How we verify virtual receptionist pricing

Where the figures come from

One rule governs this site: a rate is only published here if it was read from the vendor’s own pricing page. Not from a comparison article, not from a sales email, not from a search result summary. Every row carries the source URL and the date it was checked.

That sounds obvious. It is not how most of this category works — a great deal of published “virtual receptionist pricing” is recycled between comparison sites, and figures stay in circulation for years after the vendor changed them.

What is in the table today

6 of 10 vendors in this table do not publish an overage rate. Overage is where most of the real cost sits, so a plan's headline price is not the price. Where a rate is missing above, it is because the vendor does not publish it — not because we could not find it.

What we do when a vendor does not publish a rate

We say so.

The temptation is to fill the gap with an estimate, a range, or a figure quoted somewhere else. We do not, because a table where some cells are measured and some are guessed is not a reference — and the reader cannot tell which is which.

Two things are worth knowing about the gaps themselves. First, they cluster around overage rates, which is where most of the real cost sits once a business exceeds its plan. A headline monthly price with an unpublished overage rate is not a price. Second, an unpublished rate is usually a deliberate choice rather than an oversight: it routes the buyer into a sales conversation where the number can be set per customer.

How often it is re-checked

Every row is re-verified quarterly, and the date on each row is the date it was last read.

This is enforced rather than intended. The site fails to build if any row’s verification date is more than 90 days old, so a stale table cannot be deployed — it stops being publishable before it stops being accurate. That check exists because silent staleness is the one failure that would destroy this page’s value: a table that is quietly wrong is worse than no table.

What we do not do

We do not rank vendors. There is no “best” column and no number one. Published research on AI Overview citations found that third-party lists captured 63% of citations across commercial software queries while the recommended product’s own site supplied 12% — a self-serving ranked list tends to be discounted precisely because it is self-serving. More simply: a ranking would require us to weigh vendors against each other on criteria that vary by business, and the honest answer is that the right choice depends on call volume and call length.

We do not report a rate we could not verify. Two vendors are currently absent from the table for this reason. One publishes rates only inside a PDF brochure; another blocks automated access to its pricing page. Both will be added when we can read their figures from the source rather than infer them.

We do not adjust figures to make them comparable. Vendors meter differently — some by the minute, some by the call, one in a proprietary unit where an AI-handled minute counts as half a human one. Normalizing those into a single “effective rate” would require assumptions about your call mix that we cannot make for you, so the table reports each vendor’s own unit and names it.

Corrections

If a figure here is wrong or out of date, it is a defect and we want to know. Vendors are welcome to ask us to re-check a row; the answer is always to go and read the published page again, not to accept a supplied number.