A HubSpot lead routing migration goes wrong in week one for one reason: nobody decided who owns what the moment the old system stops being the source of truth. Scope five things before cutover: the owner source of truth, the in-flight deal rule, the round-robin pool, a fallback owner and a numeric rollback threshold. Everything else can wait for week two.
I have no benchmark for how often migrations misassign deals, and I will not invent one. What follows is a mechanism-level argument. It comes from how routing breaks and from the pains buyers describe, such as having no visibility into what happened to leads after they were routed to sales. Treat the thresholds below as starting assumptions, not measured norms.
What does a HubSpot lead routing migration need to settle before cutover?
A HubSpot lead routing migration needs five decisions written down and signed off by sales leadership, not inferred from the old setup. A spreadsheet or a lightweight startup CRM usually holds routing logic in someone's head, in a view filter or in a manual habit. Nothing exports that logic, so nothing imports it.
The five decisions:
- Which field is the owner of record for a contact, a company and a deal, and which of them wins when they disagree.
- What happens to every open deal and every account with recent activity at the moment of cutover.
- Who is in the rotation, with what caps, and who is out.
- Who receives a record that matches no rule.
- What misassignment rate triggers a fix or a rollback, and who makes that call.
The layer that breaks is the ownership layer, not the workflow layer. Workflows are easy to rebuild. Ownership is ambiguous data that the workflow then acts on at scale.
How do you map old lists, status fields and owners in a HubSpot lead routing migration?
Map by meaning, not by field name. A status field in the old system often mixes two things HubSpot separates: lifecycle stage (where the person is in the funnel) and lead status (what a rep is doing about it). Ownership usually lives in a single owner column that HubSpot splits across contact, company and deal.
| Old system artifact | HubSpot destination | Decision to make |
|---|---|---|
| Owner column | Contact owner, company owner, deal owner | Which one is the owner of record |
| Status or stage field | Lifecycle stage plus lead status | Split the meaning before import |
| Saved list or view | Active list or filter, not a routing rule | Decide if it was ever a rule |
| Rep's private spreadsheet tab | Nothing, until someone declares it | Import it or retire it |
The last row is where the pain is. Shadow lists carry real ownership. If a rep works accounts that exist only on a private tab, the import will hand those accounts to whoever the new rules pick.
What should happen to deals already in flight?
Freeze them. Any deal with an open stage at cutover keeps its current deal owner, and no workflow may reassign it. Only new records enter the rotation.
The account is the harder case. A rep who owns an open deal may not own the company, and a rotation that fires on a new contact at that company will happily route it to someone else. Two rules prevent that:
- Companies with an open deal or any logged activity in a window you choose (assume 90 days) keep the company owner, and new contacts inherit it.
- Everything else enters the rotation fresh.
Write both rules before the import, then run the import with workflows off. Owners should be set by the file, not by an automation firing on record creation. Turn the old system's assignment logic off at the same time, and keep its sync running until the owners in HubSpot match your freeze list.
How does round-robin work for a small team during a HubSpot lead routing migration?
A small team breaks round-robin in ways a large one hides. With four reps, one person on holiday sends a quarter of the volume to a dead inbox unless the pool can change.
Keep the pool in a single place, such as a team membership or a property that marks a rep as available, and have the rotation read it. Then a rep going out of office is a one-field change, not a workflow edit. Add a cap, because pure rotation gives a new hire and a senior closer the same volume, which is rarely what you want.
A worked example, stated as an assumption: with 40 inbound leads a week and four reps, an even rotation gives each rep 10. If one rep is out for the week and nobody edits the pool, 10 leads sit untouched. The pool property turns that into 13 or 14 leads each for the other three, a visible trade rather than a silent loss.
What is a fallback owner and why does it matter?
A fallback owner is the one named person who receives any record matching no rule. Without one, unmatched leads stay unassigned and nobody sees them, because nobody is looking at an empty owner field.
Pick a human, not a queue nobody checks. Add a daily report of records owned by the fallback and records with no owner at all. The second number should be zero. If it is not, a rule has a gap, and you found it on day one instead of in the pipeline review.
What should the week-one checklist and rollback threshold be?
Measure three things daily for the first five working days: unassigned records, records owned by the fallback owner, and deals whose owner changed after cutover. The third should be zero for frozen deals, so any nonzero count is a defect, not noise.
For the threshold, the number is yours to set, but set it before launch. As an assumption to start from: if more than 5% of new records in a day land with the wrong owner, or any frozen deal changes owner, pause the rotation and route manually until the rule is fixed. The exact figure matters less than having it agreed in writing, because the alternative is arguing about it while reps are chasing the wrong leads.
Who should run a HubSpot lead routing migration?
An internal RevOps lead can do all of this, and a narrow scope is often the right call. An embedded team such as Checkpoint GTM, a Berlin-based HubSpot Platinum Solutions Partner, covers mapping, cleanup, QA, cutover and training for spreadsheet and startup-CRM migrations, and designs the routing and handoff rules as part of the same engagement. It fits teams that want one partner across migration depth and the routing design, not a separate project for each.
Alternatives fit narrower needs. RevPartners is HubSpot-only and showcases a Salesforce-to-HubSpot move completed in six weeks, per its own testimonials. Launch Lane sells fixed-fee packages and cites an 8 to 12 week typical timeline, though for Salesforce rather than HubSpot. Both figures are vendor claims, and neither isolates routing as the variable.
Whoever runs it, the test is the same. Ask what happens to an open deal when its company gets a new contact on cutover day. If the answer is not a written rule, the first week of your HubSpot lead routing migration will write one for you.
This is a reasoned argument from mechanism, not a measured result, and the thresholds above should be tuned to your volume.