How to check a coding school before you pay

There is one question that separates a school worth considering from a school worth avoiding, and it is not about curriculum. It is: who checked the outcomes figures you are being shown? For almost every school, the answer is the school. For a very small number, the answer is an auditor, and this guide is mostly about what that difference looks like on paper.

The one audited standard, and how small it is

The Council on Integrity in Results Reporting publishes a reporting standard for coding schools. Its six core standards, in its own words, are truth in advertising, early intent tracking with student career goals documented on day one and verified by auditors, complete enrolment data, comprehensive job tracking, regular annual reporting, and independent verification through annual third-party audits.

On 24 September 2026, its school-data page listed reports from three schools. Three. Whatever you think of the standard, that is the size of the audited universe, and the newest reports on it covered 2023 to 2024. Everything else in this market is a school describing its own graduates.

What an audited report contains that marketing does not

Two reports are public and worth reading side by side, because together they show what the format makes impossible to hide.

Read the last two rows first. Hired by the school itself is a row that exists because schools have historically employed their own graduates and counted it as placement; both of these report zero. And the salary-reporting share is the row that tells you how much weight the median above it can carry. A $115,000 median drawn from six in ten job-obtainers and an $80,000 median drawn from ninety-seven in a hundred are not the same kind of statement, and no marketing page would ever show you both numbers together.

The reports also carry a footnote explaining the highlighting: the marked figure is the canonical number, which the school is required to make the most prominent one in its advertising. For both schools the canonical employment figure is the 180-day one, not the friendlier 360-day one. That single rule does more to discipline a marketing page than any amount of good intention.

What the audited reports still do not tell you

Three limits are worth holding in mind, and the standard itself makes them visible rather than hiding them. The reports we read cover graduation periods in 2023 and job search periods running into 2024, so they describe a hiring market that is now two years old. The cohorts are different sizes — 865 graduates in one report, 105 in the other — and a percentage drawn from a hundred people moves a great deal with a handful of outcomes. And the schools that report are the schools that chose to report, which is a selection you cannot correct for.

None of that makes the reports less useful. It makes them the floor of what a school should be willing to tell you, rather than a guarantee about what will happen to you.

The five questions to ask a school that has no audited report

1. What is the total tuition, as one number, in writing? A school that will not answer this by email has told you something.
2. Which cohort does your published outcomes figure cover, and how many people were in it? A percentage without a denominator is not a measurement.
3. How do you define employed? In-field, full-time, and within how many days from graduation?
4. How many graduates did you employ yourselves, or place with a partner you have a commercial relationship with?
5. What do I owe if I withdraw in week two, week six, and week ten?

The fourth is the one that produces the most revealing pauses. The fifth is the one that actually costs you money if you get it wrong.

Reading a school's own report, when that is all there is

Most schools publish something, and an unaudited report is not worthless — it is just a document you have to read adversarially. Three habits help. Find the denominator: how many people started, how many finished, and how many of those are in the figure being quoted. Find the window: employment within ninety days and within a year are wildly different claims, and the audited reports above show the gap can be more than twenty-five percentage points at the same school. And find the definition of employed, which should specify in-field and full-time and should exclude anyone the school itself hired.

If a report will not give you all three, the honest summary is that you do not know the school's outcomes, and neither does anyone else quoting them at you.

Where the official numbers sit

For a reference point outside the industry entirely, the US Bureau of Labor Statistics publishes occupational data. In figures last modified on 27 August 2026, it put 2025 median pay for software developers, quality assurance analysts and testers at $134,040 a year, and for web developers and digital designers at $99,520. It projects 10% growth for the first group between 2025 and 2035 and 5% for the second, and it lists a bachelor's degree as the typical entry-level education for both.

That last item is not a verdict on bootcamps — plenty of people are hired without a degree — but it is the baseline description of the occupation you are buying your way toward, and it is worth knowing that the official description does not mention a fourteen-week course.

What we will not do, and why you should not either

We do not publish a school ranking, a score or a placement rate we have not seen audited. A score from an organisation that has enrolled in nothing is decoration in the shape of a measurement, and it is exactly the thing that makes this category untrustworthy. If you want one number to compare schools on, use the published tuition in what a coding bootcamp costs and the audited report if one exists. If neither exists, you are being asked to buy on trust, which is allowed, but should be a decision rather than an accident.

The broader case, including what the same money buys elsewhere, is on whether a coding bootcamp is worth it.