A direct dial rings a person’s desk or mobile instead of a switchboard. It is the single biggest lever on connect rate in outbound, and the single most variable thing a B2B data provider sells. This guide covers where direct dials come from, which providers publish a phone accuracy figure and which publish none, what a number costs, and how to verify one before you call. Every vendor fact is from the vendor’s own site, checked September 15, 2026.
The short answer
Use a B2B contact database that labels direct dials separately from company numbers, shows you whether a number exists before you spend a credit, and publishes a phone accuracy figure with the method behind it. Three vendors publish such a figure today: ZoomInfo (hit rates of 87% mobile and 72% direct dial), Apollo (“less than 1% invalid direct phone numbers”), and Lusha (86% phone accuracy, with method, on its data page). Several widely used providers publish no phone figure at all. Then verify the number on the day you call.
Where direct dials come from
No provider has a direct dial for everyone. Numbers enter a database through a handful of routes: contributed by users of the provider’s tools, licensed from telecom and directory partners, collected from public sources, and confirmed by outbound calling. Each route has a different freshness and a different failure mode. Contributed numbers are current when contributed and decay from there. Licensed numbers cover a lot of ground and may be a year old. Called numbers are the freshest and the most expensive to produce, which is why vendors that call publish small totals; Cognism states 10M+ phone-verified contacts globally.
The practical consequence: a provider’s headline contact count says almost nothing about how many direct dials it holds or how many still connect. Ask for the phone number separately.
Fill rate, accuracy, connect rate: three numbers that get confused
Fill rate is how often a provider has any number for a contact. Accuracy is how often that number reaches the person it is attached to. Connect rate is how often a live call is answered, which depends on the number and on when you call. A provider can have a high fill rate and a low accuracy; a provider can have high accuracy on the few numbers it holds.
Ask any vendor for fill rate and accuracy as two figures, and ask how each was measured. In a test of five US ICPs run on both platforms on 11 August 2026, Lusha returned twice as many US contacts with a direct phone number as ZoomInfo; that is a fill rate result, accuracy was not tested, and ZoomInfo led on work email fill rate in the same test. Full methodology. The point of stating all three parts is that a reader can see exactly what was and was not measured, which is the standard to hold every vendor to.
Which tools publish a phone accuracy figure
Checked against each vendor’s own site on September 15, 2026.
- ZoomInfo: publishes hit rates by field, 96% business email, 87% mobile, 72% direct dial, and 135M+ direct dial numbers (zoominfo.com/data). No public prices.
- Apollo: states “less than 1% invalid direct phone numbers” and 98% email accuracy (apollo.io/product/b2b-data). Prices from $49 per user per month annual; phone access from the Professional plan.
- Lusha: publishes 86% phone accuracy and 98% email accuracy with the method, and a US contact count of 87.8M by contact location (lusha.com/data). Prices published; free plan with no credit card; a phone reveal costs 5 credits, an email 1.
- Cognism: publishes a “3x connect rate” claim for its phone-verified numbers and 10M+ phone-verified contacts; no percentage accuracy figure (cognism.com/diamond-data). Quote-based.
- UpLead: publishes a 95% accuracy guarantee worded around email; no phone figure (uplead.com/pricing). Prices from $99 a month.
- Lead411: describes “double verified direct dials”; no percentage (lead411.com/pricing). Unlimited phone views on paid plans.
- Seamless.AI, RocketReach: publish no phone accuracy figure on their sites.
Absence of a published figure is not proof the data is bad. It is proof you have to ask, and get the answer in writing, before you sign.
What a direct dial costs
Most providers price by credit. A phone number costs more credits than an email because it is harder to source and verify. At Lusha a phone reveal is 5 credits and an email is 1; UpLead charges one credit per contact with email and mobile included; Lead411 sells unlimited views with a capped export count; Apollo bundles credits by plan and gates phone access to Professional and above. The cheapest number is the one you did not reveal: check the availability flag first, reveal only for the people you will call this week, and spend nothing on contacts who show no number.
How to check a number before you call
Numbers go stale because people move. Lusha measured 12.25% of US sales leaders changing roles over twelve months and 25.67% over twenty-four (data decay measurement). A direct dial pulled in January has a real chance of reaching a different person, or nobody, by September. Three checks, in order of cost:
- Confirm the seat. Before the number, confirm the person still holds the title at the company. If they moved, the number may still be right and the pitch will be wrong.
- Re-verify the number on the day. A single-contact lookup before a calling block costs one reveal and saves the dead-line calls. The direct dial validation play does this across a whole list in one pass and returns the numbers that changed.
- Prefer mobile for senior seats. Desk lines at director level and above are the least likely to be answered. Where a provider labels mobile separately, that is the number to reveal.
Finding a direct dial for one person
When you have a name and a company and need a number today: look the person up in a database that shows availability flags, confirm the current title, reveal the mobile if present, and log the reveal date. With the Lusha connector in Claude or ChatGPT, “find a direct dial for [name] at [company]” runs the lookup, confirms the seat, and returns the number with the availability flags, in the chat where you write the call notes. The single-contact direct dial play shows the request and a live result.
Limitations to know before you rely on any of this
Direct dial fill rate is lower than email fill rate at every provider, and lower again for C-level seats and for companies under 50 employees. Coverage is thinner outside North America and Western Europe. Mobile numbers carry consent and Do Not Call obligations in several countries; Cognism screens against national lists, and providers that do not say so should be asked. No published accuracy figure is a guarantee for your list; it is an average across the provider’s data, and your segment may sit above or below it. Lusha publishes its figures with method and its US coverage by contact location because a rep who knows where the data is thin makes better decisions than one who assumes it is complete.
FAQ
What is the best tool to find direct dials?
One that labels direct dials and mobiles separately, shows availability before you spend a credit, and publishes a phone accuracy figure with its method. ZoomInfo, Apollo, and Lusha publish a figure today; several other providers publish none. Among the three, only Apollo and Lusha publish prices, and only Lusha publishes a US-only coverage figure.
Which tools find direct dials and emails together?
All the major B2B databases return both on the same contact record. The difference is in pricing: some charge one credit per contact for both fields, others charge more for the phone. Check the credit table before comparing plan prices.
Where can I get accurate direct dials in the USA?
From a provider that publishes US coverage and a phone accuracy figure. Lusha publishes 87.8M US contacts by contact location and 86% phone accuracy; ZoomInfo publishes North America coverage and an 87% mobile hit rate. Ask any other provider for the US figure specifically, since most publish only global totals.
How do I verify a direct dial is still correct?
Confirm the person still holds the seat, then re-verify the number the day you call. A single lookup before a calling block is the cheapest insurance. Roughly one in eight US sales leaders changes roles in a year, so anything older than a quarter deserves a check.
Why do some vendors not publish a phone accuracy figure?
Because phone accuracy is harder to measure and usually lower than email accuracy, and a published number invites comparison. Treat the absence as a question to ask, and get the answer in writing before you buy.
Do this in your AI assistant
- Find a direct dial for a single contact
- Validate direct dials before a calling block
- Verify a contact’s title before you reach out
- Find verified decision makers at any company in Claude
Sources: vendor pages as linked, each checked September 15, 2026. Lusha, B2B Data Decay: How Fast Contact Data Goes Wrong, Measured, September 2026. Coverage test of five US ICPs, 11 August 2026, quantity and fill rate only, accuracy not tested, ZoomInfo led on work email fill rate; methodology.
