Transparent by design
How the TIP Score works.
Six pillars, fixed weights, and the rubric that turns evidence into a number. All of it published, so you can check our working.
The six pillars
Breadth and quality of what the tool can do today, measured against published benchmarks, feature depth and real-world output quality.
Price relative to delivered value, generosity of the free tier and predictability of costs at scale.
Security posture: compliance certifications (SOC 2, ISO 27001, GDPR), data-retention controls, enterprise controls and incident history.
Ecosystem fit: APIs, SDKs, plugin marketplaces and how easily the tool composes with an existing stack.
Production readiness: stability, documentation quality, support responsiveness and operating track record.
Trajectory: release cadence, adoption growth, community activity and vendor investment.
Fit score, TIP Score, adjusted score
Fit score
How well a tool matches the job you named, from a fit table we maintain by hand. It decides who's in the running: anything well below the best fit for that job gets dropped as not really built for it.
TIP Score
Overall product strength: the six weighted pillars above, sourced and refreshed.
Adjusted TIP Score
What the advisor ranks by, and the number next to each pick. It's the same six pillars, reweighted for your industry and your answers, with fit breaking ties. Choose nothing and it's just the TIP Score.
From evidence to score
Gather
Vendor docs, benchmarks, compliance reports, pricing pages, release notes.
Score pillars
Each pillar rated 0–100; a claim without a source doesn't count.
Apply weights
The fixed formula produces the TIP Score. Nothing manual after this step.
Re-score on evidence
New releases, pricing changes and security events feed straight back into the pillars.
Grade bands
| Score | Grade | Meaning |
|---|---|---|
| 90 – 100 | Exceptional | Category-defining. Safe default choice for most teams. |
| 82 – 89.9 | Excellent | Top-tier with minor trade-offs worth understanding. |
| 74 – 81.9 | Strong | Very capable, as long as its strengths line up with what you need. |
| 65 – 73.9 | Solid | Solid in its niche. Read the limitations before you commit. |
| < 65 | Developing | Promising but early; pilot before committing. |
Scoring rubric
The weights above explain the arithmetic. This is the judgment: what a given number on a pillar actually claims. Every pillar is scored against these anchors, interpolating where the evidence sits between them.
| Anchor | What it means |
|---|---|
| 0 | No demonstrated capability or material evidence. |
| 25 | Narrow or experimental, with serious limitations. |
| 50 | Usable, with meaningful constraints or inconsistent evidence. |
| 75 | Strong production capability with documented limitations. |
| 100 | Category-leading and broadly proven, on current high-confidence evidence. |
What each anchor means per pillar
Capability
- 0
- Nothing verifiable in this category.
- 25
- Works in demos, fails on real workloads.
- 50
- Handles real work with gaps that force workarounds on common tasks.
- 75
- Strong across the category's core tasks, with documented limits.
- 100
- Sets the benchmark others are measured against, on published results and real output quality.
Value for Money
- 0
- Cost cannot be established from public information.
- 25
- Priced well above what it delivers, or unpredictable at any scale.
- 50
- Fair for the output, but a thin free tier or costs that climb sharply with use.
- 75
- Clearly priced and predictable at scale, with an entry tier that does real work.
- 100
- Best-in-category outcome per dollar, transparent pricing, genuinely generous entry tier.
Security & Compliance
- 0
- No published security posture, or an unresolved material incident.
- 25
- Basic controls only; no recognised certification, little detail on data handling.
- 50
- Documented controls and at least one recognised certification, with gaps for regulated use.
- 75
- SOC 2 or ISO 27001 equivalent, clear retention and residency controls, enterprise admin features.
- 100
- Independently certified across major frameworks, customer-controlled retention and residency, public trust centre and disclosed incident history.
Integrations & Ecosystem
- 0
- No API, SDK or documented way to connect anything.
- 25
- A minimal API, few connectors, thin documentation.
- 50
- A workable API and the common connectors, with gaps that force custom work.
- 75
- Well-documented API and SDKs, with a connector catalogue covering mainstream stacks.
- 100
- Comprehensive API, first-party SDKs and a large third-party ecosystem; composes with an existing stack without custom glue.
Maturity & Reliability
- 0
- Pre-release or abandoned; no support and no operating record.
- 25
- Early product with frequent breaking changes, sparse docs, no support commitment.
- 50
- Production-usable by teams that can absorb occasional instability; docs cover the main paths.
- 75
- Stable releases, thorough documentation, responsive support, a multi-year track record.
- 100
- Proven at scale, with published reliability commitments and enterprise support.
Momentum
- 0
- No releases or visible activity in the last year; effectively dormant.
- 25
- Occasional maintenance releases; little adoption or community signal.
- 50
- Steady but unremarkable cadence, with flat adoption.
- 75
- Regular meaningful releases, growing adoption, active vendor investment.
- 100
- Rapid substantial releases, sharply growing adoption, heavy vendor investment.
Which source wins
A claim is scored on the strongest source that supports it.
- 1Primary vendor recordOfficial product documentation, pricing pages, release notes, trust centre, model card or regulatory filing.
- 2Independent measurementPublic benchmarks or original research with a published method.
- 3Reporting that shows its basisReputable journalism or analysis that links back to a primary source.
- 4Secondary commentaryUsed only where no primary evidence exists, and labelled as secondary on the profile.
Confidence
High
Every pillar rests on tier 1 or 2 sources, each checked inside its freshness window, with no unresolved conflicts.
Medium
Most pillars rest on primary sources, but some depend on tier 3 reporting or sit close to their re-check date.
Low
Material claims are unsourced, rest on tier 4 commentary, or are past their re-check window. Treat the score as indicative.
When evidence is missing, conflicting or old
| Situation | What happens to the score |
|---|---|
| A pillar has no verifiable evidence | It is not scored above 50, and the profile cannot be marked High confidence. |
| Two sources disagree | The stronger tier wins. Within the same tier we take the more conservative reading and note the conflict. |
| A claim is past its re-check window | It is shown as needing re-check and stops supporting an increase until re-verified; the profile drops to at most Medium confidence. |
| Evidence is too thin to judge the product | No TIP Score is published. The tool shows as 'New — not yet scored' and is kept out of rankings, comparisons and the advisor. |
How long a claim stays current
Where the catalog stands against this rubric
The rubric is the standard new and re-verified profiles are held to. The catalog predates it, so here is the honest current position, counted from the live data.
profiles carry a TIP Score
of claims link directly to their source
claims are past their re-check window
Evidence standards
- ◆Every pillar score must trace to verifiable sources: vendor documentation, compliance reports, public benchmarks, release notes or funding disclosures.
- ◆Scores carry a "last updated" date. Intelligence data is a curated snapshot, pricing and terms change quickly, so always confirm with the vendor before purchase decisions.
- ◆No vendor pays for placement or scores. Ranking order is purely the weighted formula above.