How we score AI receptionists
We read each brand's official pricing and product pages, record the published plans and dated US prices, and turn them into one score out of 10 with the formula below. The same formula applies to every service, and the No. 1 of each ranking is simply the highest score.
The reference month
Plans are billed by the minute, by the call, by the unique caller or by usage, so we price every service on the same month: 200 minutes, which is about 80 calls at 2.5 minutes each. For each plan we add the published overage to the included volume and keep the cheapest way to cover the month. Per-call and per-caller prices count one caller per call (the least favorable case for per-caller billing). A plan without a published overage rate only counts if it already covers the month. Catalog medians today: $96.20 for the month, $79 for the entry price.
Criteria and weights
| Criterion | Weight | Scale (value → score /10) |
|---|---|---|
| Cost of a 200-minute month | 25% | Ratio to the catalog median: 0 → 10, 0.5 → 9, 1 → 6.5, 2 → 3.5, 4 → 1.5 |
| Entry price | 10% | Lowest paid monthly plan, ratio to the median: 0 → 10, 0.5 → 9, 1 → 6.5, 2 → 3.5, 4 → 1.5 |
| Price of an extra minute | 10% | $ per extra minute: 0.1 → 10, 0.25 → 8, 0.5 → 5, 1 → 2; "moved to the next plan" scores 3 |
| Appointment booking | 12% | native 10, higher plan 5, build 3, none 0 |
| Call transfer | 10% | warm 10, live 8, blind 6, higher plan 4, build 3, none 0 |
| Languages | 8% | Languages stated: 1 → 2, 2 → 5, 10 → 7.5, 30 → 9, 50 → 10 |
| Free trial or free plan | 8% | Free plan 10, trial of 14+ days 9, 7 days 7, free credits 6, demo agent 5, demo call only 2 |
| Setup effort | 10% | self serve 10, guided 7, developer 3 |
| Integrations | 7% | native 10, crm api 9, zapier 7, api 5 (crm api = CRM via API + Zapier) |
Booking and transfer are read on the entry plan: a feature that only exists on a higher tier scores "higher plan". A criterion the brand doesn't state on its pages scores 2/10: we can't credit what isn't published. No score without a published price or with fewer than 5 stated criteria.
From raw score to displayed score
The weighted average (raw score, 0 to 10) is shown on our editorial scale: displayed score = 7.5 + 0.42 × (raw − 4), kept between 7.5 and 9.6. Every service compared here is a serious current option, and the order of raw scores is preserved. Labels: Excellent from 9.0, Very good from 8.5, Good from 8.0, Fair from 0.0.
Rankings by need
Each ranking uses the same criterion values with different weights (each set totals 100). The weights only decide the order of that ranking: the score displayed next to a service is always its overall score, the same on every page of the site.
| Criterion | Overall | Small-business | Budget | Answering-service | Medical-office | Home-services | Law-firm |
|---|---|---|---|---|---|---|---|
| Cost of a 200-minute month | 25 | 30 | 35 | 25 | 20 | 25 | 20 |
| Entry price | 10 | 15 | 30 | 0 | 0 | 0 | 0 |
| Price of an extra minute | 10 | 10 | 15 | 10 | 5 | 10 | 5 |
| Appointment booking | 12 | 12 | 5 | 10 | 20 | 20 | 15 |
| Call transfer | 10 | 8 | 0 | 20 | 15 | 15 | 20 |
| Languages | 8 | 5 | 0 | 10 | 10 | 10 | 10 |
| Free trial or free plan | 8 | 8 | 10 | 8 | 5 | 5 | 5 |
| Setup effort | 10 | 10 | 5 | 10 | 10 | 10 | 10 |
| Integrations | 7 | 2 | 0 | 7 | 15 | 5 | 15 |
Sources and dates
Prices were checked on November 11, 2026 on each brand's own pages, linked in every review. We note exactly what each brand states and attribute claims ("Brand says…"). One sentence on our method: we compare published plans, prices and features; we don't place test calls or run the services ourselves.
12 services are compared. Synthflow was left out: its pricing page only offers annual enterprise contracts on quote, with no published price.