In short

You can implement a CRM in six steps: define goals and the sales process, set requirements including data protection, test tools with real cases, clean and import data, train the team and adjust after launch. Most of the work lies in process and data. Cloud CRM systems require a data processing agreement under Art. 28 GDPR.

  • Process before software: describe your sales workflow before you choose a tool.

  • Data protection from day one: under Art. 28 GDPR, a cloud CRM requires a data processing agreement.

  • Data quality decides: clean contacts before the import and only migrate what you need.

  • Test with real cases: run real enquiries through the trial, not demo data.

  • Clear maintenance rules keep the system from going stale within a few months.

A CRM brings contacts, conversations and sales opportunities together in one place. Whether your team will actually use it is decided before you buy: by the process, the data and clear rules. The six steps below are aimed at small and mid-sized sales teams without their own IT department.

Step 1: Define goals and process

Start by deciding what the CRM should improve. Two or three measurable goals are enough. Examples: every enquiry gets a reply within one working day. Every proposal is followed up after one week. Management sees each month which deals are likely to close.

Next, describe your sales stages from first contact to closed deal. These stages will become the columns of your pipeline. A typical B2B pipeline looks like this:

  1. Enquiry received

  2. First call held

  3. Needs clarified

  4. Proposal sent

  5. Negotiation

  6. Won or lost

For each stage, write down when a contact moves on to the next one. Without these rules, everyone sorts differently and the pipeline loses its value as a forecast.

Step 2: Write down requirements

Draft a short requirements list and separate must-haves from nice-to-haves. The table helps you sort them.

Area

Typical questions

Contacts and companies

How many records? Which fields do you really need?

Pipeline

Do you need several pipelines or custom stages?

Email and calendar

Should the CRM connect to Microsoft 365 or Google Workspace?

Integrations

Accounting, newsletter tool, website forms?

Data protection

DPA, server location, deletion periods, permissions?

Costs

Per-user subscription, minimum term, paid add-ons?

A CRM vendor processes personal data on your behalf. Under Art. 28 GDPR you need a data processing agreement for this. Our guide to the data processing agreement explains what it must contain. Also add the CRM to your record of processing activities under Art. 30 GDPR.

If the vendor is based in the US or stores data there, international data transfers become relevant. Our article on the EU-US Data Privacy Framework covers the basics.

Step 3: Compare and test tools

Shortlist two or three vendors. Our CRM comparison rates widely used systems by features, pricing model and data protection. Then use trials or free plans, with real cases from your daily work.

  • Run a real enquiry from first contact to proposal.

  • Connect a mailbox and link an email to a contact.

  • Create a report on the open pipeline.

  • Ask two colleagues to complete a task without any introduction.

After each test, note what worked and where the team got stuck. Our guide to choosing SaaS tools shows how to turn these impressions into a traceable decision.

Step 4: Clean and import your data

Before a CRM rollout, contacts are usually scattered across spreadsheets, Outlook and business cards. Clean this data before the import. Remove duplicates, delete outdated entries and standardise the spelling of companies, countries and phone numbers.

The GDPR requires data minimisation. Under Art. 5(1)(c), you may only process data that is necessary for the purpose. Do not migrate contacts "just in case" when there has been no business relationship for years.

Retention rules apply at the same time. In Germany, for example, Section 257 of the Commercial Code (HGB) requires merchants to keep received commercial letters and copies of sent ones for six years. Other countries have their own rules. Decide whether old correspondence belongs in the CRM or in an archive. For the import itself, this order works well:

  1. Define the CRM fields and map them to the columns of your old data.

  2. Run a test import with a small sample.

  3. Check the result and fix mapping errors.

  4. Start the full import.

  5. Keep the old system read-only until everything has been checked.

Step 5: Train the team and set rules

A CRM is only as good as its upkeep. Write the rules down before the team starts:

  • Required fields: what information does every new enquiry need?

  • Ownership: who maintains companies, contacts and the pipeline?

  • Timeliness: every customer conversation gets a note on the same or the next day.

  • Accountability: one person looks after fields, permissions and questions (the CRM owner).

Train in short sessions using your own data. Collect questions in one place during the first weeks. This quickly shows which rule is unclear or which field nobody understands.

Step 6: Launch, measure, adjust

Start with one team or one pipeline, not the whole company. After about four and again after twelve weeks, check the goals from step 1. Look at the data itself too: which fields stay empty? Where do opportunities get stuck?

Your next step: write your sales stages on a sheet of paper. Then look at individual vendors, for example in our Pipedrive review.

Frequently asked questions

How long does a CRM implementation take?

It depends on team size, data volume and integrations. In our assessment, small teams with clean data can complete the rollout in a few weeks. Several pipelines and many integrations make the project considerably longer.

Do I need a data processing agreement for a CRM?

Yes, if the vendor processes personal data on your behalf, which is the norm for cloud CRM systems. Art. 28 GDPR requires a contract for this. Our guide to the data processing agreement explains the details.

Can I migrate all old contacts into the CRM?

Not without checking. The GDPR requires data minimisation: you only process data you need for the purpose. At the same time, national retention rules for business correspondence may apply, such as Section 257 HGB in Germany. If in doubt, ask your data protection officer.

How much does a CRM cost?

Most cloud CRM systems charge a subscription per user and month. Some vendors offer free plans with limited features. Our CRM comparison gives an overview.

Sources

  1. Regulation (EU) 2016/679 (General Data Protection Regulation) , EUR-Lex, Publications Office of the EU , 2016-04-27
  2. § 257 HGB Aufbewahrung von Unterlagen, Aufbewahrungsfristen , Bundesministerium der Justiz, gesetze-im-internet.de
By

SaaS Column Editorial Team

Editorial team

The SaaS Column editorial team writes about business software for small and mid-sized companies, freelancers and teams, with a focus on Germany, Austria and Switzerland and the wider EU. We cover how to choose, implement and budget for SaaS tools, as well as data protection and compliance: from CRM and accounting to project management and team communication.

We work with primary sources. These include vendor documentation, pricing and trust pages, data processing agreements, legislation and publications from authorities such as the European Commission and national data protection regulators. Every verifiable statement is backed by a source listed at the end of the article.

Comparisons and reviews are editorial assessments based on a fixed criteria catalogue. We state what a score is based on and never pretend to have run hands-on tests. Our methodology explains the details.

We are independent: there are no paid placements and no affiliate links. Vendors have no influence on our ratings. If you spot an error, we correct it and note substantive changes visibly on the article.