Coding Assessment Pricing Without an Annual Contract

In one line: only two of the tools compared here sell coding tests with no contract at all: TestDome, with prepaid packs from $100 for 5 candidates (down to $7 per candidate at volume) that never expire, and Coderbyte, through its per-candidate checkout. CoderPad, CodeSignal, HackerRank, and HackerEarth each sell a month-to-month plan you can cancel any month. Most other monthly-looking prices are annual commitments displayed as a monthly number, so compute your own total before you decide.

If you hire occasionally or seasonally, start with the commitment shape behind each price. The coding assessment market runs on several shapes, and most of them are not what the headline price implies. This guide explains each shape so you can see what you would actually be signing.

Key points

• Watch the commitment shape, not the monthly headline price: several monthly-looking headline prices are annual commitments displayed as a monthly number (TestGorilla's $135/mo and CodeSignal's $79/mo are annual up front), so check the commitment shape behind every monthly figure.
• Pay-per-candidate usually wins at low and bursty volume: it has no fixed base fee and no attempt cap, so you pay only for the candidates you test. TestDome packs run from $100 for 5 candidates ($20 each) down to $7 each on the 600-pack ($4,200), with proctoring included and nothing added later.
• Pay-per-candidate is one way to avoid a lock-in, and a true month-to-month plan is the other: HackerEarth offers a month-to-month plan from $99/mo, so at low burst volume you are choosing between paying once and paying a predictable monthly fee. There is no universal crossover. Compute your own total from each vendor's base fee plus overage.
• Check the catch before you commit: a low price is wasted money if the tool misses what you need. Confirm that the catalog covers the exact roles you hire for, and that the plan includes the anti-cheat protections a 2026 hiring test requires, since some vendors gate those protections to a higher tier.

On this page

• How a 2026 buyer actually decides
• Which coding assessment tools have no annual contract?
• How the pricing compares: commitment shape and entry price
• Is per-candidate cheaper than a subscription at your volume?
• What are the tradeoffs of pay-per-candidate pricing?
• Coding assessment pricing FAQ

How a 2026 buyer actually decides

Three things are must-haves in 2026. A tool that misses one is off the list before you even look at price.

• It measures the right thing. The test needs questions for the exact roles you hire for, difficult enough and close enough to the job to predict real performance. A test that doesn't cover your stack, or asks only trivia, is useless whatever it costs.
• It stops a candidate from running an invisible overlay assistant on the test machine. An invisible overlay assistant is a tool like Cluely or Interview Coder that feeds answers in a window only the candidate can see. A webcam and plagiarism checks alone will miss that window. Stopping it takes a locked-down environment on the machine itself. A test a candidate can pass with that kind of help tells you little you can trust.
• It does not demand an annual contract. Paying $2,000 or more whether or not you hire that year is the blocker that stalls a deal for any team whose hiring comes in bursts rather than a steady stream.

Those three must-haves are not equally firm. A team that finds the right, cheat-resistant test will usually accept a contract to get it. So the no-contract requirement is the one teams relax first.

Good-to-haves come next and rank lower at this volume: deeper proctoring on top of the lockdown, broader library coverage beyond the roles you hire for, and native ATS (applicant tracking system) support. Deeper proctoring here means AI analysis of the screen and webcam that flags disallowed-resource use, a second device, or help from another person. All three matter. None is decisive at a few dozen or a few hundred candidates a year.

Once a tool clears the must-haves, cost decides between the tools that remain. For the occasional hirer, cost means four specific things:

• the total all-in price,
• the commitment shape behind a monthly-looking number,
• whether billing stops in the months you are not hiring, and
• whether the price is public at all, or only a contact-sales quote that a smaller company (SMB) cannot negotiate down.

Watch the all-in price most closely. Per-attempt overages, non-rollover credits, and renewal increases can make the real bill several times the headline price. Library size and integration counts matter too, but they belong in the head-to-head comparison pages, not here.

Every price below is the vendor's own published figure, so you can confirm each one on its pricing page. Where a subscription costs less, we say so. TestDome publishes this guide.

Which coding assessment tools have no annual contract?

Two of the tools compared here sell assessments with no contract at all: TestDome, whose whole model is prepaid per-candidate packs at a public rate, and Coderbyte through its separate per-candidate checkout. If a subscription you can cancel any month also works for you, CoderPad, CodeSignal, HackerRank, and HackerEarth each sell a month-to-month plan. The rest run on other shapes: an annual price displayed as monthly, an annual-commitment plan, an unlimited-usage subscription, or a contact-sales quote with no public number.

