AI answer monitoring · bold-reports

Headline

No engines were queried in this run, so there is still no Bold Reports AI-visibility data for 2026-10. No MCP servers were connected, which means the openrouter ask_engine tool wasn't available. Google AI Overviews and Microsoft Copilot need a manual check, which an unattended run can't do. All 42 prompt × engine rows are marked not checked. This is the second run today with the same result: the published run 20261112-1519_ai-answer-monitoring_bold-reports_8605 also had no MCP servers connected. Rerunning won't help until openrouter is connected.

Score

None. This checklist has no score, so overall_score is null.

Top 3 actions

1. Connect the openrouter MCP server before the next pbmaintain run, then rerun this week. That covers ChatGPT, Gemini, Perplexity and Claude. Two runs on 2026-11-12 have failed for the same reason, so check the MCP configuration before rerunning.
2. Check Google AI Overviews and Microsoft Copilot by hand with the seven prompts in raw/prompt-set.json. Use a clean, signed-out session and record the country.
3. Log the first successful run as the Bold Reports baseline. Look closely at P5 (licensing) and P7 (security and compliance). Any wrong claim there goes to the product marketing lead and the COE Team the same day.

Findings

AI answer monitoring: bold-reports, 2026-10

Run by: Claude | Location: n/a (no engine queried) | Date: 2026-11-12

Prompts used (fixed set; unchanged from run 20261112-1519_..._8605):

• P1: "What are the best embedded reporting tools?"
• P2: "What is the best embedded reporting tool for an enterprise / midsize company?"
• P3: "Bold Reports vs Telerik Reporting: which should I choose?"
• P4: "What are the best alternatives to SSRS?"
• P5: "What is Bold Reports, how is it licensed and who uses it?"
• P6: "How do I migrate SSRS reports to another platform?"
• P7: "Is Bold Reports secure and compliant for enterprise use?"

Summary

• Mention rate: 0 of 0 answers collected (no comparison: the only earlier run, 20261112-1519_..._8605, also collected no answers)
• Citation rate: 0 of 0 answers collected (no comparison, for the same reason)
• Wrong or outdated facts: none recorded because no answers were collected. This is unknown, not zero.

Gaps and fixes

No gaps could be found because no answers were collected.

Not checked

• P1–P7 on ChatGPT, Gemini and Perplexity (21 rows): no MCP servers were connected for this run, so the openrouter ask_engine tool wasn't available.
• P1–P7 on Claude (7 rows): openrouter wasn't connected. I didn't answer from this agent session because it isn't clean: it has the repo's Bold Reports positioning in context, so its answers wouldn't be comparable month to month.
• P1–P7 on Google AI Overviews and Microsoft Copilot (14 rows): these have no API, and an unattended run can't check them by hand.
• Phase 2 (month-over-month comparison, gap list, hand-off to the SEO/GEO and blog programs, weekly GTM review deltas): skipped because there is no data this run and no earlier run with data.

Proposed changes (need approval)

None. This run made no changes to live systems and proposes none. It's not a live-system change, but the operator should check why MCP servers aren't attached to pbmaintain runs (maintain/mcp-servers.json has uncommitted local changes) before the next attempt.

Data sources

• Fixed prompt set and category: demand-gen/search/ai-answer-monitoring.md.
• P3 and P6 prompts and competitors: products/bold-reports/gtm-plan.md. I reused them exactly as resolved in the earlier run, so the results stay comparable.
• Earlier run: reports/publish/bold-reports/ai-answer-monitoring/20261112-1519_ai-answer-monitoring_bold-reports_8605/ (no answers collected).
• Engines: none queried (no MCP servers connected).
• Exports: raw/prompt-set.json (the resolved prompts; no answers).
• Choices made without an operator:
  • Kept the earlier run's choices: P1–P2 use "embedded reporting tools", P3 compares against Telerik Reporting, and P4 asks for alternatives to SSRS.
  • The metrics period uses the run's date_range (2026-10-15 to 2026-11-11), and the logged month is 2026-10. Metric values are null (not measured), not zero.
  • run.json overall_score was already null, which is correct for an unscored checklist, so I left it unchanged.