Recruitment Software Demo Checklist: What to Test Before You Buy
A recruitment software demo checklist should make you test the tool on your own job, your own CVs and your own workflow at each stage: job setup, applications, screening, interviews, feedback, offers, reporting and data export. Score every vendor on the same sheet, watch for red flags, and ask for answers in writing before you pay.
Why you should run the demo, not watch it
A vendor demo is a rehearsed story. The job is well written, the CVs are tidy, the candidates reply instantly and nothing goes wrong. Your hiring is different: duplicate applicants, a scanned CV, a hiring manager who never logs in, an interviewer who forgets to give feedback. A purchase made on a polished demo often turns into regret in month two, when the first real role exposes the gaps. The fix is to take control of the session. Send the vendor your test brief in advance, supply your own materials, and ask to drive the screen yourself or use a trial account. Insist on seeing the real product, not slides. This guide gives you a stage-by-stage list of things to test, a scoring approach, red flags, and questions to ask. It is written for small and mid-size hiring teams who will choose without a procurement department. For background on the category see /recruitment-software and /applicant-tracking-system, and for pricing and plan comparisons on any tool, check the vendor's current pricing and features yourself.
Prepare a test kit before any demo
Build one small kit and use it with every vendor, so the comparison is fair. The kit has five parts. First, one real job description, ideally a recent role. Second, ten to fifteen CVs you already know well, anonymised if needed, including strong fits, clear misses, a scanned PDF, a CV in a different layout and one with a career gap or change. Third, a short list of your must-have features and your non-negotiables, such as export of all data or approval steps. Fourth, a scenario script of what you will do from start to finish: open a role, receive applications, shortlist, schedule, collect feedback, make an offer. Fifth, a scoring sheet, shown later in this guide. Decide who from your team will attend, and ask a hiring manager to join, because they are the person most likely to avoid a tool that is hard to use. Agree the scoring in advance so that a charming presenter does not sway the group. Send the brief to the vendor two or three days before the session and ask them to confirm they can cover it.
Tests by stage
Use the table below as your core checklist. For each stage, ask the vendor to perform the test live using your kit, and mark the result.
| Stage | Test to run | What good looks like |
|---|---|---|
| Job setup | Paste your job description and build the role | Must-haves are clear and editable by you |
| Applications | Apply as a candidate on a phone, then upload a batch | Short, easy form; many files accepted at once |
| Screening | Run your ten known CVs | Ranking close to yours, evidence shown per score |
| Unusual CVs | Include a scanned or oddly laid-out CV | Read correctly, or clearly flagged as unreadable |
| Shortlist | Compare two or three candidates side by side | Clear view, notes, a place to record the reason |
| Interviews | Book a slot as a candidate | Self-scheduling, invites, reminders work |
| Feedback | Collect scorecards from three interviewers | Interviewers can respond without a heavy login |
| Decision | Record a team decision and a rejection reason | Decision and who made it are stored |
| Offer | Draft, approve and send an offer | Approval step and signed record available |
| Reporting | Ask for time per stage and sources | Numbers come from real activity, not guesses |
| Export | Export every record | Complete, usable file without extra charge |
Do not accept answers such as we can do that. Ask to see it. If something is described as coming soon, treat it as not available and score it zero until it ships and you have seen it work.
What to test on screening and AI features
If a tool offers AI scoring or ranking, test it harder than anything else, because it affects who gets considered. Run your ten known CVs and compare the order with your own. Look for evidence: can you see which lines of each CV supported the score? Can you set the must-have criteria yourself? Run the same CV twice and see whether the score is stable. Include a scanned file and see whether the tool says so when it cannot read something. Ask what happens to people the tool scores low: are they rejected automatically, or does a person decide? Ask how it treats career gaps, changes of direction and non-traditional backgrounds. Ask where candidate data goes, whether it is used to train models, how long it is kept and how you delete it. Rules on automated decisions and candidate data differ by country, so check local law and ask the vendor for written terms. Our guide to /ai-resume-screening explains what a fair setup looks like, and /resume-parser covers the reading step.
