Migrating to VICIdial: the complete guide
A calm, end-to-end plan for leaving a hosted dialer for VICIdial - what moves, what to export, how to rebuild, and how to cut over without downtime.
Leaving a hosted dialer for VICIdial is mostly a data and planning problem, not a software problem. The dialer itself is well understood. What trips teams up is everything around it: the leads sitting in someone else's database, the do-not-call list you are legally required to keep, months of call recordings, the dispositions your reports depend on, and the phone numbers your customers already recognize. Move those carefully and the switch is boring. Move them carelessly and you spend your first week firefighting.
This guide is the end-to-end version. It covers why teams migrate, what actually moves, how to export cleanly from the old platform, how to rebuild in VICIdial, and how to cut over with a rollback plan in your back pocket. Whether you are coming off Five9, Convoso, Twilio Flex, RingCentral, Genesys, or an aging GoAutoDial box, the shape of the work is the same. The platform-specific spokes linked throughout fill in the details that differ vendor to vendor.
Why teams migrate to VICIdial
Three reasons come up again and again: cost, control, and data ownership. Per-seat hosted dialers bill a fixed amount per agent per month, often with per-minute charges layered on top. As a center grows, that line item grows linearly with no ceiling. VICIdial runs on a server you rent, so your cost is tied to hardware, not headcount - a center that doubles its agents does not double its bill. The arithmetic is laid out in detail in our look at the cost savings of moving off a per-seat dialer.
Control is the second reason. With a closed platform you get the features the vendor decided to ship, on their roadmap, behind their pricing tiers. VICIdial exposes the dialer's internals - your campaign settings, your dial method, your dialplan, your reporting tables. You can tune a Campaign (a dialing project that ties a lead list to a dial method and a script) exactly how you want it instead of choosing from a dropdown of plans. Data ownership is the third: the recordings, the leads, and the call logs live on your box, not in a vendor's account you lose access to the day you stop paying.
What actually moves
Before you touch any tooling, list every asset that has to survive the switch. Most of it falls into eight buckets:
- Leads - the contact records you dial. Each Lead carries a phone number, name, and custom fields you have built reports around. See moving leads into VICIdial.
- The do-not-call list - a DNC (do not call) record set you are legally obligated to honor. Losing it is a compliance incident, not a convenience problem. See importing your DNC list.
- Call recordings - audio files plus the metadata that ties each one to a call. See migrating call recordings.
- Dispositions - the outcome codes agents pick after each call. See mapping dispositions to statuses.
- Call scripts - the on-screen text agents read. See rebuilding scripts in VICIdial.
- Phone numbers - the inbound DIDs and outbound caller IDs your customers know. See porting phone numbers.
- Integrations - CRM hooks, webhooks, and the API glue that pushes results downstream.
- Agents - the people, their logins, and the retraining they need on a new screen. See retraining agents.
One thing that does not move cleanly is historical reporting. Every vendor calculates metrics slightly differently, and you cannot perfectly recreate last quarter's dashboards in VICIdial's reporting tables. Plan for a reporting gap and decide early how you will bridge it - we cover the options in handling the reporting gap.
Exporting from the old platform
The export is the part most teams underestimate, because hosted platforms make it deliberately quiet. Start a written checklist on day one and treat it as the gate before any cutover date is set. A complete walkthrough lives in the data export checklist, but the essentials are these:
- Pull leads to CSV with every custom field included, not just name and number. A dropped column is a broken report later.
- Export the DNC list separately and verify the row count matches the platform's own number. This list is your legal shield - check it twice.
- Bulk-download recordings before you cancel anything. Many platforms purge audio on account close, and your Recording retention obligations do not pause because you switched vendors.
- Document your disposition list and your script text. These rarely export as files - screenshot or copy them by hand.
- Record which DIDs you own and which the vendor owns. You can only port numbers you control.
Rebuilding in VICIdial
With exports in hand, you rebuild from the carrier up. The carrier (your Carrier, the telco that actually delivers calls to the phone network) connects to VICIdial over a SIP trunk - the authenticated channel between your dialer and that carrier. VICIdial is BYOC: you bring your own carrier and register the trunk to your server's IP. This is the first thing to stand up, because without a working trunk you cannot test anything else.
Next, recreate your structure. Build a campaign for each dialing project and a list for each Lead list (a named batch of leads loaded into a campaign). Choose a Dial method - manual, ratio, or Predictive dialing, the mode that dials several numbers ahead of free agents to drive idle time toward zero. Load leads with the Lead loader, VICIdial's CSV import tool, mapping your exported columns onto VICIdial's fields and your custom fields into the lead's custom layout. Set up an Ingroup for inbound queues if you take calls, and attach an Agent script - the on-screen prompt agents read - to each campaign.
