What Codility costs in 2026: real per-invite math.

Codility pricing (2026).

Codility meters usage by assessment invite rather than by credit or seat. Each paid plan includes a capped number of invite credits, and one credit covers a candidate in Screen (an assessment) or Interview (a live coding session). Pricing checked October 2026 on codility.com/pricing.

Starter carries a limited task library and basic proctoring with candidate snapshots; Scale adds a bigger task library spanning engineering, AI and system design, custom task building, standard analytics and 3 platform users; Custom adds unlimited users, Skills Intelligence, AI follow-up questions, custom assessments from Codility’s Assessment Science team, integrations and dedicated support. Codility does not publish a Custom price, so the quote you receive is the only reliable figure.

Per-invite amortisation math.

Because the platform bills by invite, the honest cost metric is annual plan price divided by the included invites. The result has a counterintuitive shape: the cheaper Starter plan has the lower per-invite cost, because Scale’s premium buys seats and features rather than a volume discount.

So Starter is the cheaper plan per invite at $10, and Scale runs $20 to $24 per invite depending on billing cadence. You move up to Scale for the 3 platform users, the bigger task library, custom task building and analytics, not to lower the per-invite rate. The per-invite price only falls below the Starter $10 once you negotiate a high-volume Custom contract.

Codility vs HackerRank vs CodeSignal on price.

The three self-service coding-assessment platforms sit close together on price. Codility carries the highest sticker price at both tiers, but its entry tier also includes the most volume, which makes it the cheapest of the three per candidate at that level.

All three columns are annual-billing list prices from each vendor’s pricing page, checked October 2026; more detail on our HackerRank cost and CodeSignal cost pages. Because each vendor meters differently (Codility by invite, HackerRank by attempts, CodeSignal by credits), compare on your own annual candidate volume rather than on tier names: convert each plan to a cost-per-candidate at your real throughput.

The cost-justified hiring-volume threshold.

The right question is whether Codility is cheaper than the engineer-hours it replaces. An auto-scored screen saves roughly 10 to 25 minutes of reviewer time per submission versus a manually scored coding exercise. At a senior IC loaded rate of $122 per hour, that is roughly $20 to $50 per assessment in engineer-time saving, above the $10 to $24 per-invite platform cost.

On that arithmetic Starter at $1,200 per year pays for itself once you run roughly 25 to 60 reviewed assessments a year, within its 120 included invites. Scale at $6,000 billed annually pays back at roughly 120 to 300 reviewed assessments, so at the low end of the saving it needs close to its full 300-invite allowance. Below about 50 assessments a year, a free live-coding tier or a manual approach is usually the cheaper route. The math improves as engineer loaded rates climb and the scoring is actually used to filter expensive onsite stages.

A useful test: count actual completed assessments in the previous 12 months and multiply by your reviewer-time saving per assessment in dollars (roughly $20 to $50 at a $122 senior IC loaded rate). If the result exceeds your Codility annual cost, the platform pays for itself. If not, drop to Starter, a cheaper live-coding tool, or a manual approach.

Where Codility makes most sense versus alternatives.

Codility makes most cost-justified sense for teams that run real volume on standardised, validated screening assessments and want a single platform spanning auto-scored screens and live coding interviews. Its capped-invite model makes the spend predictable, which mid-market finance teams tend to prefer over open-ended credit balances.

It makes less sense for teams that primarily run live coding interviews (cheaper on a live-coding-only tool like CoderPad), for teams that will not use the standardised scoring as a structured filter, and for low-volume teams under the per-invite payback threshold. The right evaluation is a short pilot that measures whether the scoring actually reduces expensive onsite volume, then a decision on real data rather than a feature-comparison spreadsheet.

Cross-references.

For HackerRank cost (the direct competitor), see the HackerRank cost page. For CodeSignal (credit-based model), see the CodeSignal cost page. For Karat (outsourced interviews, different model), see the Karat cost page. For CoderPad (cheaper live-coding-only), see the CoderPad cost page. For the broader tooling cost framework, see the tools page.

Related reading