Skip to content

Latest commit

 

History

History
49 lines (39 loc) · 2.75 KB

File metadata and controls

49 lines (39 loc) · 2.75 KB

Quarterly Competitor Refresh Process (WS6-003)

Last reviewed: 2026-06-06 Cadence: Quarterly, or before any public positioning / comparison claim Owner: Strategy + documentation

Competitor products change quickly. Volatile facts (stars, pricing, model names, feature flags) must be source-dated or omitted. This process keeps TeaAgent's public comparisons honest and maintainable.

Artifacts to refresh

Artifact Path Action
Signal survey competitor-signal-survey-YYYY-MM-DD.md New dated file; supersede prior survey
Self-comparison matrix competitor-self-comparison-matrix-YYYY-MM-DD.md Update rows from official docs only
Landscape / positioning competitive-landscape-and-positioning-*.md Revise anti-personas and gaps if needed
Release checklist release-checklist.md Bump Last reviewed on quarterly section
When not to use when-not-to-use-teaagent.md Confirm non-fit scenarios still accurate

Refresh checklist

  1. Pick review date — Use ISO date in new survey filename (YYYY-MM-DD).
  2. Official sources only — Product docs, help centers, release notes. No forum rumors without labeling as anecdotal.
  3. Timestamp volatile metrics — GitHub stars, pricing tiers, model lists: include "as of DATE" or drop the metric.
  4. Run automation:
    python3 scripts/refresh_competitive_docs.py
    python3 scripts/validate_docs_consistency.py
    python3 scripts/generate_docs_inventory.py
    ./scripts/verify_docs.sh
  5. Supersession notes — Old surveys link forward; do not delete historical reasoning.
  6. Gap workflow — New competitor capabilities → signal-to-acceptance-gap.md.
  7. Record evidence — Update docs/release-evidence.json with refresh date and commit.

Source rules

Allowed Avoid
Vendor documentation URLs checked on review date Undated "X is better" claims
TeaAgent codebase + test references for our claims Copying competitor marketing superlatives
Explicit "not refreshed since DATE" on stale rows Implied parity without evidence

Current baseline (2026-06-06)