and 360° — the two categories where they differ — before deciding. |
If CLEAR Talent is on your list
How CLEAR Talent answers the questions that usually decide a shortlist
We would rather you scored CLEAR Talent against this RFP than against our own feature list, so here are the answers that most often decide a shortlist — including the two that rule us out for some buyers. Reviews run in stages — self-assessment, manager review, calibration and final rating — each with its own deadline and automated reminders on whatever is outstanding. Calibration happens inside the cycle, with live heatmaps and outlier flags across managers and forced distribution as an optional guardrail, and every change to a rating or comment is logged with who, what and when. Goals cascade from company OKRs to individual goals, and 360° feedback is structured against your competency framework rather than run as a separate survey.
Ava, the AI assistant, labels every output as AI-generated, leaves it editable and approved by a person, never makes the final rating, and can be switched off per cycle, team or module; your employees’ data is never used to train models for other customers. CLEAR Talent is hosted on Google Cloud Platform with regional requirements agreed per deployment, provides data subject access request exports and right-to-be-forgotten workflows, and offers a data processing agreement you can request for your DPO to review. The two answers to know now: there is no SAML single sign-on or SCIM provisioning today, and no prebuilt connector to a named HRIS — people and structure come in by CSV or Excel bulk import or API access, and records go out as CSV, Excel or PDF on every plan.
On price, the plans are published: Basic is USD 39 a month for 1 admin user and 10 staff users, Pro is USD 295 a month for 3 admin users and 50 staff users, with published rates for additional users, and Enterprise is quoted. Annual billing saves 15%, setup fees are quoted depending on the package, and there is a free 1-month trial — long enough to run the scripted demo scenarios above with your own managers.