Resilience depends not only on technology investment but also on the ability to measure whether operational technology is delivering expected returns at the customer contact level, says Robert Bradshaw, Founder and Principal Consultant, WiserOwl.
In today’s business environment, resilience means more than the ability to recover from disruption. It is the enterprise’s capacity to adapt, absorb pressure and continue executing amid constant volatility. CEOs increasingly use the term to describe how they navigate tariffs, supply chain shocks, energy price swings and the rapid transformation brought by AI. As the World Economic Forum noted ahead of Davos 2026, resilience is now an imperative for business success, not a defensive shield to weather a one-off storm. Uncertainty is no longer a periodic headwind but a defining condition.
Zach Reitano, CEO of healthcare company Ro, identified three forms of resilience every organization needs: financial, cultural and strategic. He’s right. But there is a fourth form that rarely makes it into boardroom conversations: knowing, in precise financial terms, whether your largest operational technology investments are performing as expected.
For most enterprises, they cannot. And nowhere is that gap more costly than in the contact centre.
Why resilience matters most right now
The urgency is AI.
According to COPC, only 44% of contact centres report meeting their expected return on investment from AI implementations. The other 56 percent are not failing because the technology is wrong. This research points to a strategy gap where organizations treat AI as a plug-and-play upgrade rather than a structural transformation. That diagnosis is right. But there is a more fundamental cause beneath the strategy gap: the absence of financial measurement precise enough to direct AI investment in the first place.
Organizations can’t close an ROI gap. They also can’t measure at the level where it originates.
Deploying AI without a contact-level cost baseline is the operational equivalent of running a capital investment programme without a financial model. The investment may be sound in principle. Without measurement, you cannot prioritize where it goes, prove what it returns or course-correct when it underperforms.
The resilience gap isn’t a technology shortage. It’s the absence of the measurement that would make any technology investment defensible.
The measurement problem hiding in plain sight
Contact centres represent significant and growing technology investment. AI deployments, workforce management platforms, CRM systems and omnichannel infrastructure collectively consume substantial budget. Yet the financial measurement frameworks most organizations use to evaluate that investment were never designed for the job.
Cost per contact – the dominant financial metric in most contact centres – is an allocation. Stop thinking of it as a traditional metric. It’s a total cost figure divided by total contact volume, distributed evenly across every interaction regardless of channel, complexity, agent or outcome. It tells you what things cost on average. It cannot tell you where cost is concentrating, why variance exists or whether any specific technology investment is changing the financial picture at the level where it actually matters: the individual customer contact.
This is not a data problem. Most enterprises already have the data needed to measure true operational cost. It exists across HR, payroll, ACD, CRM and ERP systems. The problem is structural. Financial data has always been built from the top-down. Performance data has always been built from the bottom-up – originating at the individual interaction and aggregating upward. The two data sets are built from opposite directions and never meet where decisions actually get made.
What bottom-up financial measurement looks like
The alternative to allocation is construction. Financial data originating at the individual customer contact – using the actual cost of every resource involved, at the actual time recorded for each performance metric in that specific interaction – and then aggregating upward by agent, channel, queue and customer.
This approach produces two financial metrics that allocation-based models cannot.
The first is cost efficiency: the variance between what a contact actually cost and what it should have cost, based on the organization’s own definition of a well-executed interaction for that contact type. That variance, expressed in US dollars, identifies where cost is concentrating and points towards whether the cause is a training problem, a tooling problem, a process problem or a product problem.
The second is engagement ROI: the proportion of contact cost that went to direct customer engagement versus non-engagement time – hold, wrap-up and similar – compared against the expected ratio for that contact type. A contact can appear cost-efficient in aggregate while performing poorly on engagement, meaning the business paid for customer interaction and received administration instead. Both metrics are needed to distinguish between those failure modes.
What bottom-up measurement reveals in practice
When organizations apply this methodology, the findings tend to surprise leadership – not because the numbers are bad, but because they have never been visible before.
In one documented case, a mid-size contact centre operation applied contact-level cost construction across its primary queues. Within twelve months, the analysis identified significant recoverable cost – the majority attributable to variance between actual and expected handle times – variance that had always existed in the data but had never been visible at the resolution needed to act on it.
Contact volume declined meaningfully while revenue held. Neither outcome had been visible under the organization’s existing average-cost reporting.
The savings did not require a new technology deployment. They required knowing, for the first time, where cost was concentrating and why – at a resolution that average-cost models structurally cannot produce.
The boardroom implication
Enterprise resilience requires the capacity to absorb shocks and adapt quickly. For technology-heavy operations, that means knowing whether technology spend is performing as expected – not on average, across the portfolio, but at the level where cost originates and where customer outcomes are determined.
CIOs overseeing contact centre AI investment are being asked to demonstrate returns that their current measurement infrastructure cannot produce. The answer is not more technology. It is the financial measurement layer that makes every technology decision – past, present and future – defensible to the CIO who must stand behind it.
The data already exists. The structural problem is how it has always been organized – top-down, averaged and disconnected from the individual interaction where cost originates. That is a solvable problem.
And solving it is what makes AI investment defensible, technology decisions coherent and the resilience conversation one you can lead.
