← All posts

How to Import Customer Data Without the Mess

A customer data import can either give your team a clean start or quietly create months of admin. The difference is rarely the spreadsheet itself. It is the decisions made before you upload it. Knowing how to import customer data means knowing what belongs in your CRM, what does not, and how each record will help someone take the next useful action.

That matters when you are moving on from scattered spreadsheets, inboxes and notes held in people’s heads. A CRM should make connections clearer. It cannot do that if it begins with duplicate contacts, missing company names and deals that nobody owns.

Start with the job your data needs to do

Do not import every field simply because it exists. Start by asking what your team needs to see when speaking to a prospect, managing an account or reviewing the pipeline.

For most B2B teams, the essentials are straightforward: company name, contact name, role, work email, phone number where relevant, account owner, relationship status and useful context from previous conversations. If you are importing opportunities, you will also need a deal name, value, expected close date and pipeline stage.

The right level of detail depends on your business. An agency may need service interests and renewal dates. A consultancy may need buying committee roles, referral sources and contract values. A startup with a short sales cycle may need far less.

Keep one rule in mind: if a field will not inform a decision, a conversation or a report, do not bring it across yet. Empty fields make a CRM feel complicated. Useful fields make it feel organised.

Audit the source before you import customer data

Most customer data lives in more than one place. There may be a master spreadsheet, a previous CRM export, a finance system, event sign-up lists and individual sales trackers. Treat none of them as perfect.

Choose a primary source for each type of record. For example, use your existing CRM export for contacts and companies, and your current pipeline tracker for open deals. This avoids merging conflicting versions of the same information without a clear rule for which version wins.

Then review the file carefully. Look for obvious problems: rows with no company or contact name, emails in the wrong column, merged cells, inconsistent dates and old test records. Remove people who should not be contacted and records that have no business purpose.

This is also the moment to agree on a cut-off point. If one team member continues adding leads to an old spreadsheet while another prepares the import, your data is out of date before it arrives. Set a clear date and decide where new records will be added from then on.

Clean the details people rely on

Good data is not about perfection. It is about consistency. A sales colleague should not have to guess whether “Acme Ltd”, “ACME” and “Acme Limited” are the same account.

Standardise company names, job titles, country names and date formats. Use a consistent format for telephone numbers. Check that email addresses are in their own field rather than mixed into notes. Split full names into first and last names if your CRM uses separate fields.

Pay particular attention to labels such as lead source, industry, customer type and lifecycle stage. “Referral”, “referal”, “Partner referral” and “Introduced” may all mean the same thing, but they will fragment reporting. Pick one approved value for each category before importing.

Free-text notes need more judgement. A concise note that explains a live relationship can be valuable. A decade of copied email threads is usually not. Bring across context that helps the next owner understand the account, not historical clutter that nobody will read.

Decide how records will connect

Customer data is more useful when the relationships are intact. Contacts should sit under the right company. Deals should be linked to both the right company and the right primary contact. Activities should be associated with the record that gives them meaning.

To make this work, identify a reliable unique value for each record. Email address is often the best identifier for an individual contact. For companies, use a consistent company name or an existing account ID where you have one. Do not rely on a contact’s name alone. There are plenty of David Smiths in B2B sales.

Duplicates are inevitable, especially when several colleagues have maintained their own lists. Merge them before importing where possible. If two records are genuinely different, make that clear with details such as company, email or location. Guessing later costs more than checking now.

A little caution is useful here. Two contacts can share an email inbox. One company can have several trading names. Parent companies and subsidiaries may need separate records. The cleanest structure is the one that reflects how your team actually sells and supports customers.

Map fields with purpose

Field mapping is where your spreadsheet columns are matched to the fields in your CRM. It sounds technical. It is mostly common sense.

Match standard information to standard fields wherever possible. First name belongs in first name. Email belongs in email. Deal value belongs in deal value. Avoid putting core information into a generic notes field just because it is convenient. That makes filtering, reporting and handovers harder.

Create custom fields only when a detail is genuinely specific to your business. For example, a recruitment consultancy may need a field for hiring volume, while a design agency may need project type. A custom field should answer a question your team asks repeatedly.

Before you upload, make a simple mapping sheet. It should show the source column, the CRM field it will populate, the expected format and any transformation required. For example, a column labelled “Potential” may need to become a numerical deal value, while “Hot”, “Warm” and “Cold” may need to map to a defined qualification status.

This document is not bureaucracy. It gives everyone one shared answer when questions arise later.

Test a small import first

Never make the first import the full import. Start with a small, representative sample: a few companies, several contacts, a couple of deals and records with edge cases such as incomplete data or shared domains.

Review the result in the CRM as your team will use it. Can you find a company quickly? Are contacts connected correctly? Do deal values display as expected? Are dates understood correctly? Can you filter a group of records by owner, stage or source?

Ask someone who did not prepare the file to check it too. Fresh eyes spot mistakes quickly. They may notice that job titles have landed in the company field or that a stage called “Proposal sent” has been imported as a closed deal.

Once the sample looks right, proceed in sensible batches. Start with companies and contacts, then add deals and other related information. Keep the original source file safely stored, along with a copy of the cleaned import file. If something needs correcting, you will know exactly what was used.

Treat consent and privacy as part of the import

A CRM is not a storage cupboard for every email address your business has ever collected. If you operate in the UK, your customer data handling must reflect UK GDPR and the privacy commitments you have made to people.

Check why each record is being held, whether the contact details are accurate and whether marketing preferences have been captured. A sales relationship, an active customer contract and a newsletter subscription are different things. Do not treat them as interchangeable.

Limit access to sensitive information. Keep personal details out of open notes when they do not belong there. If a former employee’s records are imported, make sure account ownership is reassigned rather than leaving valuable relationships stranded.

Privacy is practical, not performative. Clear permissions and clear ownership protect your customers and your team.

Give every record an owner and a next step

An import is complete when the data is usable, not when the progress bar reaches 100 per cent. Each active company and deal should have a clear owner. Without one, follow-ups disappear and valuable accounts become everyone’s problem and no one’s responsibility.

Set a short review window after the import. Over the first week or two, ask the team to flag duplicates, missing fields and confusing labels. Fix patterns at the source rather than correcting the same issue record by record.

Then create a simple standard for new entries. Agree which fields are mandatory, when a deal should be created and what a good account note looks like. A clean CRM stays clean because the team has a shared habit, not because someone performs a heroic tidy-up every quarter.

The goal is not a database with the most fields. It is a trusted place for the relationships that move your business forward. Get the first import right, and every call, handover and opportunity review starts from firmer ground.

Your Details: