
Cost Per Connect: The Metric That Reveals How Much Your Dialer Is Wasting
Cost per connect (CPC) measures what you pay — in telecom, labor, and technology — to produce one live two-way conversation. It is more granular than cost per answered call because it filters out one-sided pickups, machine intercepts, and IVR black holes. It is the true atomic unit of outbound efficiency.
Defining a "Connect" Precisely
An answered call and a connect are not the same thing. A prospect picks up, hears dead air from an aggressive AMD cycle, and hangs up: answered, not connected. A prospect answers, your dialer bridges to an agent, the agent says hello and the prospect immediately hangs up: that is a one-second connect — technically connected, commercially worthless.
A meaningful connect requires:
- The prospect answers
- The call bridges to a live agent within two seconds of the prospect's greeting
- The prospect engages for at least 20 seconds
That third filter removes garbage connects. Your dialer logs will show more connects than your agents actually work. Reconcile against agent talk time before running CPC calculations.
The CPC Formula With All Cost Layers
Full CPC requires three inputs: total dials, connect rate (on the filtered definition above), and combined cost per dial.
Combined cost per dial = (telecom cost per dial) + (labor cost per dial) + (technology cost per dial)
For a team of 15 agents, 450 dials per agent per day, at a loaded labor rate of $38/hour (benefits included):
| Cost component | Daily total | Cost per dial |
|---|---|---|
| Labor (8 hours × $38 × 15 agents) | $4,560 | $0.675 |
| Technology ($6/agent/day × 15) | $90 | $0.013 |
| Telecom (flat-rate, $4.95/agent/day × 15) | $74.25 | $0.011 |
| Total | $4,724.25 | $0.699 |
At a 14% connect rate (filtered):
CPC = $0.699 ÷ 0.14 = $4.99
Labor is 97% of that number. Telecom is 1.6%.
Why Telecom Still Matters in the CPC Model
If telecom is only 1.6% of CPC, why optimize it? Three reasons:
First, it compounds at scale. A 50-seat outbound floor at $99/seat/month on flat-rate pays $4,950/month in telecom. The same floor on a per-minute model at $0.012/minute, with each agent making 450 dials per day at 90-second average attempt duration (mixing short unanswered and long connected calls), pays approximately $6,075/month. That $1,125/month delta is $13,500/year — enough to fund a full-time QA analyst.
Second, the telecom variable affects labor utilization. When telecom is per-minute and expensive, managers over-rotate toward AMD to reduce connected-call handle time on machines. AMD introduces 1–3 second latency on genuine human picks. That latency increases the rate of prospect hang-ups before the agent speaks. Fewer clean connects per hour means lower agent utilization. Lower utilization means more labor cost per connect.
Third, on per-minute billing, every configuration change has a hidden price tag. Testing a higher pacing ratio, extending campaign hours into lower-connect windows, or adding a new market all have direct telecom cost implications. On flat-rate pricing, those tests are free to run.
Local Caller ID Impact on CPC
The fastest way to reduce CPC — holding labor and technology cost constant — is to raise connect rate. Local caller ID is the most reliable lever for doing that.
| Caller ID type | Typical connect rate (US cold outbound) | CPC at $0.699/dial labor+tech+telecom |
|---|---|---|
| Toll-free (800/888) | 8–11% | $6.36–$8.74 |
| Out-of-state local | 12–15% | $4.66–$5.83 |
| Local match (same area code) | 18–24% | $2.91–$3.88 |
The difference between a toll-free number and a locally matched number at scale — 30 agents, 400 dials/day — is 24–48 more connects per agent per day. At a 3.5% close rate, that is 0.84–1.68 additional closes per agent per day before any change in script, offer, or agent skill.
UnlimCall provisions on-demand local caller ID across 33 live markets. The number is generated fresh at dial time rather than drawn from a rotating pool. Fresh numbers maintain answer rate over the full campaign lifecycle; rotated pools degrade as numbers age onto carrier spam registries.
Tracking CPC Trends Over a Campaign's Life
CPC is not a static number. It worsens predictably as a campaign ages:
- Days 1–7: Fresh list. Connect rate high (18–22% on a good local-match setup). CPC near floor.
- Days 8–21: Mid-penetration. Easy-to-reach contacts are largely exhausted. Connect rate drops to 12–15%. CPC rises 20–40%.
- Days 22+: Deep penetration. Connect rate falls to 8–11%. CPC climbs toward or above the break-even threshold for the offer.
When CPC crosses your offer's break-even, the campaign should be paused and the list refreshed, not squeezed harder. Increasing pacing ratio to compensate for falling connect rates does not lower CPC — it raises it, because more dials are fired for the same number of connects.
Outbound call center solutions built on flat-rate telecom surface this inflection point earlier, because telecom cost is not masking the true CPC trend.
CPC vs. Cost Per Acquisition: Which Drives Decisions
CPC tells you whether your dial strategy is efficient. CPA tells you whether the offer is viable. Use CPC to tune the calling operation; use CPA to decide whether to run the campaign. See also how to model cost per acquisition for outbound programs.
Takeaways
Cost per connect is primarily a labor efficiency metric, but telecom structure determines whether you can run experiments freely and whether your caller ID strategy is depressing connect rates at the source. Flat-rate telecom removes the cost penalty for high-volume dialing and local caller ID on 33 markets raises the connect rate baseline that drives CPC down.
Run the CPC Math on Your Current Setup
See flat-rate seat pricing across all 33 markets and calculate the telecom component of your current CPC.