• True self-serve pay-per-candidate, no contract: TestDome, where the whole model is prepaid packs at a public per-candidate rate, the packs never expire, and large packs can be paid by wire transfer or purchase order (PO). Coderbyte also sells candidates this way through a separate checkout.
• True month-to-month option: CoderPad's $120/mo Starter, CodeSignal's $99/mo Build plan, HackerRank's plans ($199/mo Starter, $449/mo Pro), and HackerEarth's plans (from about $99/mo). Each has a true month-to-month rate alongside a cheaper annual commitment.
• Monthly-looking price that is really annual: TestGorilla's $135/mo and CodeSignal's headline $79/mo are a single annual invoice divided by twelve for display. CodeSignal's true month-to-month rate, if you decline the annual discount, is $99/mo. The TestDome vs. TestGorilla comparison walks through that annual-only structure in detail.
• Annual commitment only: Codility. Its Starter card says annual billing only. Its Scale plan offers monthly billing at roughly $600 a month, marked 'Annual commitment, billed monthly', so neither plan is a true month-to-month.
• Unlimited-usage subscription: Coderbyte's flat monthly and yearly plans give unlimited total candidates. They cap how many candidates run at once, and they gate some features to add-ons, up to a $12,495-a-year unlimited Bundle. Coderbyte is the one vendor here that also sells candidates individually, so its cost depends on which option you pick.
• No published price at all: iMocha and Mettl (Mercer | Mettl) require a sales call before you see any number. Most other vendors also route their largest deals to a contact-sales quote at the top end. They at least publish an entry price, though, and let you buy at that price without talking to anyone. With iMocha and Mettl there is no self-serve figure to start from.

Two options avoid buying a dedicated tool at all, and a cost-conscious small team should weigh both. The first is the screening built into your ATS, the software you already use to manage candidates. That screening is usually free or bundled into a seat you already pay for, so it adds nothing to the bill. The catch is quality. ATS question banks tend to be thin and generic. The screening rarely includes any lockdown or proctoring to stop an overlay assistant. And the software is built for tracking candidates first, with skill measurement as a side feature.

The second option is building and maintaining your own test. There is nothing to license, and you can tailor the test to your exact stack. The cost is ongoing engineering time to write and refresh questions, and there is no proctoring unless you build that too. The hardest part is weak predictive validity: a home-made test is rarely calibrated against real hiring outcomes the way a vendor's item bank is. Both can be the right call for a very small or very specialized team. Both trade a license fee for gaps in coverage, integrity, and validity that a paid tool is meant to close.

How the pricing compares: commitment shape and entry price

TestDome runs entirely on no-contract prepaid packs at a public per-candidate rate. Every competitor price billed per month is worth checking for a hidden annual commitment before you compare it to TestDome's entry price.

The price column pairs each tool's entry price with its commitment, since a monthly-looking number often carries a 12-month contract. The middle column is the decisive one for 2026 developer screening. It names the plan and the price where a machine lockdown that stops an overlay assistant first becomes available. Several vendors offer a lockdown but gate it well above their entry plan. No lockdown catches everything. An overlay assistant fed from a second physical device can still get through, so treat a lockdown as a strong barrier rather than a guarantee. The last column answers the question that matters most for a team that hires in bursts: does the tool keep billing you in the months when you are not hiring?

Across these coding assessment tools, TestDome's published price is the whole amount owed, with no annual term, no attempt cap, and no overage added later.

A machine lockdown that stops an overlay assistant is rarer. Several tools offer only the video-and-telemetry kind of proctoring. Among the tools that do lock down the machine, HackerRank gates the lockdown to its Pro plan (the TestDome vs. HackerRank comparison works through that gating), Codility gates it to a contact-sales quote, and iMocha and Mettl publish no price at all.

Two tools clear the lockdown and no-contract must-haves at a low, published price: TestDome from $100 for 5 candidates with no contract, and HackerEarth at $99 a month for 10 candidates on a month-to-month plan. Check both catalogs against the roles you hire for, since the first must-have is yours to confirm. After that, the choice comes down to billing shape. For a single burst of around 10 candidates, HackerEarth's $99 undercuts buying two TestDome 5-packs. HackerEarth's flat monthly fee is also easy to budget when your hiring follows a steady rhythm. TestDome wins when hiring is sporadic. You pay once, there's no monthly fee to remember to cancel, and you keep whatever you don't use, so nothing is wasted in the quarters you do not hire.

Is per-candidate cheaper than a subscription at your volume?

Usually yes at low and occasional volume. A subscription bills its full base fee before you test anyone, caps how many candidates that fee covers, and then charges overage for each candidate above the cap. Pay-per-candidate has no base fee and no cap, so you pay only for the candidates you test.

The number on a subscription pricing page is rarely the number you pay. A flat plan bundles a fixed allotment: 25 attempts a month on one vendor, 60 credits a year on another. The plan bills overage above that allotment, so its real cost is the base fee plus whatever your volume triggers, and the headline price hides that second part. TestDome's packs have no base fee and no cap: $100 for 5 candidates, $400 for 25, $1,000 for 100, $2,400 for 300, and $4,200 for 600. Each pack price is the whole cost, with proctoring included.

Where a subscription starts winning depends entirely on its base fee and its cap, so there is no single crossover number. To find your own number, pick the TestDome pack that matches or just exceeds the candidates you expect to test. Unused candidates never expire and carry over to your next hiring cycle. Then add the subscription's base fee to the overage its cap will trigger at that same volume, and compare the two totals. Ask for the overage rate before you sign anything.

