Switch to Wholesale VoIP: The 8-Step Migration Guide

Most carriers and providers know they're overpaying for voice termination. Few execute the migration cleanly. The pattern that breaks deployments isn't choosing the wrong wholesaler — it's flipping traffic in one cutover instead of routing in parallel, or signing a deck without auditing billing increments, or losing a corridor's ASR overnight because the new vendor's underlay on that destination wasn't tested.

This guide walks the eight-step migration sequence carriers actually use to land on a new wholesale VoIP termination provider without breaking customer experience. The whole sequence takes four to eight weeks if executed properly.

The first three weeks are evaluation. Weeks four through eight are the staged cutover. Skip the staging and you'll be debugging customer complaints in production with no rollback path.

Step 1 — Audit your current call patterns and costs

Before you talk to a single new vendor, pull six months of CDR data from your existing softswitch or wholesaler portal. You're looking for four things:

• Destination distribution. What percentage of your minutes go where? A surprising number of operations think they're "global" when 80% of their traffic is to five destinations. The pricing leverage is in those five.
• Average call duration by destination. Short-ACD destinations (under 30 seconds) get hit hardest by 30/6 or 60/60 billing increments.
• Time-of-day distribution. If your traffic concentrates in specific windows, you can negotiate time-of-day rate optimization. Carriers won't offer this unless you ask.
• Current ASR, ACD, PDD per destination. Your baseline. The new wholesaler needs to match or beat each metric per destination, not in aggregate.

This audit takes a day if your CDR data is clean. If it's not, fix the data pipeline before you start vendor evaluation.

Step 2 — Evaluate your current billing mechanics

Pull your existing wholesaler's invoice and reconcile it against your CDRs for the last 90 days. You're looking for the deltas that aren't in the rate deck.

• Billing increment effect. If you're on 60/60 billing and your average call duration is 38 seconds, your effective per-minute cost is 58% above the deck rate.
• FAS contamination rate. Calls under 6-second ACD as a percentage of total billable minutes. Above 2% suggests false-answer supervision in the route mix.
• Per-CDR or per-attempt fees. Real wholesalers don't charge any of these. They're all negotiating points.
• DID rental separate from termination. Vendors who charge $0.20/DID/month vs $0.80/DID/month is a meaningful spread at 5,000+ numbers.

What you're producing is a true cost-per-completed-minute that includes everything the rate deck obscures.

Step 3 — Understand the wholesale rate structure

Three pricing models dominate the wholesale market:

• A-Z rate deck — one row per destination prefix, per-minute price. Carrier standard, updated weekly.
• Tiered routing — explicit route classes (Basic CC, Platinum CC, Custom CLI) with published per-tier rates.
• Flat-rate / blended — single per-minute rate covering a defined geography.

Most carriers want a mix. Tiered routing for premium corridors where ASR matters; A-Z deck for everything else; flat-rate only for specific use cases where the bundle math works in your favor.

Step 4 — Request trial SIP trunks from 2-3 wholesalers

Don't shortlist on documentation alone. Provision actual SIP test trunks to your top two or three candidates and put live traffic through them. Two weeks of test data per vendor is the minimum useful sample.

The shortlist for most operations comes from a mix of:

• A Tier-1 wholesaler with destination depth matching your traffic (Bandwidth for US-domestic, BICS or MCM for global, Tata for South Asia)
• A self-service-led wholesaler for operational simplicity (Telnyx, MCM)
• A pure-play CPaaS for comparison if your team values the API surface (Twilio, Plivo)

Real wholesalers return a custom A-Z deck inside 24-72 hours. If a vendor takes a week to deliver a deck, they're not the right partner. For the operator-by-operator breakdown see our wholesale VoIP termination providers comparison.

Step 5 — Calculate true savings

With test trunks running and decks in hand, build a side-by-side cost model:

If the savings number doesn't justify a 6-week migration plus integration risk, the migration isn't worth doing yet. Successful evaluation typically shows 15–40% reduction in total wholesale spend at comparable or better ASR per destination.

Step 6 — Choose your integration method

Three integration patterns cover most wholesale migrations:

• Direct SIP trunk to your existing softswitch. Standard pattern; no custom integration work. The new vendor becomes one of N termination paths in your LCR engine.
• API-driven provisioning. If your operation provisions SIP trunks programmatically, the new wholesaler needs full REST APIs for trunk provisioning, route management, and CDR retrieval.
• SIP-to-SIP gateway. For commercial situations where direct trunk replacement is awkward.

For most operations, direct SIP trunk + LCR is the right answer. Integration work is hours, not weeks.

Step 7 — Implement gradually with parallel routing

This is the step that breaks most migrations when skipped. Do not flip a single switch. Run the new vendor in parallel with the old one for at least four weeks.

• Week 1 — 5% traffic. Route 5% through the new vendor, balanced across top destinations. Monitor ASR, ACD, PDD, NER per destination per day. Compare against baseline.
• Week 2 — 25% traffic. If week 1 metrics held, graduate to 25%. The new vendor's NOC starts seeing your traffic patterns and tuning routes.
• Week 3 — 50% traffic. Same monitoring. By now you have enough volume to negotiate rate sharpening if your original quote was based on lower forecast volume.
• Week 4 — 100% traffic with old vendor on standby. All traffic on new; old SIP trunk remains provisioned and accepting traffic for failover. Run for two weeks before terminating the old contract.

Step 8 — Monitor quality and renegotiate

The migration doesn't end at cutover. Three operational practices keep wholesale costs and quality optimized post-migration:

Monthly billing reconciliation. Pull the wholesaler's CDRs and reconcile against their invoice. Discrepancies above 0.5% are common and almost always resolvable in your favor.

Quarterly rate review. As your volume grows or destination mix shifts, your wholesaler should be willing to sharpen pricing. Quarterly is the right cadence.

Annual competitive benchmark. Even if you're happy, request a deck from one alternate provider annually. The threat of substitution is the only thing that keeps a wholesale relationship competitively priced over multiple years.

Common pitfalls to avoid

The single most common cause of migration failure is cutting over without a parallel run. Two days of customer complaints later you're rolling back to the old vendor with degraded credibility on both sides.

• Optimizing on deck rate alone. A vendor with $0.0028 deck rate and 60/60 billing is more expensive than $0.0035 deck rate and 1/1 billing on sub-30-second ACD traffic.
• Ignoring billing increment in the contract. If it's not in the master agreement, it's whatever the vendor decides next quarter.
• Skipping the FAS audit. Vendors with weak FAS detection inflate your bill silently.
• Not negotiating volume tiers annually. Your traffic profile changes year-over-year.
• Treating compliance as a checkbox. STIR/SHAKEN, RMD, traceback, Section 214 — these are operational obligations, not paperwork.

FAQ