What to test on the hiring manager and interviewer experience
Software is only useful if the people who must use it actually do. Ask the vendor to show the product through the eyes of a hiring manager who has ten minutes a week. Can they see the shortlist and give a yes or no without learning a whole system? Ask to see the interviewer's view: how do they get the scorecard, how many clicks to submit, and is it usable on a phone? Check the candidate's view too: how the application looks, how scheduling works, whether they get confirmations and reminders, and whether the messages look professional. Test how the tool handles reschedules and no-shows. Finally, ask how permissions work: can you give someone view-only access, hide salary data or restrict a role to certain people? A tool that only the recruiter can use leaves you chasing feedback by email again, which defeats much of the purpose. Look at the tool with the least patient colleague you have, and watch where they hesitate.
Red flags during a demo
Some patterns should slow you down. The vendor will not use your CVs or your job description. Features are shown only in slides or recorded videos. Pricing is hidden until late in the process, or the answer changes between calls. Answers about scoring are vague, such as proprietary model, with no evidence view. The tool rejects or hides candidates automatically with no human step. There is no export of all data, or export costs extra. Contract terms push a long commitment with a hard auto-renewal. Support is described as available but cannot be shown. Key features are labelled coming soon. The vendor criticises competitors without specifics, or states a competitor's price or feature as fact, which you cannot rely on, so check each competitor's current pricing and features yourself. They avoid questions about data location, retention or deletion. They pressure you with an end-of-month discount before you have tested anything. One red flag is a conversation; several together are a reason to walk away.
Questions to ask every vendor
Ask the same questions of everyone and write the answers down. Pricing: what exactly is included in the plan we would use, what costs extra, and what will the price be at renewal? Seats: how many people can use it, and is there a charge for hiring managers? Data: where is it stored, who can see it, how long is it kept, how do we export and delete it? Security: what controls exist, such as access permissions and audit records? Support: how do I reach a person and how quickly do they respond? Onboarding: how long to import our existing candidates and set up? Contract: what is the minimum term, how do we cancel, and what happens to our data afterwards? Roadmap: which features are live today, and which are planned? Integrations: how does it connect to our other systems, through a native link, an API, webhooks or an automation tool? References: can we speak to a team of our size? Treat a refusal to answer in writing as information.
A scoring sheet you can copy
Score each vendor from zero to three on each line, where zero is no, one is partly, two is yes and three is yes and demonstrated on your kit. The weights are example weights, so change them to suit your priorities.
| Line | Weight | Vendor A | Vendor B |
|---|---|---|---|
| Ran on our CVs and job | 3 | ||
| Evidence shown for every score | 3 | ||
| Human decision before any rejection | 3 | ||
| Hiring manager and interviewer ease of use | 3 | ||
| Full data export | 3 | ||
| Scheduling and reminders | 2 | ||
| Offer approval and signing | 2 | ||
| Reporting on real activity | 2 | ||
| Price clarity and fit | 2 | ||
| Support and onboarding | 1 |
Multiply each score by its weight and add up. Treat any zero on a weight-3 line as a veto to discuss, whatever the total. Have two people score separately before comparing, then discuss the largest differences. A tool that wins on total but fails a veto line should not win.
A worked example with example numbers
The numbers below are examples only. A 30-person company tests two tools on the same job and the same ten CVs. The team's own ranking puts CVs 3, 7 and 9 at the top. Tool A places two of the three in its top three, shows the CV lines behind each score, reads the scanned file correctly and lets the hiring manager vote from a simple page. Tool B places one of the three in its top three, shows only a number with no evidence, and cannot read the scanned file. Tool B's presenter says the scanned file would be supported in the next release. Scored on the sheet above, with the example weights adding up to 24 and a top score of 72, Tool A reaches 55 and Tool B reaches 33, and Tool B also scores zero on the evidence line, a veto. The team asks Tool A for written data terms and a quote with renewal pricing, runs a one-week trial on a live role, and decides after the trial. These results are invented to show the method, not a verdict on any real product.
