Hierarchical publish gate · The UltimateSEO Framework
The publish gate: hard gates first, then the quality score
The gate is not a ranking formula. 52 of the 128 controls are hard gates. They are decided first and never scored: if any one fails, the page is fixed before anything else. The 100-point quality score is always reported with its evidence coverage and a possible range, and no result authorises publication by itself.
1. Hard gates (not scored)
If any hard gate fails, the decision is gate failed, whatever the total score. If none has failed but any is still UNKNOWN or BLOCKED, the decision is insufficient evidence. Every P0 control is a hard gate, and so are E-02 and E-06.
8 of the 52 hard gates can be checked automatically, 9 give a partial signal and 35 need your evidence. Until that evidence is supplied the gates cannot pass, so the decision is insufficient evidence unless a gate has already failed.
The 52 controls that act as hard gates
- Integrity (7): G-01, G-02, G-03, G-04, G-05, G-06, G-07
- Technical eligibility (10): T-01, T-02, T-03, T-05, T-07, T-09, T-11, T-12, T-13, T-20
- Intent and strategy (4): I-01, I-04, I-07, I-08
- Quality and evidence (4): Q-03, Q-04, Q-05, Q-06
- Retrieval and usability (2): R-14, R-18
- Business identity (6): E-01, E-02, E-05, E-06, E-08, E-09
- Authority and distribution: none
- Measurement (2): M-14, M-16
- Execution safety (9): S-01, S-02, S-03, S-04, S-05, S-06, S-07, S-08, S-09
- Proportionate optimisation (8): O-01, O-02, O-03, O-04, O-05, O-06, O-07, O-08
2. Quality score
| Score bucket | Weight |
|---|---|
| Intent and user-task fit | 15 |
| Accuracy and evidence chain | 15 |
| Original, non-commodity value | 15 |
| Technical eligibility | 12 |
| Retrieval and usability | 10 |
| Business identity | 10 |
| Business usefulness and conversion readiness | 8 |
| Authority and distribution readiness | 5 |
| Freshness and maintenance plan | 5 |
| Measurement and experiment readiness | 5 |
| Total | 100 |
How UltimateSEO computes it
- 94 of the 128 controls feed a score bucket. The other 34 do not change the score: 28 of them are hard gates and 6 are recorded as evidence only (T-17, T-18, T-19, A-12, M-15, M-17).
- Only controls in state PASS or FAIL count. UNKNOWN, NOT_APPLICABLE and BLOCKED controls are excluded, never assumed.
- Per bucket: score = pass ÷ (pass + fail) × weight. A bucket whose controls are all NOT_APPLICABLE is left out.
- The total compares the weight earned by passing controls with the weight that has evidence at all. It is shown with its evidence coverage (evidenced controls ÷ 128) and a possible range: the lowest and highest total the missing evidence could still produce.
- If a hard gate fails the decision is gate failed: the score is still shown, labelled "fix gates first".
- No result marks a release as authorised. Publishing a website change needs its own approved workflow.
3. Decision bands
| Score | Decision |
|---|---|
| 90–100 | Publish |
| 80–89 | Publish with minor fixes |
| 70–79 | Rework |
| 0–69 | Reject |
| Any hard gate FAIL | Gate failed: fix the gate first |
| Any hard gate UNKNOWN or BLOCKED | Insufficient evidence |
Publish and Publish with minor fixes also need PASS or FAIL evidence for every applicable scored control. With gaps, the decision is insufficient evidence.
4. Page-level dependency order
An agent or a reviewer assesses a page in this order. Each question is only worth asking once the one before it has a satisfactory answer.
- G Is there a policy or spam problem?
- T Is the URL technically eligible?
- I Is it the right user, intent and page type?
- Q Is the information accurate, sufficient and original?
- R Can people and retrieval systems understand it easily?
- E Are the author, brand and entity trustworthy?
- A Is there an authority or distribution opportunity?
- M How will the result be measured and kept up to date?
Two areas sit outside this page-by-page order. Execution safety (S) applies to every change an agent or a person makes, and Proportionate optimisation (O) decides whether a change is justified at all, including NO_CHANGE. All 17 of their controls are hard gates.
5. Change classes used in findings
Every recommended change in a report carries one of these classes so you know what can be applied safely and what needs evidence or approval first. A change without a recognised class is treated as APPROVAL_REQUIRED.
- READ_ONLY: Inspect authorised evidence.
- AUTO_SAFE: A small reversible candidate inside the authorised scope.
- AUTO_WITH_TEST: A candidate requiring relevant tests.
- EVIDENCE_REQUIRED: Authoritative facts must be supplied.
- APPROVAL_REQUIRED: Required exact authority must be valid before the consequential action.
- DO_NOT_DO: Prohibited activity cannot be released.