The detail that bites people is dispositions. In VICIdial every outcome is a Status (lead status) code attached to a campaign, and the Disposition an agent selects writes that status onto the lead. If your old platform's outcome names do not map one-to-one onto VICIdial statuses, your conversion math breaks silently. Map them deliberately - the mechanics are in the dispositions-to-statuses guide. While you are at it, load your DNC entries into the DNC list and confirm the campaign is set to honor it before a single test call goes out.
Want the server build to disappear entirely? A VICIfast managed plan stands up a secured, branded VICIdial in under 40 seconds, so the only work left is loading your data and pointing your carrier at the box.
The cutover: parallel run versus hard cutover
There are two ways to flip the switch. A hard cutover stops the old platform and starts the new one on the same day - fast, cheap, and unforgiving if anything is wrong. A parallel run keeps both systems live for a window, moving agents over in batches while the old dialer stays available as a safety net. For most centers the parallel run wins, because it turns a high-stakes single event into a series of small, reversible steps. We compare the two approaches in detail in running old and new dialers in parallel.
Whichever you choose, write the steps down in order and rehearse them. A zero-downtime plan sequences the work so calls are always being handled somewhere - you move agents, not the whole floor, and you keep the lead Hopper (the buffer of numbers VICIdial pre-loads for dialing) full on the new box before you send anyone to it. The full runbook is in the cutover plan. The migration phases look like this:
stateDiagram-v2
[*] --> Plan
Plan --> Export
Export --> Rebuild
Rebuild --> ParallelRun
ParallelRun --> Cutover
Cutover --> Decommission
Cutover --> Rollback
Rollback --> ParallelRun
Decommission --> [*]Notice the branch from Cutover back to a rollback state. Every plan needs an undo. Decide in advance what failure looks like - call quality below a threshold, a broken disposition, agents unable to log in - and what single action reverts to the old system. Keep the old platform reachable until you have decommissioned it on purpose, not by accident. The discipline of writing that escape hatch down is covered in building a rollback plan.
Porting numbers and retraining agents
Number porting is the slowest single item in any migration and the one you cannot rush, so start it first even though it finishes last. Moving a DID (direct inward dialing) (a direct inward dial number that routes inbound calls to a queue) from one carrier to another is a regulated process that takes days, sometimes weeks, and runs on the losing carrier's schedule, not yours. The trick is to keep the number answerable on the old platform right up until the port completes, then route it to your VICIdial ingroup. The step-by-step is in porting phone numbers to VICIdial.
Agents are the other long pole. VICIdial's agent screen looks nothing like a polished hosted UI, and the first day on a new tool is always slower. Train before cutover, not during it. Walk agents through logging in, taking and placing calls, selecting dispositions, and using a Scheduled callback (a callback the agent books for a specific future time). A few hours of practice on a test campaign saves a chaotic launch day. The retraining playbook is in retraining agents on VICIdial.
Where managed hosting fits
Everything above is the migration. Notice that none of it is server administration - it is data, carrier, dispositions, scripts, and people. The Linux build, the Asterisk compile, the database tuning, the TLS certificate, and the firewall hardening are pure overhead that adds nothing to your dialing. That is the part managed hosting removes. VICIfast provisions a dedicated, single-tenant VICIdial on a branded subdomain over HTTPS in under 40 seconds, so the clock that used to be a half-day of install work is gone. You spend your migration time on the things that actually matter: getting leads loaded, the trunk registered, and dispositions mapped right.
The platform-specific spokes cover the quirks of leaving each vendor. If you are moving off a per-seat hosted suite, see Five9 to VICIdial, Convoso to VICIdial, Genesys to VICIdial, NICE CXone to VICIdial, or Talkdesk to VICIdial. Coming from a CPaaS or UCaaS tool? See Twilio Flex to VICIdial, RingCentral to VICIdial, Aircall to VICIdial, Dialpad to VICIdial, or Vonage to VICIdial. Already on an open-source dialer? We cover GoAutoDial, OSDial, and self-hosted ViciBox to managed too.
Migration is a planning exercise, not a leap of faith. Export carefully, rebuild from the carrier up, run in parallel, keep a rollback ready, and port your numbers early. When you are ready for the box itself, browse VICIfast managed plans - a secured VICIdial server in under 40 seconds means your migration is about your data and your customers, never about building a server.
About VICIfast LLC
VICIfast LLC operates a managed VICIdial hosting + BYOI service for outbound and inbound call centers. We run the dialers, the carriers, the recordings pipeline, and the compliance plumbing so operators don’t have to.
Citing this article
VICIfast Engineering. “Migrating to VICIdial: the complete guide”. VICIfast LLC, June 29, 2026. Retrieved from https://vicifast.com/blog/migrating-to-vicidial-guide
Have questions?
Related posts
You might be interested in
VICIfast newsletter
Liked this? Get the next one in your inbox.
We ship the kind of stuff you just read — concrete, numbers-first, no drip. One email when a new post goes live. Unsubscribe in one click.
Comments
No comments yet — be the first.