Subscription vs Pay-As-You-Go: Pricing Models for Candidate Testing

Two ways to pay for candidate testing

Every buyer of candidate assessment tools eventually faces the same commercial decision: do you commit to a subscription, or do you pay per test as you go? The pricing model you choose has less to do with the quality of the assessment and everything to do with how, when and how often you actually hire.

Get it right and your testing budget scales cleanly with headcount plans. Get it wrong and you either overspend on an allowance you never use, or you hit a wall mid-campaign because you have run out of committed capacity. This guide breaks down how each model works, where each one wins, the hidden costs to watch for, and a simple break-even framing so you can match the model to your hiring pattern rather than a vendor's preferred contract.

How each pricing model works

Subscription (committed volume). You agree an annual or multi-year contract, usually with a fixed number of assessments per period or a per-seat licence for the people who send and review tests. You pay whether or not you use the full allowance. In return you often get a lower per-test headline rate, account management and bundled reporting. Human-graded vendors tend to sit here: for example, Pipplet is typically quote or subscription-based, with human-graded results returned in around 24 hours.

Pay-as-you-go (credits). You buy assessment credits and spend one each time you test a candidate. There is no contract, no minimum and no per-seat fee. When hiring is quiet, you spend nothing; when a campaign lands, you top up. International English Test uses this model: credit-based pricing of roughly £8.99–£11.99 per test depending on volume, with no contracts, no sales calls and fully self-serve onboarding.

The core trade is predictability versus flexibility. Subscriptions reward you for committing in advance. Pay-as-you-go rewards you for only testing who you need, when you need to.

When a subscription makes sense

A subscription can be the right call, but only under fairly specific conditions:

• Steady, high volume. You test a large, consistent number of candidates every month and can forecast that number with confidence.
• Predictable pipeline. Your hiring is not seasonal or campaign-driven; the flow is broadly flat across the year.
• Many active reviewers. You have enough hiring managers testing regularly that per-seat or committed pricing genuinely lowers your blended cost.
• You value bundled services. Dedicated account management, custom reporting or integration support matter enough to justify the commitment.

If all four are true, a committed rate can beat pay-as-you-go on unit cost. The keyword is consistent: subscriptions are efficient when your usage curve is flat and full.

When subscriptions waste budget

The same model that rewards steady volume punishes variable hiring. Watch for these patterns:

• Spiky or seasonal hiring. If you run two big recruitment pushes a year and go quiet in between, an annual allowance sits idle for months while you still pay for it.
• Pilots and one-off projects. Testing a single cohort, trialling a new role or validating a market does not justify a year-long commitment.
• Small teams paying for unused seats. Per-seat licensing means you pay for occasional reviewers who log in a handful of times a year.
• Uncertain headcount plans. If the board might freeze hiring or double it, committing 12 months ahead is a gamble either way.

In all of these cases you are buying capacity you will not use. The lower per-test rate is irrelevant if you only consume a fraction of the allowance.

The hidden costs of subscriptions

The headline per-test price rarely tells the whole story. Before you sign, price in:

• Annual minimums. You commit to a floor of spend regardless of actual hiring. Under-hire and you still pay.
• Lock-in and auto-renewal. Multi-year terms and automatic renewals make it hard to leave if your needs change or the tool underperforms.
• Overage rates. Exceed your allowance and the per-test cost above the cap can be far higher than the committed rate.
• The sales cycle itself. Quote-based tools require calls, proposals and procurement sign-off. That is days or weeks before you can test your first candidate, plus internal time cost.
• Seat sprawl. Licences bought for a busy quarter keep billing when the hiring slows.

None of these appear on the per-test line. All of them affect your true cost per hire.

Why pay-as-you-go suits variable hiring

Credit-based pricing removes the commitment risk entirely. You test only the candidates you need to test, which means your spend tracks your actual pipeline rather than a forecast made months earlier. There is no minimum to hit, no seat to justify and no renewal to negotiate out of.

It also removes the sales cycle. A self-serve, credit-based tool lets a hiring manager buy credits and send an assessment the same day, which matters when a strong candidate is in play and you cannot wait a week for procurement. This is exactly the profile of a self-serve alternative to Pipplet: automated and AI scoring return a CEFR result across all four skills — listening, reading, speaking and writing — in minutes rather than around 24 hours, with no quote required.

That said, faster and human-graded are different design choices rather than better or worse in the abstract. If you want to weigh accuracy, turnaround and cost against each other, our comparison of AI vs human-graded tests covers the trade-offs in detail.

Comparing the models at a glance

A rough break-even framing

There is no universal cut-off, but the logic is simple. Estimate your realistic annual testing volume, not your optimistic one, then compare two numbers:

1. Pay-as-you-go cost = expected tests × per-credit rate. Nothing is wasted, so this is your true spend.
2. Subscription cost = committed allowance × committed rate, plus seats, including the portion you will not use.

A subscription only wins when its committed rate, applied to the volume you will genuinely consume, beats your per-credit price after you strip out unused allowance and admin overhead. Because most organisations over-forecast their own hiring, the crossover point is higher than vendors imply. If your volume is uncertain or lumpy, pay-as-you-go is usually the safer commercial bet — you keep the flexibility and forfeit nothing when a quarter goes quiet.

A buyer's checklist

Before committing to any pricing model for candidate assessment tools, work through these questions:

• Is my hiring volume genuinely steady, or does it spike around seasons and campaigns?
• Can I forecast next year's testing volume within a tight range, honestly?
• How many people actually need to send and review tests — and how often?
• What is the annual minimum, and what happens if I under-use it?
• What are the overage rates above the cap?
• Is there auto-renewal or multi-year lock-in, and how do I exit?
• How long from signature to sending my first test?
• Does the price cover all four skills and a recognised CEFR output, or only part of the assessment?
• Is the provider a credible standards body member — for instance, International English Test is an ALTE Associate Member?

Map your answers against the two models. If your hiring is predictable and heavy, price a subscription carefully and negotiate the minimums. If it is variable, seasonal, exploratory or small, credit-based pay-as-you-go will almost always protect your budget better while giving you the same CEFR-graded assessment on demand.

Ready to test candidates without a contract or a sales call? Explore English assessment tests for companies.