How to run a one-week trial after the demo
If a vendor passes the demo, do not sign yet. Ask for a trial account or a short paid pilot and run one live or recently closed role through it for a week. On day one, load the job and import or upload the CVs. On day two, review the scores against your own judgement and note every disagreement. On day three, book two mock interviews with colleagues acting as candidates and check invites, reminders and the candidate view on a phone. On day four, ask two interviewers to submit feedback and time how long it takes. On day five, draft a sample offer and push it through approval. Keep a log of every moment of confusion, every missing feature and every time you needed support, and note how quickly support replied. At the end, ask each participant three questions: what was easy, what was annoying, and would you use it for the next role. Their answers, with your scoring sheet, are the evidence you take to your manager or finance lead, and they are far more reliable than the demo.
Where Rezumz fits
Rezumz is a hiring workspace for small and mid-size hiring teams, and you can run this checklist on it. Start with the free plan, which includes 2 free candidate analyses, or the live demo at /demo. It scores each CV against the must-haves you set and shows the lines it relied on, collects applications on a link, schedules interviews, gathers scorecard feedback through personal links, records the team decision, and handles offer letters with approval and a simple typed-name electronic signature. Hiring managers get a view-only review page. People make every decision, and nothing is rejected for you. It is not a job board and has no one-click multiposting, and it is not an HRIS or payroll system, so large enterprises needing deep systems will want something else. It connects to other tools through webhooks and a read-only API rather than native links. Prices: $29 a month, $199 a year, a limited $499 lifetime option, and Team at $79 a month or $699 a year for up to 10 teammates, suited to 4 or more recruiters. See /pricing and /compare. Score it on the same sheet as everyone else.
Common mistakes and when not to run a full evaluation
Common mistakes: judging by the vendor's sample data, letting one enthusiastic person decide, skipping the hiring manager's view, forgetting export and exit terms, comparing tools on different tests, buying for features you will not use, and ignoring the cost of switching later. A full evaluation like this one is not for everyone. If you hire one or two people a year, a spreadsheet or a free tracker may be enough, and a long vendor process costs more time than it saves. See /free-applicant-tracking-system. If your group already mandates a system, test whether it covers your needs and escalate the gaps instead. If you are not ready to define your own process, a demo will not fix that, and you should start with a simple written process and a scorecard before you look at tools at all.
Frequently asked questions
What should I test in a recruitment software demo?
Test each stage with your own materials: job setup, applications, screening on ten CVs you know, shortlist, scheduling, interviewer feedback, decisions, offers, reporting and full data export. Ask to see each live, not described. Score every vendor on the same sheet and treat features marked coming soon as not available.
How do I test AI screening in a demo?
Run ten CVs you already know, including a scanned one. Compare the ranking with yours, check that every score shows the CV lines behind it, run one CV twice to test consistency, and confirm a person decides before any rejection. Ask where data is stored and how it is deleted.
What are the biggest red flags from a vendor?
Refusing to use your CVs, hidden or shifting pricing, vague answers about scoring, automatic rejection with no human step, no full data export, long auto-renewing contracts and pressure to sign before testing. One red flag is worth a question; several together are a reason to walk away.
How many vendors should I compare?
Two to four is usually manageable for a small team. More than that costs time without improving the choice. Use the same test kit, the same brief and the same scoring sheet for each, and check each vendor's current pricing and features yourself.
Should hiring managers join the demo?
Yes. They are the people most likely to avoid a tool that feels heavy. Ask to see the product from their view, with ten minutes a week and no training, and from an interviewer's view on a phone. If they cannot use it easily, feedback will drift back to email.