Keep the total in perspective too. The amount you compute here is small next to what a screening step is meant to prevent. A per-candidate cost in the tens of dollars, or a monthly fee in the low hundreds, is tiny against the salary, ramp time, and re-hiring that a bad hire or a slow hiring process costs. The exact figure depends on your role and market, so no single ROI number fits every team.

A lower headline price sometimes buys a weaker test. A self-serve plan with only basic snapshot proctoring can undercut a per-candidate cost on paper. But it leaves out the lockdown and AI-analysed monitoring that stop an overlay assistant on the test machine, and in 2026, without that control, a coding test's score is not evidence of unaided skill.

TestDome's own weak point sits at the lowest volume, and it is worth naming. The smallest pack, 5 candidates for $100, works out to $20 per candidate, the highest per-candidate rate across our pack sizes. The $7 headline rate appears only on the 600-candidate pack. The $100 pack still beats any annual commitment at that volume, though a month-to-month plan like HackerEarth's $99 can beat it on a single small burst.

What are the tradeoffs of pay-per-candidate pricing?

Pack pricing trades the familiar flat monthly fee for a lower total cost. That trade has limits worth knowing before you commit.

Before any of the cost math, settle the first must-have: confirm that the library covers the roles you hire for. TestDome spans 130+ skills across coding and non-coding work, but a test is only worth its price if it covers your exact stack at real depth. If you hire for a stack that catalogs often cover thinly, say a specific data-science, C++, or Cypress role, or a less common framework, check that your stack is in the catalog before you buy, the same way you would check ATS support below. This check is the one no pricing table can make for you.

Some buyers find pack pricing less intuitive than a flat monthly fee. A plan at $449 a month gives a budget meeting one recurring number to approve, and a pack does not. The mitigation is that the per-candidate rate and every pack size are public, so you can compute your exact annual cost in advance instead of guessing at usage-based billing after the fact.

TestDome does not sell a flat subscription at all. If what you want is a predictable flat monthly fee with unlimited testing regardless of volume, that product doesn't exist here. Once you are testing well past the few-hundred-candidate range every year, most vendors move you onto a custom quote with no public price. TestDome's pack prices stay public at every volume. Packs combine, so 2,000 candidates come to about $12,000, and a custom quote is available if you prefer one.

Also check that your ATS is supported before you commit. TestDome has four native integrations, Greenhouse, SmartRecruiters, Recruitee, and TalentLyft, plus a Zapier connector. Anything outside those four either goes through Zapier or needs developer work against the REST API. API work is not something a non-technical team can set up alone. If your ATS is not one of the four, ask support whether an integration is available or on the roadmap, since new integrations are usually built when customers ask for them. Confirm that Zapier can connect your ATS in the meantime.

If the pay-per-candidate shape fits your volume, you can price TestDome yourself, with no sales call and no annual signature required. See TestDome's per-candidate pricing

Coding assessment pricing FAQ

What is the cheapest coding assessment for occasional hiring?

For occasional or seasonal hiring, TestDome's pay-per-candidate packs are usually the lowest total cost. You pay only for the candidates you test, and nothing in the months you don't hire. A 5-candidate pack is $100, a 25-candidate pack is $400, and a 100-candidate pack is $1,000, all with proctoring included and no renewal. A flat annual subscription bills its full fee whether you test five candidates or five hundred, so it rarely wins for a team that hires in bursts.

At what volume does a subscription become cheaper than pay-per-candidate?

There is no single crossover, because the answer depends on the subscription's base fee and the size of its included attempt cap. A plan with a low annual base fee and a generous cap can win at steady high volume. A capped plan with a high base fee usually loses at the volumes most small teams run. At the low and bursty volumes most small teams run, pay-per-candidate usually wins because there is no base fee to absorb before you test anyone.

Is TestDome's per-candidate price expensive at low volume?

At the smallest size, yes, relatively: $20 a candidate, TestDome's highest rate. Even so, that price buys the complete product, with proctoring included and no annual commitment attached. The rate drops to about $10 a candidate at 100 candidates and $7 at 600.

Which coding assessment tools are pay-per-candidate?

Two of the established tools sell candidates individually: TestDome, where prepaid packs at a public rate are the whole model, and Coderbyte, through a per-candidate checkout that sits alongside its subscriptions. Several subscriptions meter per attempt above an included cap, but that is overage billing on top of a base fee you pay either way, not pay-per-candidate pricing.

How can I tell if a coding test's monthly price is really annual?

Look for the phrase "billed annually" next to a monthly figure. TestGorilla's $135/mo and CodeSignal's $79/mo are both a full year's fee divided by twelve for display. You cannot cancel either one month to month. A true month-to-month option is one you can leave at any point, like CoderPad's $120/mo Starter, CodeSignal's $99/mo month-to-month rate, or HackerRank Pro's $449/mo. TestGorilla's Plus lists only a starting price, with credit counts settled on a sales call. Ask directly for the annual total and the per-candidate math before you sign anything.