Intro
AI visibility work fails when it skips steps.
A few prompts. A content brief. A report. Then silence.
The Capston Core methodology is built differently. Five stages, in order, with evidence at every stage and verification at the end. No stage is optional. No claim is made without a source.
This is the process every Capston Core engagement follows — for hospitality groups, premium experience brands, and certified agency partners.
Why five stages
Most AI visibility work collapses into a single stage: “we wrote some content, hopefully AI picks it up.”
That is not enough.
Visibility requires a measured before state, a clear diagnosis, a prioritized fix, a tested verification, and a long-tail monitor. Each stage answers a different question and produces a different artifact.
Skip a stage, and the work becomes opinion.
Stage 1: Baseline
Question answered: How do AI engines see us today?
The baseline captures the current state. It runs a structured prompt set across the priority AI engines for the brand’s markets. It records every answer, every citation, every competitor mention, every factual claim.
The baseline produces:
- A prompt library, organized by intent (discovery, comparison, trust, conversion)
- A competitor set, defined and agreed upfront
- Captured AI answers, dated, with model and version
- A citation map: which URLs are cited, on which prompt, in which engine
- Factual accuracy notes per claim
- The first scorecard snapshot
No conclusions yet. Just evidence.
Stage 2: Diagnosis
Question answered: Why are we missing, weak, or misdescribed?
The diagnosis explains the baseline.
It identifies why the brand is absent on certain prompts, why competitors dominate others, why factual errors persist, why citations route to intermediaries instead of the brand’s own site.
Diagnostic categories include:
- Entity gaps — the brand does not exist as a clean entity across the open web
- Schema gaps — pages lack the structured markup AI engines parse
- Content gaps — answer-shaped pages are missing for high-intent prompts
- Source authority gaps — the sources that describe the brand are weak or outdated
- Competitor framing — competitors are described with sharper, more specific language
- Distribution leaks — OTAs, aggregators, or partners capture the answer
The diagnosis is not a list of complaints. It is a ranked set of root causes.
Stage 3: Action plan
Question answered: What do we change, in what order?
The action plan converts the diagnosis into concrete work.
Each action carries:
- A category (entity, schema, content, source, partner, technical)
- An owner (Capston Core, certified partner, client team)
- An effort estimate
- An expected lift on which scorecard dimension
- A success criterion
Priorities are set against the largest gap with the lowest effort. The action plan is not exhaustive — it is the next 8 to 12 weeks of work, sequenced.
What is out of scope is just as important. Capston Core does not bundle every marketing recommendation into the action plan. Content production, PR, design, paid media, and CMS rebuilds are partner work, not method work.
Stage 4: Verification
Question answered: Did anything actually change?
Verification reruns the same prompt set, against the same engines, after the action plan has been executed.
The same scorecard. The same dimensions. The same evidence format.
Verification produces a before-and-after report. It shows which prompts moved, which dimensions improved, which competitors lost ground, which citations now point to the brand instead of intermediaries.
Verification also documents what did not move. That is equally valuable — it tells the next cycle where to focus.
Without verification, AI visibility work is just activity.
Stage 5: Monitoring
Question answered: Are we holding the gain?
AI answers change. Models update. Competitors react. Source authority shifts.
The monitoring stage keeps the same prompt set alive over time. Quarterly retests. Stable dimensions. Trend lines.
Monitoring produces:
- Quarterly scorecard refreshes
- Alerting on dimension regressions
- Competitor movement reports
- Trigger points for the next baseline cycle
This is how Capston Core engagements become long-term programs instead of one-off audits.
The non-negotiables
Across all five stages, three rules hold.
- Every claim is backed by evidence. Prompts, captures, citations, dates, models. If it is not in the evidence layer, it is not in the report.
- The prompt set does not change between stages. Adding or removing prompts between baseline and verification invalidates the comparison.
- Out-of-scope work stays out of scope. Capston Core does not ship content, design, or paid media. Partners do.
These rules are what makes the methodology repeatable across clients, verticals, and certified partners.
How this fits into Capston Core
The methodology is one of six things Capston Core owns. The other five — scoring, data, platform, QA, certification — exist to make the methodology work at scale.
→ Back to Capston Core
→ Explore the AI visibility scoring system
→ See the AI answer evidence layer
FAQ
How long is a full cycle?
Baseline plus diagnosis takes two to four weeks. Action plan execution depends on scope, usually 8 to 12 weeks. Verification adds one to two weeks. Monitoring is ongoing.
Can we start with just a baseline?
Yes. The baseline is the first step. Many engagements start there and progress to action plan and verification once the diagnosis is clear.
Who runs each stage?
Capston Core runs baseline, diagnosis, and verification. Execution can be Capston Core, a certified partner, or the client’s internal team. Monitoring is Capston Core.
What if the prompt set is wrong?
The prompt set is co-designed at the start of the baseline and signed off before capture begins. Adjusting it mid-cycle invalidates the comparison, so it is locked once set.
How is this different from a content brief?
A content brief tells you what to write. The Capston Core methodology tells you whether the brand is visible, why it is not, what to do about it across more than just content, and whether the work moved the needle.
Does this work outside hospitality?
Yes. The methodology is vertical-agnostic. Hospitality is the first proof market, but the same five stages apply to any premium brand whose visibility in AI answers carries commercial weight.
Final CTA block
See the methodology applied to your brand.
Run a baseline
View the hospitality scorecard