One formula, the same for all 14 tools. It reads what each vendor publishes about its entry plan, the cheapest plan you can pay for month to month, and turns it into a score out of 10. Data read on November 14, 2026.
Our reviews and scores are built from what each vendor publishes: pricing pages, plan comparison tables, documentation and product pages, read on the date shown on each page.
What are the criteria and their weights?
| Criterion | Weight | What we read |
|---|---|---|
| Engines covered | 20% | How many of the six reference engines the entry plan tracks at once |
| Entry price | 15% | Cheapest paid plan, billed monthly |
| Price per 1,000 answers | 15% | Entry price divided by the AI answers it collects in a month |
| Refresh schedule | 10% | Hourly, daily or weekly checks |
| Insight depth | 15% | Citations and sources, sentiment, competitor benchmarking, recommendations: 2.5 points each when included in the entry plan |
| Data access | 10% | API (4 on the entry plan, 2 as an add-on or on a higher plan), CSV export (3), MCP or BI connector (3) |
| Try before you pay | 10% | Free plan, free trial and its length |
| Seats | 5% | Users included in the entry plan |
Each criterion gives 0 to 10 points. The score before scaling is the weighted average. When a criterion cannot be read (a custom plan has no price per answer, for example), it is removed and the other weights are scaled up to total 100%.
How are the points given?
- Engines covered. We count how many of six reference engines the entry plan tracks at the same time: ChatGPT, Perplexity, Gemini, Google's AI answers (AI Overviews or AI Mode), Claude and Microsoft Copilot. Points: 1 → 3 · 2 → 5 · 3 → 6.5 · 4 → 8 · 5 → 9 · 6 → 10. A plan that lists eight engines but runs three at once counts as three.
- Entry price, in dollars a month, billed monthly: 30 → 10 · 60 → 9 · 100 → 8 · 200 → 6.5 · 300 → 5 · 500 → 3, with straight lines in between. No public price: 2 points.
- Price per 1,000 answers. Entry price divided by the AI answers the plan collects in a month, times 1,000. In dollars: 8 → 10 · 15 → 9 · 25 → 8 · 50 → 6 · 90 → 4 · 140 → 2.5. When a vendor gives prompts and a daily schedule but no answer count, we use the minimum (prompts × 30).
- Refresh schedule. Hourly 10, daily 9, weekly 6; 7 when the vendor meters responses or credits without stating a schedule.
- Insight depth. 2.5 points each for citation and source tracking, sentiment analysis, competitor benchmarking and recommendations, when they are in the entry plan.
- Data access. API: 4 points on the entry plan, 2 as an add-on or on a higher plan. CSV export: 3. MCP or a BI connector: 3.
- Try before you pay. Free plan 10, trial of 14 days or more 9, 7 to 13 days 8, trial of unstated length 7, no trial 3.
- Seats. Unlimited 10; otherwise 1 → 3 · 2 → 5 · 5 → 7 · 10 → 9.
How does the calculation become the score you see?
Displayed score = 4.4 + 0.57 × weighted average, rounded to one decimal and kept between 7.5 and 9.6. The scale is the same for every tool, so it never changes the order.
Tie-break: two tools never show the same score
When two tools would print the same score, the one with the cheaper price per 1,000 answers keeps it (then the lower entry price, then the longer trial) and the next one is lowered by 0.1. Today this applies to SE Ranking, Promptwatch, LLMrefs, Nightwatch, Profound, Scrunch, Evertune.
Prices in another currency
A price is always shown in the currency the vendor lists. For the two price criteria only, a euro price is converted with the European Central Bank reference rate (1.1052 dollars per euro, Nov 13, 2026).
Why does the order change in the rankings by need?
The rankings by need use the same criteria with other weights. Only the order changes: the score next to a tool is always its overall score.
| Ranking | Engines covered | Entry price | Price per 1,000 answers | Refresh schedule | Insight depth | Data access | Try before you pay | Seats |
|---|---|---|---|---|---|---|---|---|
| Overall | 20% | 15% | 15% | 10% | 15% | 10% | 10% | 5% |
| Answer engine optimization | 15% | 10% | 10% | 5% | 35% | 10% | 10% | 5% |
| LLM tracking | 30% | 10% | 20% | 20% | 10% | 5% | 5% | 0% |
| Agencies and white label | 15% | 10% | 20% | 5% | 10% | 20% | 5% | 15% |
| Free and low cost | 15% | 30% | 15% | 5% | 10% | 5% | 20% | 0% |
Does money change a rank?
No. Some links on this site may be affiliate links; the formula does not know which ones. A tool with no program can be No. 1, and today's No. 1 is Rankscale at 9.5/10 because of what its entry plan includes. We sell no visibility tool of our own. See the affiliate disclosure.
