Skip to content

Penetration testing cost, explained

A traditional web application penetration test commonly costs $5,000 to $30,000, according to Astra. This guide shows buyers how pentests are priced, what moves the number, the hidden costs that never appear on a quote, and how to compare vendors side by side.

How much does a penetration test cost?

A traditional web application penetration test commonly costs $5,000 to $30,000, and more for complex applications with many user roles, according to Astra's penetration testing cost guide. Where you land in that range depends on how much there is to test and how many tester days the vendor estimates.

The range is wide because the product being sold is time. A small app with one user role and a handful of endpoints sits near the bottom. A multi-tenant platform with admin, staff and customer roles, a public API and an AI assistant sits at the top or beyond it.

The quote is one cost. The calendar is another. Most buyers discover the second after signing.

How are penetration tests priced?

Most penetration tests are priced by effort. The vendor estimates how many tester days your scope needs and multiplies by a day rate. Scope, the number of user roles, reporting and retesting are then added or bundled. A smaller group of vendors publish fixed prices or sell testing as a subscription.

Scoping itself takes effort. Expect a questionnaire, a call or two and a statement of work before you see a number. A vendor that publishes prices or quotes a fixed figure removes that step.

  • Day rate: the price of one tester for one day. Senior testers and specialist skills raise it.
  • Scope: the number of applications, hostnames, endpoints and features in the test.
  • Roles: each user role is tested against the others, so every added role multiplies the access control checks.
  • Reporting: writing, review and a readout call are often billed as extra days.
  • Retest fees: verifying your fixes is frequently a separate line item or a separate engagement.

What drives the price up or down?

Eight things move a penetration test quote: the size of the scope, the number of user roles, whether APIs and AI features are included, how deep the testing goes, how fast you need it, whether retesting is included and how often you test. Each one adds or removes tester days.

Read your quote against this table. If a line is missing, ask whether it is included or excluded. The cheapest quote often leaves out the API, the AI features or the retest.

Cost driverWhat it meansHow it moves the price
Scope sizeApplications, hostnames and endpoints in the testMore surface means more tester days
User rolesAdmin, staff, customer and tenant accounts tested against each otherEvery added role adds access control checks and days
APIsREST or GraphQL endpoints behind the appOften quoted as a separate test with its own days
AI featuresChat, RAG and agents with toolsOften sold as a separate AI red team engagement
Authenticated depthTesting behind the login with supplied accountsDeeper coverage costs more days than an outside view
UrgencyA start date sooner than the vendor's queue allowsRush fees, or no availability at all
RetestingVerifying fixes after the reportBilled per retest unless bundled
FrequencyOne annual test or recurring runsRepeat engagements multiply the cost unless sold as a subscription

How long does a penetration test take?

A traditional engagement commonly takes 2 to 4 weeks from accepted quote to start, then 5 to 10 business days of testing, with the report arriving 1 to 2 weeks after that, according to Stingrai. End to end, that is roughly one to two months before you can act on the findings.

Time has a price. A deal waiting on a pentest report waits the whole stretch. A release that shipped while you sat in the queue goes untested until the engagement begins. An audit window keeps moving while you wait for a slot on someone's calendar.

Autonomous testing changes the timeline. Grapefruit starts once you verify your domain with one DNS record and sign online. Results land within hours of kickoff.

What are the hidden costs of a penetration test?

The hidden costs are the ones outside the quote: the weeks of lead time before testing starts, the wait for the report, the fee to retest after you fix things, and the engineering hours spent on findings nobody can reproduce. Together they can outweigh the difference between two quotes.

Reproduction is the big one with AI-driven testing. In a Pentest-Tools survey of 158 respondents in June 2026, 87.8% hit AI pentest findings that needed significant manual validation, and 53% ranked proof of exploit as a top buying factor. Unproven findings move the cost from the vendor's invoice to your payroll.

Grapefruit removes that cost. Every finding arrives with the exact request and response that proves it. Your engineer pastes the request, runs it and watches the attack work. After the fix, they run it again and watch it fail.

