Methodology
A fixed model, not a generated opinion.
Every GuruLynk recommendation comes from the same weighted calculation. Run the audit twice with the same answers and you get the same result.
The GuruLynk Match Score
Each product is scored 0–100 per dimension, then combined with these fixed weights.
| Dimension | Weight | What it measures |
|---|---|---|
| Business Fit | 25% | How well the product serves your business model and the segments it was designed around. |
| Goal Fit | 20% | Coverage of the specific outcomes you selected, measured against the product's supported goals. |
| Team Size Fit | 15% | Whether the product performs well at your headcount, penalised by distance from its proven range. |
| Budget Fit | 15% | Price level compared with your stated monthly software spend. |
| Ease of Use | 10% | Time to competence for a non-specialist operator. |
| Integrations | 10% | Breadth and quality of connections to the rest of a typical stack. |
| User Preference Fit | 5% | Your stated decision factors, such as price, AI capability or scalability. |
Category status: keep, optimize, replace, add
After scoring, each category receives one status. KEEP means your current tool is the strongest match for your profile. OPTIMIZE means the tool works but the setup is inefficient, usually overlap between products. REPLACE means a materially better fit exists — at least a 12-point difference. ADD means the category is missing and your answers indicate you need it.
Independence
GuruLynk is capable of recommending a tool that has no commercial programme at all, and does so regularly. Affiliate availability is stored as a separate data field and is never an input to the match score. It cannot raise, lower, or reorder a recommendation.
Affiliate disclosure philosophy
When affiliate relationships are added in a future version, they will be disclosed on every page where a monetised link appears, and the scoring model will remain unchanged. If a recommendation would change because of a commercial relationship, we consider that a defect, not a business model.
Data and verification
- Product attributes are reviewed and dated; each record carries a last-verified date.
- Pricing bands are stored as levels rather than exact prices, which change frequently.
- Scores are recalculated whenever attributes change, not manually adjusted.
Limitations we state openly
- The catalogue covers a curated set of products, not every tool on the market.
- Savings ranges are estimates based on typical stack overlap, not guarantees.
- The audit reflects what you tell us; incomplete answers produce a less precise result.