A practical guide to CRM migration for SMEs
Picking a new CRM feels like the big decision. It isn't. The big decision comes a few weeks later, when someone asks: "Do we just export everything from the old system and import it?"
Most companies say yes. It's quick, nobody has to make a call on a single record, and the project stays on schedule. Six months later the new CRM holds the same duplicates, the same companies that closed years ago and the same contacts nobody recognises. Only now it's in a nicer interface, and the team has lost faith in it again.
We've seen this often enough that it's become the one piece of advice we give on every migration: clean your data before you move it.
Once data sits in a new system, it gathers momentum. Workflows start using it, reports get built on it and emails go out to it. Every week the mess stays, it gets more expensive to remove, because deleting a record now means checking what depends on it.
Before the move, it's the cheapest it will ever be. Nothing depends on the old records yet. A salesperson can look at a list of their contacts and say "keep, merge, delete" in an afternoon.
There's a second reason. The people who know whether a contact is still relevant aren't in IT. They're your salespeople, account managers and project leads. Before the migration, you can ask them. After it, the question gets lost in everything else.
1. Decide what the new system has to answer. Start from the reports, not the data. Which questions should the CRM answer a year from now? Revenue by segment, open deals by region, customers who haven't ordered in twelve months? That list tells you which fields matter and which ones you can leave behind.
2. Give every record an owner. Split the export by owner, so each salesperson or account manager gets the companies, contacts and deals they're responsible for. Records nobody claims are usually the first ones to go.
3. Let owners clean their own data. Keep, merge or delete. Owners know which customer was bought, which contact left and which company appears twice under two spellings. It's faster than it sounds, and it gives people a reason to trust what ends up in the new system.
4. Agree the rules before the import. Which fields are mandatory? How do you write company names? Which categories exist for industry, language or lead source? Decide once, write it down, and build it into the new CRM as required fields and dropdowns instead of free text.
5. Import once, then protect it. Import the clean set into a fresh system, not on top of the old mess. Then keep it clean with required fields, duplicate checks and a short data quality review every few months.
Servix Plantedekor AS was moving off a legacy CRM to match the rest of its Nordic group. Before anything was exported, each salesperson went through the companies and contacts they owned, removed what was outdated and signed off what was valid. Only then did we format the data and import it into a fresh HubSpot. The result is a CRM the team trusts, with reporting on the same standard as the rest of the group.
Convergence Partners, a Swiss venture capital firm, had its network spread across several people's Outlook address books. We extracted the contacts, deduplicated and structured them first, and only then imported them into HubSpot, segmented into the investor and stakeholder groups the team actually works with.
Sofi Filtration had leads landing in scattered inboxes and spreadsheets. We gathered the data from every stakeholder and salesperson, cleaned it and moved it into one place, with custom properties and a pipeline that matched how Sofi actually sold.
You can read more about this in the customer stories.
Less than most people expect. The technical import is usually the quick part. The clean-up depends on how many records you have and how many people own them, but because each owner only reviews their own data, the work spreads well.
The step that takes longest is usually the first one: agreeing what the system has to answer. That's a good problem to have, because it's the conversation that decides whether the new CRM gets used at all.
If you can tick all seven, you're ready. If you can't, that's exactly where to start.
If you're planning a CRM move, or you suspect the data in your current one isn't telling you the truth, our Digitalization Audit is built for this moment: two to four weeks, a clear picture of where your data actually lives, and a written roadmap before you commit to any platform. If HubSpot turns out to be the right fit, here's how we set it up. And once the data is clean, you can finally ask it questions in plain language.
Still running the business on spreadsheets? Start with why SMEs should ditch spreadsheets.
Not sure where your data stands? Start with the Digitalization Audit →