What Data to Store in a CRM (and What to Leave Out)

The data to store in a CRM is the data you will act on: who the customer is, how to reach them, who owns the relationship, where they are in their lifecycle, what they have bought, and what was last said or promised. A good rule is that every field should have an owner and a use. If nobody reads a field when deciding what to do next, it adds effort and clutter without adding value.

This guide gives you a field list to start from, rules for keeping the data clean, and a short section on data minimisation, which is both good practice and, in many places, a legal expectation.

The principle: every field needs a job

Before you add a field, answer three questions:

1. Who fills it in? A person, an import, or an integration?
2. When is it filled in? At first contact, after a sale, after each call?
3. What decision does it support? A follow-up, a segment for an email, a report?

If you cannot answer the third question, leave the field out. You can always add it later. Removing a field that half your records have filled in is much harder.

Core contact fields

These are the fields almost every small business needs on a person record.

Field	Example	Why it matters
First and last name	Maria Lopez	Personal communication, duplicate detection
Email	[email protected]	Primary channel; also the best unique key
Phone	+1 555 0100	Calls, SMS where permitted
Company (linked)	Lopez Interiors	Groups people at the same customer
Role / job title	Office manager	Know who decides and who uses
Owner	Sam (account manager)	One person responsible for next steps
Lifecycle stage	Lead, customer, lapsed	Drives which communication fits
Source	Referral, website, trade fair	Shows where good customers come from
Created date	2026-03-14	Cohorts, response times
Last contact date	2026-09-28	Spot silent accounts
Communication preferences	Email yes, phone no	Respect choices and consent

Use a phone format that includes the country code. Use a fixed list (a dropdown) for lifecycle stage and source, never free text, or your reports will contain "Referral", "referral" and "Ref." as three different sources.

Company fields (for B2B)

If you sell to businesses, store company-level information on a company record, not on each contact:

• Company name and website
• Industry (fixed list)
• Size band (for example 1–9, 10–49, 50–249 employees)
• Billing address and country
• Customer since (date of first purchase)
• Account owner
• Contract or renewal date, if you have contracts

Size bands are better than exact numbers because nobody keeps exact headcounts current, and a band is all you need for segmenting.

Purchase and relationship fields

These turn a contact list into something you can use for retention work:

Field	Type	Use
First purchase date	Date	Cohort analysis, anniversaries
Last purchase date	Date	Lapsed-customer segments, win-back timing
Number of orders	Number	Frequency, loyalty segments
Total revenue to date	Currency	Prioritise accounts
Main product or service	Fixed list	Relevant cross-sell, onboarding content
Satisfaction (latest score)	Number	Follow up on low scores
Open issues	Linked tickets or a yes/no flag	Do not send a sales email during a complaint

With first and last purchase dates and order counts, you can calculate the inputs for customer lifetime value and build simple segments such as "bought twice, nothing in the last six months". If you sync these fields from your shop or accounting system rather than typing them, they stay accurate.

Activity data: what to log

Interactions are where CRMs become cluttered. Log the things a colleague would need to pick up the relationship without asking you:

• Meetings and calls with a two-line summary: what was discussed, what was agreed.
• Promises and next steps as tasks with a due date, not buried in notes.
• Quotes and proposals with value and status.
• Complaints and their resolution (see how to handle customer complaints).
• Important preferences such as "prefers calls in the morning" or "invoices to accounts department".

A summary note template keeps notes short and scannable:

Date / channel: 2026-09-28, phone Who: Maria Lopez (office manager) Summary: Happy with the last delivery. Considering a second location next spring. Next step: Send brochure for the multi-site package by Oct 5 (owner: Sam).

What to leave out

Some data creates risk without adding value. As a general rule, avoid storing:

• Sensitive personal data that has nothing to do with the business relationship, such as health, religion or political views. Many data protection laws treat this kind of data with special strictness.
• Payment card numbers and passwords. These belong in dedicated, secured payment systems, never in notes.
• Opinions about the person you would not want them to read. In many jurisdictions customers can ask to see the data you hold on them.
• Data "just in case". Birthdays, hobbies or family details only make sense if you have a concrete, welcome use for them, and the customer has shared them for that purpose.

Data protection rules differ by country and change over time. Treat this section as general guidance, and check your specific obligations (for example consent for marketing, retention periods and data processing agreements with your CRM provider) with a qualified professional.

Data hygiene: rules that keep a CRM usable

Clean data is mostly about prevention. Agree on a handful of rules and write them down.

Entry rules

1. Search before you create. Check by email and company name before adding a new contact.
2. Use fixed lists for stage, source, industry and country.
3. One owner per record. Unowned records are nobody's job.
4. Minimal required fields. Name, email or phone, owner, stage. Everything else optional.
5. Dates in one format, ideally set by the system rather than typed.

Monthly clean-up (30 minutes)

• Merge duplicates flagged by the system
• Fix or archive contacts with bounced emails
• Assign an owner to records without one
• Close or update deals with no activity for 60 days
• Check new imports for odd values

Twice a year

• Review which fields are actually filled in and used; remove or hide the rest
• Archive or delete contacts you no longer have a reason to keep, according to your retention policy
• Review user access and remove people who have left

A worked example: trimming a bloated CRM

Say a small agency has 2,400 contacts and 46 custom fields. A quick audit shows:

Finding	Count
Fields filled in on fewer than 10% of records	19
Fields nobody could name a use for	11
Contacts with no activity in over 3 years	900
Likely duplicates (same email)	140

The agency hides the 19 rarely used fields (after exporting their data), deletes the 11 unused ones, merges 140 duplicates and, after checking its retention policy, archives the 900 dormant contacts. The CRM goes from 46 to 16 fields and from 2,400 to about 1,360 active contacts. Reports become faster to read, and the team stops skipping fields because the form is short enough to fill in properly.

The numbers are illustrative, but the pattern is very common: a smaller, cleaner CRM gets used more.

How this fits into choosing or switching a CRM

If you are still choosing a system, take your field list into the trial and check that the CRM supports fixed lists, duplicate detection and clean export. These are listed as must-haves in our guide on how to choose a CRM. If you are still on a spreadsheet, the same field list makes a good column structure; see spreadsheet vs CRM for a template.

Your next step

Export your current contact list and, for each column, write down who fills it in and what decision it supports. Anything without an answer goes on the "remove or hide" list. Then write your five entry rules on a single page and share it with everyone who touches customer data.