A CRM migration quote is really three jobs priced as one number: mapping your fields and objects into the new system, deduplicating and cleaning the records you bring over, and rebuilding the automations, sequences, and templates that ran your pipeline in the old tool. Most quotes bundle all three into a flat fee, which is exactly why two vendors can quote the same contact count 3x apart. This breaks the three apart so you can price each one and tell when exporting the data yourself on a weekend beats hiring anyone at all.
What CRM migration services actually do
Strip away the sales language and a migration is a sequence of mechanical steps. A vendor pulls an export from the old CRM, by CSV or API, covering every object class you use: contacts, companies, deals, tickets, notes, attachments. They map each field to its equivalent in the new system, or build a new custom field where none exists. They run a dedupe pass, then rebuild your automation rules, sequences, and reports before cutting your team over to the new login.
That's it. There's no step in there that requires deep technical skill, which is also why the price varies so much. It's mostly hours, and the hours depend entirely on how messy your specific setup is, not on how complex switching CRM providers is in general.
The quote is three jobs wearing one price tag
Ask what's driving a vendor's number and most can't tell you, because their own estimate was built on a rough guess, not a line-by-line breakdown. That's the real problem with CRM migration pricing: a single lump sum hides whichever driver the vendor underpriced, and you only find out which one when the invoice grows mid-project.
The three drivers are data mapping, dedupe and hygiene, and workflow rebuild. Each scales on a different variable, field count, record count, and automation count, so a vendor who prices by contact count alone is guessing on two of the three. A CRM migration quote built from three separate estimates is one you can actually sanity-check against your own setup.
Driver one: data mapping, or matching fields that were never built to match
No two CRMs structure data the same way, even when they look similar on the surface. Salesforce calls it an Opportunity, HubSpot calls it a Deal, and the fields attached to each rarely line up one to one. Before you accept a quote, check these specifically:
- Custom fields and whether they exist at all in the target system, or need to be built from scratch
- Picklists and dropdown values, since "Qualified Lead" in one tool might be three different values in another
- Object relationships, like which records roll up under a company versus sit as standalone contacts
- Multi-select fields, which some import tools silently flatten into a single value
- Currency and date formats, which break silently on import rather than throwing an error
If you're moving into HubSpot specifically, check the tier you're landing on before you map anything. HubSpot's free and Starter tiers cap custom properties and marketing contacts, so a mapping plan built for an Enterprise feature set can force an unplanned upgrade, or a round of trimming fields, mid-migration. That's a concrete thing to ask about before the mapping work starts, not after.
Driver two: dedupe and data hygiene, the part most quotes undercount
This is where estimates go wrong most often, because duplicate counts don't scale in a straight line with contact count. A company with three lead sources, a web form, a trade show list, and a cold outreach tool, can easily generate three separate records per real customer. Fuzzy matching on name and company, rather than exact email match, is the only way to catch most of them, and fuzzy matching takes a human reviewer's judgment on the ambiguous cases, which is where the hours actually go.
Merge logic matters as much as matching rules. Which record wins when two contacts merge, the most recently modified, the one with the most complete fields, the one tied to an open deal? A vendor who can't answer that question hasn't actually built a dedupe process. They've budgeted for one.
CRM hygiene isn't a one-time event either. A migration is a single clean sweep, but without a matching rule enforced on record creation afterward, like blocking a new contact with a duplicate email, your clean CRM fills back up with duplicates within a few months. If CRM data cleansing is part of your quote, ask whether it includes that ongoing rule or just the one-time cleanup. Keeping one clean system of record after the move is its own problem, covered in our single source of truth use case.
Driver three: rebuilding the workflows, not just the records
Records transfer. Workflows don't, not cleanly. Automation rules that trigger off a specific field value, email sequences tied to list membership, quote templates pulling from merge fields, all of it has to be rebuilt by hand in the new system's logic, because the underlying rule engines aren't compatible. This is usually the largest line item in a migration quote, and it's the hardest one to estimate up front, because the vendor doesn't know how many automations you have until they're inside your account.
A HubSpot CRM migration service differs from a straight CRM-to-CRM swap here, because HubSpot workflows often touch marketing sequences and lifecycle stages that a sales-only CRM doesn't have, which adds rebuild hours a like-for-like contact count wouldn't predict.
