Most CRMs are full of names and emails but missing the phone numbers that actually matter, especially direct lines and cell numbers. This guide looks at why phone data decays so quickly and how extracting numbers directly from email signatures keeps contact records current.
This page focuses on extracting phone numbers from email signatures. For a complete guide to contact extraction across Outlook, Gmail, calendar meetings, and more, see Email Address Extractor: Complete Guide to Extracting Email Addresses from Outlook, Gmail & More.
Phone numbers tend to be the field most likely to be missing, wrong, or outdated in a typical B2B contact database, for a few specific reasons.
They're rarely captured at the point of first contact. A rep receives an email from a prospect, saves the email address, and moves on, while the direct dial sitting in the signature goes unnoticed.
They also change more often than people realize. When someone moves to a new company or role, their direct line usually changes with it. Third-party databases tend to update these fields periodically rather than continuously, so the gap between a database and reality tends to widen over time.
Cell numbers are the least likely of all to end up in a CRM. Direct office lines occasionally get logged, but cell numbers, often the most useful number for actually reaching someone, almost never do unless a rep manually types one in after a phone call.
Email signatures address all three of these gaps. Contacts tend to include the numbers they actually want to be reached on, and they update their signatures when those numbers change.
Phone numbers in email signatures follow no standard format. Across a large volume of email, you'll see wide variation in how contacts label and format their numbers:
A rule-based approach that only looks for fixed patterns tends to break down against this level of variation. Reliable extraction generally depends on models trained on real-world signature data that can identify and correctly label phone numbers regardless of format.
Many extraction approaches capture whatever number appears in a signature without distinguishing type. That distinction matters in practice.
A direct line connects a caller straight to the contact without going through a receptionist or switchboard, which makes it considerably more useful than a general company number for reaching a specific decision-maker. Cell numbers are often even more valuable, since they reach a contact wherever they are, outside office hours, and are increasingly the primary number for senior professionals who work remotely or travel often.
Storing direct and mobile numbers as separate fields, rather than lumping them together, gives a sales team a fuller picture of how to actually reach each contact, not just confirmation that a phone number exists somewhere in the record.
For most businesses, the immediate value of phone number extraction isn't building a brand-new contact list from scratch. It's filling gaps in the one they already have.
A common scenario looks like this: a CRM contains thousands of contacts, most of them originally captured with just a name and email address at the point of first interaction. Years of subsequent email correspondence have generated dozens of signatures containing direct dials and cell numbers, none of which ever made it into the CRM. Processing that history and matching extracted phone numbers back to existing contact records can fill in those gaps automatically, often flagging the most recently seen number for each type as primary while retaining older numbers as history.
The general process starts with connecting to Gmail, Outlook, or Office 365 through secure OAuth, which avoids sharing or storing passwords. From there, historical email, sometimes going back up to ten years, gets scanned, and every signature encountered is processed. Phone numbers are extracted, labeled by type, and formatted consistently.
Where a contact has different numbers across different emails, such as an old direct line from a previous role and a current cell number from a more recent exchange, a well-built system captures all of them, flags the most recently seen number for each type as primary, and retains the full history on the contact record.
Enriched records can typically be exported to CSV or Excel, pushed into a CRM like Salesforce, HubSpot, or Microsoft Dynamics, synced to an email platform like Mailchimp or Constant Contact, or accessed through an API for custom integrations.
Third-party vendors are generally useful for finding phone numbers on net-new prospects, meaning contacts a team has never interacted with before. For that specific use case, they fill a role that signature extraction doesn't.
Where vendor databases tend to fall short is enriching relationships a team already has. They rarely include cell numbers, they update periodically rather than continuously, and they have no visibility into the specific people a given team actually corresponds with. Signature-based extraction addresses that gap by working from communications that have already happened, which tends to produce numbers that are more current and more specific to the relationships a business actually has.
Yes, when the signature provides enough context to differentiate. Direct or office numbers and mobile or cell numbers are typically stored as distinct fields on the contact record.
This is one of the most common scenarios in practice: an old direct line from a previous role, a cell number from an early exchange, and a new direct dial from a current position. A good extraction system captures all of them, flags the most recently seen number for each type as primary, and retains the full history, so you have the most current number without losing historical context.
International formats, including country codes, local formatting conventions, and extensions, can generally be captured as they appear in signatures and standardized for consistency across records.
Yes, when the system is built to detect this. If a new phone number appears in a contact's signature, for example after a job change, the contact record updates while the previous number is retained in history.
Fax numbers can be captured where present and stored as a separate field, flagged distinctly from direct lines and mobile numbers so they don't mix in with primary phone data.
Whatever is available gets extracted, and the phone field is simply left blank where no number is present. The contact record can still be created from email header data, and other enrichment sources, like LinkedIn, can help supplement missing fields where a profile is available.
Once phone data has been extracted, it typically supports a few common workflows: updating a CRM so sales teams always have current contact details, prioritizing direct outreach by combining phone data with engagement frequency to identify high-value contacts, and building complete contact profiles by pairing phone numbers with job titles, LinkedIn profiles, and location data.
SigParser extracts phone numbers from email signatures across Outlook, Gmail, and Office 365 history automatically, labeling direct and mobile numbers separately and keeping records current as signatures change. SigParser scans 90 days of email history at no cost if you want to see how many phone numbers are already sitting in your team's signatures.
For the broader picture of contact extraction, including calendar data and full inbox scanning, see Email Address Extractor: Complete Guide to Extracting Email Addresses from Outlook, Gmail & More.
SigParser can securely scan one to thousands of employee mailboxes to extract email addresses and more. Get a demo to learn more.