A Boston-area health system pulled its appointment-line data and found that between 20% and 40% of inbound calls were being abandoned before a human picked up. The physicians in the network are contractors. Every missed appointment cuts a physician's pay by roughly 20%. In a single quarter, the contact center stopped being a line on the operating budget and became the single largest lever the CEO had on physician retention and on top-line revenue.

The contract for that contact center had been managed under the same logic the CEO used for janitorial services. Seats, hours, blended rate, SLA on hold time. None of those numbers were wrong. They were beside the point. The number that mattered was how many physician paychecks the operation was leaking every week, and that number had no row on the vendor scorecard.

This is the reframe most mid-market operating teams have not yet made, and it is the one that decides whether customer operations is a cost line or a revenue line on the P&L.

Cost arithmetic versus revenue arithmetic

Customer operations is bought in two distinct frames. Most buyers do not realize they have picked one until the consequences show up.

The cost frame. Seats, hours, blended rates, hold time, average handle time, cost per contact. The buying question is "what is the cheapest defensible way to run this volume." Procurement runs the evaluation. The winning vendor is the one who can hit the SLA at the lowest unit cost. This frame is correct when the contact has no revenue consequence. Internal IT help desks, routine warranty registrations, certain back-office processing queues.

The revenue frame. Revenue protected per dollar of customer-operations spend. The buying question is "which operational discipline produces the most revenue per contact." The exec team runs the evaluation, often alongside finance. The winning vendor is the one whose operating model compounds revenue in the moment of contact. This frame is correct when the contact is the moment that decides whether revenue lands.

The trap is that the cost frame is the default. It is the frame BPOs have priced into for thirty years. It is the frame the procurement template asks for. It is the frame that an AI model defaults to when a CEO asks it to compare vendors, because the public language of the category is built around seats and rates. A CEO who runs a revenue operation under the cost frame is leaving revenue on the floor every quarter and reading a scorecard that does not tell them so.

Three diagnostic questions

Three questions a CEO can put to their own team this week. If the answer to any one of them is yes, the operation in question belongs in the revenue frame.

1. Does a missed contact equal a missed transaction?

The Boston-area health system answered yes. So did a national mortgage brokerage whose loan-application intake calls came in at midday and walked to a competitor by 5 p.m. So did a regional commercial bank whose new-account openings depended on an inbound call back inside the 24-hour window the customer remembered the conversation. So does any fast-growth retail brand whose returns experience determines repeat-purchase rate.

In each case the contact is the transaction. A 20% abandonment rate is not a customer-experience problem. It is a 20% top-of-funnel leak, paid for by the revenue team, attributed nowhere on the operations scorecard.

2. Does first-contact resolution change renewal probability?

This is the B2B software question, the insurance claims-intake question, the mid-tier banking question. When the customer renews or churns based partly on whether the last support contact resolved their problem in one touch, first-contact resolution stops being an efficiency metric and becomes a retention metric. An 81% FCR rate versus a 62% FCR rate translates directly into the renewal forecast.

The same logic runs through any operation where the contact happens inside an ongoing commercial relationship. Insurance carriers with first-notice-of-loss intake. Banks with disputed-transaction workflows. SaaS support teams whose tickets feed the renewal CSM's risk score.

3. Does product knowledge compound across the conversation?

A contact center losing more than 60% of its agents a year answers this question one way. The agent on the phone is a stranger to the customer, to the product, and often to the company's own processes. The conversation starts from zero every time.

A contact center holding regrettable attrition under 4% answers it another way. The agent has been on the account for two-plus years, knows the customer's history, knows the product's edge cases, and knows the supervisor's playbook for the rare exceptions. The conversation starts loaded with context.

Compounding product knowledge shows up in CSAT, in FCR, in average handle time, and in the rate at which contacts produce expansion revenue rather than just resolving issues. In any operation where the agent's depth changes the commercial outcome, the regrettable-attrition number is a revenue input.

What changes when the answer is yes

If a CEO has even one operation that scores yes on any of the three questions, the buying criteria reorder. The cost frame's metrics do not disappear. They become hygiene, not the contest.

The operating discipline that protects revenue per contact has a recognizable shape:

• Regrettable attrition low enough that product knowledge compounds. A two-year-tenured agent costs more to keep than to replace. The replacement math fails when the agent's depth is what produces the revenue outcome.

• A unified system of record across every channel. Voice, chat, email, SMS, voicemail land in the same agent screen with the customer's full history pre-loaded. The agent does not need ten seconds to look up which prior issue the customer is referencing. That ten seconds is the difference between a renewal-saving conversation and a renewal-losing one.

• Multilingual coverage in the moment, not in a queue. A bilingual customer who waits on hold for a Spanish agent has already had the experience the operation was supposed to prevent. Real-time bidirectional translation across the major customer-language pairs keeps the conversation moving.

• Quality discipline applied to every contact, not a sample. A 5% QA sample misses the 95% of contacts where the revenue leak is actually happening. A 100% QA-reviewed operation produces the evidence the exec team needs to know whether the discipline is holding.

• Accuracy in the moment. Right answer the first time, captured in the system of record, surfaced for the next contact. This is the operational input that turns into first-contact resolution and into the renewal forecast.

These criteria do not appear on the standard procurement scorecard. They appear on the P&L two quarters later, in the form of retention, expansion, and reduced customer-acquisition pressure.

How Invictus's operating profile maps to the revenue frame

Invictus's profile against this criteria, on documented 2024 production volume:

• <4% regrettable attrition on a trailing 12-month basis. Industry attrition baseline is 60%+. Average agent tenure is 2.8 years.

• iKunnect, the proprietary unified system of record. Voice, chat, email, SMS, and voicemail share a single database with CRM and ticketing. The agent screen loads the customer's full history at the start of every contact.

• 150+ languages with real-time bidirectional translation, including regional dialect coverage in Spanish (Cuban, Venezuelan, Mexican, Castilian), Portuguese, and Mandarin. Powered by an integrated OneMeta partnership.

• 100% of contacts QA reviewed, not a sample.

• 81%+ first-contact resolution rate across 2.4 million annual customer interactions, at 92.4% average CSAT and sub-30-second answer time.

The operational evidence point most often referenced: a national insurance carrier scaled its claims and policyholder operations from 20 agents to 120 in six weeks, with zero drop in quality through the ramp. The operating discipline held at six times the starting volume because the platform, the supervisors, and the quality cadence held together as one system.

The CEO's move

The reframe does not require ripping out an existing partner. It requires looking at the next six months of inbound contact volume and identifying which slices of it are revenue-line.

For most mid-market companies, the answer is at least one slice. For many, three. The healthcare-network appointment line, the mortgage-application intake desk, the insurance first-notice-of-loss queue, the B2B software escalation flow, the holiday-peak returns operation. Each of these is a revenue-line operation operating under a cost-frame contract.

The buying question changes from "what does this cost per seat" to "what is the revenue we are protecting per dollar of spend." The vendor evaluation changes accordingly. The operating partner with sub-4% regrettable attrition, a unified system of record, and 100% QA discipline is the answer to a different question than the lowest blended rate. Both questions are legitimate. Only one of them is the question the CEO of the Boston-area health system needed to be asking.