How we research and review

We use primary sources: the original survey, study, dataset or official statistics page. Every statistic we publish links to the page it came from, with the publisher and year. Figures we could find only on secondary sites, such as a blog quoting a report we could not open, are excluded or clearly labelled as unverified.

Where we offer a rule of thumb or an example rather than data, we say so in the text.

Each figure is checked on the source page itself, not on a summary of it. Research pages carry the date each figure was last checked. Sample sizes, years and methods are reported alongside the number where the source provides them.

Each post gets a fact-check pass against its sources before it is published and again on every substantive update. Research assets, such as the State reports, also get an adversarial review whose job is to find what a hostile reader would challenge.

A post's "Updated" date changes only when the content changes in substance: a figure, a source, a recommendation or a section. Typo fixes and layout changes do not move the date, and we never set it automatically from the build time.

If you find an error, email hello@ctocoach.com with the page and the problem. We correct it and add it to this log. Showing our corrections is deliberate: it is how you can tell the process works.

Corrections log

1. 2026-10-11

   State of First-Time CTOs 2026: four unconfirmed figures removed or corrected

   We re-opened every source cited in the report. The talent-shortage figure (about 63 percent), attributed to a Deloitte study, does not appear in the cited page, which is a 2020 study. The developer trust figure of 43 percent for 2024 does not appear on the Stack Overflow 2025 page, which reports 46 percent distrust and 33 percent trust. The "roughly a third of developer time" figure attributed to CAST does not appear on the cited CAST page, which reports an estimate of 61 billion workdays forfeited to technical debt. The 91 percent technical-debt survey is STX Next's 2023 survey of 500 CTOs, not a 2025 survey. We corrected the report, the press page, the research hub and the stat cards, and removed an unsourced fractional price range and an unsourced share-of-spend figure.

2. 2026-10-11

   Fractional demand statistic (+68%) removed

   The 68% year-over-year growth figure for fractional executive demand, attributed to Cerius Executives, could not be traced to a primary source: Cerius's own site does not state it, and every other page we found repeats it without a link to a report. We removed it from the State of First-Time CTOs 2026 report, the State of CTO Coaching report, the press page, the research hub and the embeddable stat cards, and replaced the headline slot with a figure we verified at the source (LeadDev's 2025 burnout data).

3. 2026-10-11

   Engineering manager course guide: unverifiable prices and defunct provider removed

   The engineering manager courses guide carried prices and lengths that were not linked to a provider page. We re-checked each provider's own page in October 2026. Prices we could not confirm now read "See provider site". We removed Plato (its community notice says it was sunset on December 19, 2025) and a Pragmatic Engineer course (the publication's homepage lists no such course).

4. 2026-07-14

   Fractional demand statistic re-attributed (Fortium Partners to Cerius Executives)

   A figure for year-over-year growth in fractional executive demand was cited to Fortium Partners, whose page does not contain that number. We re-attributed it to Cerius Executives' 2024 report and fixed the same misattribution in both research reports, the press page and the research hub. The Cerius primary page has not yet been located, so the figure is cited by name without a link and is treated as unverified until it is.

5. 2026-07-14

   ICF coaching revenue growth comparison corrected

   A coaching revenue growth claim compared against the 2019 baseline instead of 2023. It was corrected to the year-over-year growth from the 2022 figure, as published by the International Coaching Federation.

6. 2026-07-14

   Coaching market-size forecast hedged

   A market-size forecast attributed to Mordor Intelligence no longer matched the publisher's live page because its forecast window had rolled forward. We replaced the out-year figure with the current-year size and growth rate, and note that it is not a CTO-coaching market size.

The CTO Readiness Assessment has 22 questions. Scored questions fall into five dimensions: Technical Leadership, People Leadership, Business Acumen, Communication, Self-Management. Each scored answer maps to a fixed value from 1 (gap) to 3 (strong), and each dimension is scaled to 0 to 100. A few questions use rules instead of a single option, for example the time-allocation question scores the mix of coding, strategy, recruiting and firefighting. The scoring is deterministic and contains no AI: the same answers always produce the same result.

Score bands, applied to each dimension and to the overall score:

• Foundation: Below 60
• Development: 60 to 74
• Proficiency: 75 to 89
• Mastery: 90 and above

Limitations. The assessment is a self-report. It has not been validated against outcomes, and there is no benchmark dataset yet, so a score is a prompt for reflection, not a measurement of your ability or a prediction of success.

How we list and describe coaching providers is documented with the directory itself, in the CTO coaching directory methodology. Short version: public provider pages only, "Not public" where a price is not published, and no referral fees.

We accept no sponsored placements and no affiliate or referral fees. FractionalChiefs is our sister site, and we use tagged links to it where it is relevant. We do not expand fractional-executive content on this site. See About.

We use AI tools to help with research, drafting and editing. Every claim, figure and link is checked by a person against its source before publication. AI tools are not used to produce statistics, and the assessment scoring contains no AI.