How we evaluate every product.
A published 10-criteria framework, weighted per category, traceable per product. No black box. No vibes. Updated quarterly.
Review Criteria
What goes into a review.
Each review is the output of a structured evaluation. We test the product hands-on across a set of category-defined workloads, conduct interviews with at least three reference customers (where available), audit the vendor's published documentation and security disclosures, and benchmark against competing products in the same category.
The narrative review is signed by a named analyst. The score is computed from the weighted dimensions below. Every numeric claim is cross-referenced to the test, the interview, or the document that produced it.
Scoring Framework
10 dimensions. Published weights.
The overall score is a weighted average of 10 per-dimension scores (0–10). Weights are category-tuned (e.g. Security weighs heavier in healthcare-regulated verticals than in marketing tools), but the default weighting below applies unless a category-specific override is published in the article foot.
How We Test
Hands-on, sourced, signed.
Every review applied to this site follows the same hands-on testing protocol — no exceptions, including for products owned by our parent company.
- Trial access. Analyst access via vendor-provided trial, public demo, or standard customer onboarding. No special VIP access from any vendor. Minimum 2-week active evaluation per product before scoring.
- Category-defined workloads. Each category has a documented set of reference workloads (e.g. for KM: 5,000-article migration · 500-user onboarding · ServiceNow integration test · agent search-relevance suite). Every product in the category runs the same workload set.
- Customer interviews. Minimum 3 reference customers per product (when available). Conversations cover deployment timeline, hidden costs, integration friction, support quality, and would-buy-again signal.
- Documentation + security audit. Vendor docs, API references, security disclosures (SOC 2 Type II, ISO 27001, HIPAA BAA availability), compliance posture, and incident history reviewed against the same checklist.
- Score computation. Per-dimension scores assigned with evidence citations. Composite score is a weighted average using category-tuned weights documented above. No score adjustment after evidence is locked.
- Analyst sign-off + editor review. Named analyst signs the score on the record. Editor-in-chief reviews for evidence completeness, balance (Pros + Cons), and methodology adherence before publication.
- Quarterly re-scoring. Every product re-evaluated on the same workload set at next-review-date. Score changes documented as evidence-of-change, not opinion drift.
Editorial Independence
No vendor pays for placement.
Vendors do not pay for inclusion, score adjustment, sponsored placement, or removal. Affiliate relationships, partner programs, and sponsorships — when they exist — are disclosed at the foot of the affected article. We do not accept payment for review consideration, payment for prioritized publishing, or payment for early access at the cost of editorial control.
When a reviewer has a personal or commercial conflict of interest, the conflict is disclosed in-line on the review and an independent secondary reviewer co-signs the score. If the conflict is material to the recommendation, the review is reassigned.
Upland ownership — the structural disclosure
Enterprise Software Review is an editorial property owned by Upland Inc. (NASDAQ: UPLD). Several products covered in our reviews — RightAnswers, Panviva, Qvidian, InterFAX, BA Insight, PowerSteering, Cimpl, RO Innovation, Upland PSA — are part of the Upland portfolio. We acknowledge this is a structural conflict of interest. The mitigations are specific:
- Same rubric, weights, and 15-section template apply to Upland products and competitors alike. Weights are fixed before any vendor scoring.
- Per-criterion scores are tied to specific evidence. When a competitor leads on a criterion, that result is published.
- Analysts report to the editor-in-chief, not Upland commercial leadership.
- For every Upland product reviewed, we publish reviews of at least 2 direct competitors in the same category with the same depth.
- When a per-dimension average and the displayed rank disagree (e.g. on listicles), the article includes an explicit "How We Chose" section explaining the category-specific weighting.
Read the full Editorial Independence & Conflict of Interest Policy for the complete control list and how to audit our work.
"We score software the way critics score books — methodically, signed, and on the record. The vendor doesn't get a vote."
— From the masthead
Update Cadence
Reviewed every quarter.
Every published review carries a Next review date in its frontmatter. The default cadence is quarterly (90-day cycle), with category-specific exceptions for fast-moving spaces (AI search, generative tools — re-reviewed monthly) and slow-moving spaces (ITFM, TEM — re-reviewed semi-annually).
Out-of-cycle re-reviews are triggered by major product releases, significant security incidents, ownership changes, or material customer feedback. Re-reviews retain the same dimensional structure and are signed by the same analyst when possible.
Read a review applying this framework.
Browse the full library of 14+ reviews scored against this rubric.