Migrating to VICIdial: What Actually Happens When You Move Off a Hosted Dialler
Most UK contact centres don't go looking for a new dialler because they're bored. They go looking because the per-seat licence has crept up at every renewal, because a feature they need sits behind another paywall, or because they've realised that if the supplier went under tomorrow, their entire operation would go with it. At some point the question becomes: what would it actually take to move?
For a large share of those firms, the destination is VICIdial — the mature, open-source predictive dialler built on Asterisk that runs a great many contact centres precisely because it has no per-seat licensing and no lock-in. This guide is an honest walk-through of what a migration actually involves: what's straightforward, what's fiddly, and where the risk really sits.
Why firms move to VICIdial
Three reasons come up again and again:
- Cost and licensing. Hosted SaaS diallers typically charge per agent, per month, often with usage on top. VICIdial has no per-seat licence — you pay for the infrastructure and the people who run it, not for each headset. Above a certain floor size, the difference is substantial.
- Control and lock-in. With a proprietary platform, your data model, your integrations, and your recordings all live inside someone else's system. Open source means the database is yours, the dialplan is yours, and nobody can switch you off.
- Feature ceilings. Blended inbound/outbound, custom dispositions, granular reporting, deep CRM integration — things that are either extra-cost or simply unavailable on a locked-down hosted product are open on VICIdial.
The trade is that VICIdial gives you control and the responsibility that comes with it — which is exactly the part a migration has to get right.
What a migration actually involves
A dialler migration is not a single switch. It's several workstreams that have to land together:
Lead and list data. Your contact data has to come out of the old system and into VICIdial's structure. This is usually straightforward to export, but the data model rarely maps one-to-one — custom fields, dispositions, and callback states often need remapping rather than a clean copy. Plan for a transformation step, not a drag-and-drop.
Numbers, DIDs and SIP trunking. Your inbound numbers either get ported to a new SIP provider or re-pointed to the new platform. Porting has lead times — often days to weeks — and has to be scheduled, not rushed on cut-over day. Outbound CLIs and any number-rotation setup need rebuilding too.
Call recordings. Decide early whether historic recordings come with you or stay archived on the old system. Migrating large recording sets is doable but takes planning around storage, indexing and retention — and your retention obligations don't pause during a move.
Campaigns, dispositions and dialplan. Campaigns, lists, dispositions, and the call routing itself are rebuilt in VICIdial. This is where platform knowledge matters most: the dialplan controls everything from how calls are presented to agents to how abandoned calls are handled.
Compliance carry-over. DNC and TPS suppression, recording configuration, and — critically — your abandoned-call rate settings under Ofcom's rules all have to be configured correctly on the new platform from day one. A misconfigured predictive ratio on go-live is both a compliance risk and a reputation risk.
Agents. The part most plans underestimate. Agents who know one interface cold will be slower on a new one for a week or two. Training, a quick-reference guide, and a calmer go-live — a single campaign first, not the whole floor at once — protect your connect rates through the transition.
Where the risk actually sits
The technical export is rarely the hard part. The risk lives in three places:
- Timing. Number porting and SIP changes have lead times outside your control. The cut-over has to be sequenced around them, not the other way round.
- Cut-over downtime. Done badly, a migration means a floor sitting idle. Done well, the old system stays live until the new one is confirmed working, and you switch campaign by campaign with no gap.
- Configuration on go-live. Predictive ratios, abandon limits, recording, and suppression all have to be right on day one — not tuned afterwards while live calls go out.
None of this is a reason not to move. It's a reason to move deliberately, with the old system kept warm until the new one has proven itself.
Self-hosted or managed?
VICIdial can be self-hosted — you run the servers, the Asterisk layer, the SIP, the backups, and you take the 2am call when a trunk drops. For some firms with in-house telephony engineers, that's the right answer.
For most, the point of moving off a hosted dialler was never to acquire a second full-time infrastructure job. A managed VICIdial deployment keeps the cost and control benefits of open source while someone else carries the migration, the hosting, the compliance configuration, and the out-of-hours support. The migration itself is also simply easier when it's run by people who've done it before and know where the platform hides its sharp edges.
A realistic timeline
A clean migration for a small-to-mid floor is usually a matter of a few weeks end to end — most of which is number porting and parallel-running, not build time. The build can move fast; the dependencies outside your control set the pace. Anyone promising an overnight switch for a live floor is underestimating the porting and the go-live configuration.
Telebyte runs managed VICIdial / Asterisk for UK contact centres, and migrations from Amcat/Noble, Touchstar, Ultra and Onstate — keeping the old system live until the new one is confirmed working, with compliance configured from day one. If you're weighing a move, tell us what you're running today and we'll give you a straight assessment.