Quick Verdict
RoleMath is a decision engine, not a directory, bootcamp, job board, or affiliate blog. Where a figure comes from an official source, the page shows that source and the date we read it — and where we have not recorded something, the page says so instead of filling the gap. Which certification suits you depends on your goal and your background, so nothing here is a single universal answer, and nothing here is a funnel toward a product we sell.
Where Our Data Comes From
| Source | What it backs | Currency |
|---|---|---|
| BLS OEWS | Occupation-level wages | May 2025 national release |
| BLS Employment Projections | 10-year outlook, annual openings | 2024–2034 cycle |
| O*NET 30.3 | Skills, tasks, interests, job zones | O*NET 30.3 |
| Official vendor pages & exam guides | Exam codes, objectives, prerequisites, cost, credential type | As-verified, dated per page |
Why There Is No RoleMath Score
RoleMath used to publish a 0–100 difficulty score. We retired it on 2026-07-28 — our own decision about our own metric, recorded in our project history rather than sourced from anyone else — and the reason is worth stating plainly. One term in the formula — the tier term — was read straight from the level word in a product name, "Fundamentals", "Associate", "Expert", and could contribute up to 80 of the 100 points on its own. Adding that tier term up across every certification in the catalogue and dividing by the sum of all the scores, it supplied 86.8% of every point the formula awarded; the arithmetic is that ratio and nothing more. That figure is RoleMath measuring its own retired formula against its own catalogue; it is not a finding about certifications, and there is no external source for it because there is no external thing being measured. The practical effect was that a word-processing exam scored harder than Security+, CCNA and RHCSA. A number that reproduces a marketing label is our opinion wearing a number's clothes, and it was not repairable by reweighting.
What replaced it is duller and checkable. Certifications are grouped by the experience the vendor itself publishes — quoted, linked, and shown with the date we read the page — and a credential we hold no such statement for is absent rather than defaulted, so the length of that page is itself a statement about how much RoleMath has verified and recorded. We do not estimate exam pass rates. Where a certifying body published one we would quote and date it like any other vendor figure. Across the official vendor pages RoleMath has actually read for the certifications in this catalogue, none has stated a pass rate — that is a record of what our own reading found, not a claim that no vendor anywhere publishes one, and we have not searched beyond the pages we cite.
What We Will Not Claim
- No certification salary or ROI claims — a credential does not earn you a specific amount. We show occupation-level BLS wages as context, separate from any certificate.
- No job, placement, or pass-rate guarantees.
- No "best certification for everyone" — which credential to take depends on the job you are aiming at and what you have already done.
- Where BLS publishes an employment projection for an occupation, the role page shows that occupation's own figure, including when it is a decline, and frames the role as a launchpad rather than an endpoint. This page names no projection of its own: the previous wording said "general IT support is projected to shrink", which is not a BLS occupation title and carried no figure or citation, so a reader could not check it.
Credential-Type Honesty
We keep these distinct on every page and never call a course a "certification": proctored certification · professional certificate · certificate of completion · badge · course / learning hub.
How We Make Money — And What We Don't Do
- Ratings and recommendations are never pay-to-rank; no vendor pays for placement.
- We are vendor-neutral — we compare credentials across providers on the same evidence, not as a single-vendor funnel.
- Any affiliate or referral link would be labeled inline and would never influence a rating; resource lists put official/free options first.