Recruitment agencies hit a version of this that's easy to miss: an ATS and a CRM aren't the same system, even when a vendor like Bullhorn or RecruitCRM blurs the line by offering both. If your agency runs a CRM for client relationships and a separate ATS for candidate pipelines, migrating one without touching the other changes the scope significantly, and a quote that treats "CRM migration" as covering both without saying so is underscoping the job. Rebuilding this kind of automation logic is close to what we do as ongoing AI automations work, just applied once instead of continuously.
The arithmetic a vendor's quote is built on
Here's a worked example, built from assumptions stated up front, not a quote you should expect to match. Say you're moving 5,000 contacts with 40 custom fields and 15 active automations.
Data mapping at roughly one hour per five custom fields runs about 8 hours. Dedupe on 5,000 records with a 15% duplicate rate, assuming half of those need manual review at ten minutes each, runs around 20 hours. Workflow rebuild at 2 to 4 hours per automation for 15 automations runs 30 to 60 hours. That's 58 to 88 hours total. At a blended agency rate of, say, €90 to €140 an hour, a purely illustrative range, that's roughly €5,200 to €12,300.
The number matters less than the method behind it. Swap in your own contact count, field count, and automation count, and you'll land closer to a real estimate than any flat industry figure could get you. Our admin cost calculator runs the same kind of hours-times-rate math for recurring manual work, if you want to see where migration cost sits next to the ongoing admin cost it's meant to replace.
When a weekend export job beats hiring anyone
Here's the honest walk-away case. If your contact count is small, say under a few hundred, your fields are mostly standard with few or no custom picklists, and you have no automations to rebuild because you never built any, you don't need a vendor. Export the CSV from your old CRM, clean the obvious duplicates by sorting on email, and re-import using the new system's own import wizard. HubSpot's and Salesforce's import tools both handle this case natively, and a weekend is a realistic timeline for it.
The tradeoff is real: native import tools only catch exact-match duplicates, not fuzzy ones, so you'll do a manual pass on anything that looks similar but isn't identical. That's a few hours of tedious work, not a few thousand euros. If that describes your setup, paying for CRM migration services is paying for a job you could finish yourself before Monday.
What to ask before you sign a migration quote
Force any vendor to price the three drivers separately before you sign anything. A single number tells you nothing about which part they've underbid.
- Ask for data mapping, dedupe, and workflow rebuild as three line items, not one total
- Ask what their dedupe matching rule actually is, exact email, fuzzy name and company, or phone
- Ask who tests each rebuilt automation before cutover, and what that test looks like
- Ask what happens to historical activity, calls, notes, email threads, not just current field values
- Ask for hours and hourly rate per driver, not just the total, so you can check their math against your own
- Ask what's explicitly excluded, reporting dashboards, email templates, and third-party integration reconnects are common gaps
If you're weighing a migration against just moving to a tool built for your industry, check published pricing where it exists rather than guessing. RecruitCRM, for example, lists its per-seat pricing on its own site, which gives you a real anchor instead of a vendor's verbal estimate.
Common questions
How much does a CRM migration cost?
It scales with three things, record count, how many custom fields and picklists need remapping, and how many automations have to be rebuilt, so a flat industry number is less useful than pricing those three separately.
Can I migrate my CRM myself without hiring a service?
Yes, if your contact count is small, your fields are mostly standard, and you have few or no automations to rebuild, a CSV export and re-import over a weekend usually covers it.
What's the biggest hidden cost in a CRM migration quote?
Workflow and automation rebuild. A vendor prices it from a guess until they're inside your account, so it's the line that's most often wrong, and almost always wrong in the direction of more hours, not fewer.
How long does a CRM migration take?
It tracks the same three drivers as the cost, more custom fields and more automations mean more time, independent of how many contacts you're moving.
A CRM migration is a one-time version of a problem most sales and marketing teams carry every week: manual data entry, records that don't match across tools, and workflows nobody fully trusted to begin with. If that sounds like your actual day-to-day rather than a one-off project, the Revenue Roadmap scores exactly that kind of work, including CRM cleanup, on value against effort and puts a fixed price on fixing it.
