Public example · Our own website · September 17, 2026
Can buyers find the answers they need?
Follow a buyer question from website evidence to a concrete improvement and an acceptance check. Use our own site as an example, or open SIGNAL directly.
Live · Public
Send your public URL, target market, and the decision you need to make. Or start directly at signal.weavist.ai.
In WEAVIST ACE, this example belongs to the Acquire stage (SIGNAL). See how WEAVIST ACE fits together
Page text revised · observations from , not re-run
Completed assessment · SIGNAL Expert v2
Five questions, with the gaps visible
WEAVIST AI assessed its own public website using SIGNAL. The assessment completed; that does not mean every capability passed. Some continuation and independent-source evidence remained unverified. Scores can vary between runs: check the evidence before treating a change as an improvement or regression.
- AIO
- 44 / 100
- GEO
- 39 / 100
- SEO
- 83 / 100
These scores describe weavist.ai as it was on the observation date, before the October 2026 rebuild. The rebuilt site has not been re-assessed, so no newer score is shown.
Search observation: 0 findings in 5 completed queries using Gemini with Google Search grounding. This bounded sample does not establish absence from all searches or AI answers. The run did not directly observe every target engine.
Use the report to inspect evidence and choose work to verify. These scores do not measure rankings, visits, qualified enquiries or ROI.
The assessed questions
English · US market · submitted pages: /signal and /spatial
- AI visibility diagnostic tool
- How to evaluate AI search readiness for a public website without installing plugins
- Top AI visibility diagnostic platforms for web marketing teams
- Is GEO readiness diagnostic analysis worth implementing for public site content
- How to turn AI discovery audit recommendations into developer actionable tasks
The editorial walkthrough below explains how to act on the dated self-assessment above. The September 9 archive uses different questions and an earlier mode, so its scores cannot be compared directly.
01
Buyer question
“Can a buyer tell whether SIGNAL fits their website, what they will receive, and how to try it?”
02
Observed answer or gap
The assessment retained public-page evidence for SIGNAL’s purpose, inputs, no-plugin procedure and developer handoff. It did not establish complete continuation paths or independent corroboration. The five grounded searches found no target-site citation in this sample. Check these gaps against the cited pages before assigning work.
03
What to do next
Explain the required public URL, market, and buyer questions; show the prioritized task list as the deliverable. Link cost and integration answers to the right product section, and give buyers a direct product entry instead of another enquiry detour.
04
How to accept the result
Ask a colleague to find the required inputs, understand the cost factors, inspect the example, and open SIGNAL. Record missing answers and fix their destination pages. Measure indexing, visits, and enquiries separately; completing this journey does not establish an uplift.
Earlier assessment · September 9, 2026 · different questions and mode
WEAVIST AI assessed its own website with the SIGNAL engine. The run completed September 9, 2026 at 01:30:51 Asia/Taipei (September 8 UTC), using schema 0.3.3. This is a historical first-party example, not an independent customer result or a fresh assessment of the current site.
Input: one website and three declared questions
The URL was https://weavist.ai. The goal was to help visitors understand SPATIAL and SIGNAL, inspect demos, and contact the team. The agent supplied these English, US-market questions as audit inputs; they were not observed buyer demand.
- spatial AI platform for retail spaces that learns which experiences work
- AI search visibility audit tool for company website discoverability
- WEAVIST AI spatial SIGNAL demos
- Submitted paths: /about/, /signal/, /spatial/, /demos/. The assessment sampled the homepage and bounded supporting resources.
Result: accessible evidence, substantial unknowns
The 39 checks yielded 13 pass, 2 partial, 24 unknown, and 0 fail. The historical readiness scores were AIO 57.33, GEO 56.50, and SEO 57.33 out of 100, all with low confidence.
- The submitted paths were linked from the homepage and allowed by robots.txt.
- The sitemap contained 18 URLs; homepage metadata, canonical consistency, JSON-LD presence, and hreflang checks passed.
- Visible-text density and external identity coverage were partial. Corroboration and business-outcome evidence remained largely unknown.
Evidence: recorded responses and public source links
The engine completed 13 planned transport probes. Requests using five tested AI crawler user-agent strings received full HTTP 200 responses matching the control page. These were test requests, not observed visits by those crawlers.
- A separate OAI-SearchBot test also received HTTP 200 with content matching the control response; it was not added to the scored checks.
- The downloadable summary preserves selected findings, inputs, timestamps, and source URLs. It is an editorial extract, not a demo export feature.
- Source links open the current website. They help inspect the sources but do not recreate the historical response bytes.
Historical comparison: text density increased
Across the two September 9 runs, visible-text/HTML byte ratio moved from 4.3% to 5.7%, changing that check from fail to partial. This framework-sensitive heuristic does not measure indexability, ranking, or performance.
- The before run submitted two key paths; the after run submitted four. The input sets were not identical.
- No successful grounded-search observations informed either run. The after run did not execute grounded research.
- The comparison cannot establish better AI discovery, more citations, visitor growth, or conversion uplift.
The next decision
Use the remaining gaps to choose evidence to collect, rather than treating the score as an outcome.
- Review whether product pages answer fit, inputs, deliverables, cost, and evaluation questions.
- Verify external company references before treating identity links as corroboration.
- Agree a new assessment scope and separate measurement for indexing, citations, visits, or enquiries.
Limits of this example
This is a bounded public-site check of our own website. It does not validate every localized page, demo workflow, or structured-data claim.
- The target-engine list was an input, not proof that Google AI Overviews, AI Mode, ChatGPT, or Perplexity answers were observed.
- Unknown means the evidence was not established. Zero failed checks does not mean the website is fully verified.
- No customer results, ROI, ranking gains, or current indexing claims are inferred from this example.
Evidence and source pages
WEAVIST SIGNAL
Request a SIGNAL trial
Share your public website, use case, intended market and language, and any pages outside the assessment scope. You can also try SIGNAL directly at signal.weavist.ai.
Request a SIGNAL trialSend your public URL, target market, and the decision you need to make. Or start directly at signal.weavist.ai.