  • Lead time: 2 to 4 weeks from accepted quote to start is common, per Stingrai. Deals and audits wait with you.
  • Waiting for the report: an issue found on day two may reach your engineers weeks later, once the report is written and reviewed.
  • Paying again to retest: the fix is yours, but proof that it worked is often a new invoice.
  • Findings nobody can reproduce: a vague write-up sends an engineer hunting for hours before any fixing begins.
  • Stale results: the report describes the release that was tested. Your product keeps shipping.

How do you compare penetration testing quotes?

Compare quotes on what you receive and when, as well as the total. Line up the scope, the roles covered, the start date, the report date, the evidence attached to each finding and the retest terms. Two quotes at the same price can differ by a month and by half the coverage.

Ask for a sample report before you sign. It shows you the quality of the evidence faster than any sales call. If a finding in the sample cannot be reproduced from what is on the page, yours will read the same way.

  • Is the API in scope, or quoted separately?
  • Are LLM features and agents tested in the same engagement?
  • How many user roles are tested, and are they tested against each other?
  • When does testing start, and when do I see the first finding?
  • Does every finding come with a request and response I can replay?
  • Is the retest included, and how soon after a fix does it run?
  • What does a second test later this year cost?

Does testing AI features cost extra?

Usually, yes. Most vendors sell AI red teaming and application penetration testing as separate engagements, each with its own scope, quote and report. A product with a web app, an API and an LLM feature can end up buying two or three tests and stitching the results together.

Separate tests also miss what sits between them. Picture a chat assistant that reads a poisoned document, then calls an internal API using its own elevated access. The AI team sees a prompt injection. The app team sees an API working as designed. The real breach is the chain, and nobody was paid to test it.

Grapefruit attacks the web app, the APIs, the LLM features and the agents in one test and reports them in one place. One scope, one quote, one set of findings.

How does Grapefruit price penetration testing?

Grapefruit has two offers. Maintenance plans re-test your verified domains every week or every month, with prices published on the pricing page and card checkout. A full penetration test is quoted at a fixed price, so the number you approve is the number on the invoice. See pricing for both.

There is no demo call and no scheduling queue. Add your hostname, publish one DNS TXT record, sign the testing authorization online and testing starts. Setup takes a few minutes.

Retesting is built in for subscribers. Fix something and it is re-tested on the next scheduled run. Each run is dated, kept in the portal, and shows what is new and what got fixed since the last one. That history doubles as an evidence trail for a SOC 2 Type II audit window.

Questions

How much does a web application penetration test cost?
Traditional web application penetration tests commonly cost $5,000 to $30,000, with complex multi-role applications costing more, according to Astra. The figure tracks tester days: more endpoints, more roles and more features mean more days. APIs and AI features are often quoted separately on top of that.
Why do penetration testing quotes vary so much?
Because vendors estimate effort differently and include different things. One quote covers the web app alone. Another adds the API, a retest and a readout call. Day rates differ with seniority. Line up scope, roles, dates, evidence and retest terms before comparing totals, or you are comparing different products.
Is a retest included in the price of a penetration test?
Sometimes. Many vendors bill retesting as a separate line item or a new engagement, so ask before you sign. On a Grapefruit maintenance plan, a fix is re-tested on the next scheduled weekly or monthly run, and the result is recorded with a date in your portal.
How long does it take to get a penetration test report?
Stingrai reports that traditional engagements commonly take 2 to 4 weeks from accepted quote to start, 5 to 10 business days of testing, and another 1 to 2 weeks for the report. With Grapefruit, results land in your portal within hours of kickoff, and the report is downloadable.
Is a cheap penetration test worth it?
A low price is worth it when the evidence is strong. Judge the test on proof: does each finding include the request and response that demonstrates it, and can your engineer reproduce it in minutes? A test that leaves your team validating findings by hand costs more than its invoice says.
Where can I see Grapefruit's prices?
On the pricing page. Maintenance plan prices for weekly and monthly re-tests are published there, with card checkout. Full penetration tests are quoted at a fixed price. You can start a plan without a demo call: verify your domain with one DNS record, sign online and testing begins.

Standards and sources