Deskspark

Migration guide

Moving a live desk without dropping anything

Switching CRM is not risky because the data is hard to move. It is risky because you have to keep billing, placing and answering candidates while you do it. This is the order we would do it in, with an honest list of what does not come across.

01

Export, before you cancel anything

Half a day

Pull a CSV of each of these out of your current system, in this order: companies, contacts, jobs, deals or placements. Most systems bury this under Admin, Reports or Data Export; if yours only offers it by support ticket, raise that ticket now — it is usually the longest-lead part of the whole move. Export while your licence is still live and keep the files. An export you cannot re-run is a bad position to be in on day three.

If you are coming from Bullhorn, export the Candidate and ClientCorporation objects as they come. Deskspark recognises a Bullhorn export on sight and fills in the whole column mapping for you.

02

Import companies first, then people

An hour

Order matters, and it is the one thing most migrations get wrong. Import companies before contacts and jobs, and Deskspark links each new record to the right company automatically by matching the name. Do it the other way round and you get a database of people and roles floating free of the businesses they belong to, which then has to be relinked by hand. Every import shows you the mapping and a preview before a single row is written.

Notes come across with contacts if your export has them — map the column and they land on the record rather than in a spreadsheet nobody opens.

03

Bring the CVs across

Minutes

Bulk-upload your CV folder from the Contacts page. Deskspark matches each file to the right candidate by the email address in the filename, uploads the ones it can place, and tells you plainly which it could not rather than dropping them silently. Most systems export CVs named by email already; if yours names them by candidate ID, a bulk rename is worth the ten minutes.

04

Connect the live rails

Half an hour

Connect your mailbox so sent history and replies sync into the record, set up SMS and WhatsApp if you use them, and invite your desk. Everyone works from the same contacts, jobs and pipeline from the first day, so there is no window where half the team is on the old system and half on the new one because the invites hadn't gone out.

Outlook and Microsoft 365 connect for any account. Gmail is in limited release while Google finishes reviewing our app — tell us and we'll get you added.

05

Run both for two weeks, then cancel

Two weeks

Keep the old system read-only and work entirely in Deskspark. Two weeks covers a full cycle of the things that only surface occasionally: an invoice going out, a placement hitting its guarantee window, a monthly report. Check those specifically rather than checking that contacts exist. Cancel once a full cycle has passed with nothing missing, not on the day the import finishes.

Your trial is 14 days, which is deliberately the same length. Run the parallel fortnight inside it and decide with real evidence.

What comes across

  • Companies, with their own records and offices
  • Candidates and client contacts, with tags and status
  • Notes attached to contacts
  • Jobs and roles, with salary, location and status
  • Deals and placements, with stage and value
  • Business-development prospects
  • CVs, matched to candidates in bulk
  • Structured addresses, so radius search works from day one

What doesn't, and why

  • Historic email threads

    Your mailbox is the system of record for those. Connect it and everything from that point syncs into the contact; older threads stay searchable where they already live.

  • Invoices and timesheets already raised

    Deskspark drafts invoices from placements going forward. Closing out billing in the old system is cleaner than migrating half-paid history into a new one.

  • Interview records and stage history

    There is no import for these yet. Open placements are worth re-entering by hand; closed ones are usually not.

  • Anything from JobAdder or Vincere as a one-click preset

    Bullhorn publishes its export field list, so we auto-map it. JobAdder does not publish one. Vincere lets each agency choose, reorder and rename its own export columns, so there is no fixed set anyone could preset — a canned mapping would be wrong for most of you. Both still import fine: you map the columns once, we remember that mapping, and the next file maps itself.

Most of this is an afternoon

The import itself takes a couple of hours. The fortnight of parallel running is the part that makes it safe, and it fits inside the free trial.

Start freeAsk us about your export

Weighing up more than one system? Score your shortlist against what actually matters on a desk.