Failure Demand in Contact Centres
Failure demand is customer contact caused by the business failing to do something — or failing to do something right — the first time. It's the calls that shouldn't exist: customers chasing status updates, re-explaining their issue, correcting the organisation's mistakes, or calling back because the previous interaction didn't actually resolve anything. In most contact centres, it's a larger share of volume than anyone admits, and it's the single biggest lever available for reducing cost while improving service.
Why it matters
Failure demand typically accounts for 20–60% of contact centre volume — and closer to 80% in public sector and utilities. It's the invisible expense on every cost line, because the business treats failure demand as "normal demand" and staffs to it instead of designing it out. Every call you remove is pure margin.
Why it hides
Failure demand looks exactly like value demand on the dashboard. A customer calling "how do I check my account balance?" is value demand. A customer calling "why hasn't my refund arrived after two weeks?" is failure demand. The ACD counts both the same way — as one call — and the business staffs to the total.
What this guide covers
A precise definition of failure demand vs value demand, current industry data, the systemic causes that create it, a worked example of the commercial impact, how to identify it in your own operation, and the specific design changes that reduce it.
What is Failure Demand?
The concept was codified by management thinker John Seddon and the Vanguard Consulting team, who drew a distinction between the two reasons any customer contacts a service organisation. Every single customer interaction falls into one of two categories:
✓ Value demand
- Demand you want — why the customer is here from their point of view
- The customer's original need (new account, product enquiry, quote)
- The reason the contact centre exists
- Genuine interactions that create value for both sides
✕ Failure demand
- Demand caused by a failure to do something, or do something right, for the customer
- "Where's my order?" / "You charged me twice" / "I explained this to the last person"
- A symptom of the organisation failing, not the customer needing something
- Work that wouldn't exist if the service was designed properly
Plain-English Definition
Failure demand is the calls that shouldn't exist. If the business had done its job right the first time, the customer wouldn't be contacting you now. Every "I'm just following up…" is failure demand. Every "it didn't work…" is failure demand. Every "I was told…" is failure demand. The customer is doing the business's work for it — chasing, correcting, re-explaining — and paying for the privilege with their time.
Why the name matters
Some organisations prefer to call it "repeat contacts," "avoidable contacts," or "non-value-added interactions" because "failure" feels uncomfortable. That discomfort is exactly the point. Calling it failure demand forces the business to own it — because it is the organisation's failure, not a customer characteristic to be managed around. Softening the language softens the accountability, and the problem persists.
Failure Demand vs Value Demand — How to Tell the Difference
The distinction looks obvious in the abstract but gets messy in practice. The test is always from the customer's point of view: would this contact exist if the business had done its job properly? If no, it's failure demand. If yes, it's value demand.
Clear value demand
"I want a quote for home insurance." "Can you tell me about your business banking products?" "I need to open a new account." These are all genuine first-time interactions where the customer is coming to you with a need the business is designed to serve.
Clear failure demand
"Where's my order, it's been three weeks?" "I was told last Tuesday someone would call back." "You've charged me for something I cancelled." "This is the third time I've explained this." All are the organisation's failures showing up as customer contacts.
The tricky middle ground
"How do I reset my password?" could be value demand if it's genuinely new, but it's failure demand if the reset flow is so badly designed that customers routinely need a human. The test is prevalence: if a significant share of customers need help with the same self-service task, the self-service design has failed.
The counterintuitive insight
Most contact centres are staffed and measured as though all demand is value demand — that every call is a legitimate interaction requiring a response. The failure demand lens flips this: a meaningful proportion of the work shouldn't be coming in at all, and the right response isn't to handle it faster, it's to stop creating it. This is why failure demand is the single biggest performance lever most contact centres have — not because the calls are handled badly, but because they shouldn't exist.
The Scale of Failure Demand — What the Data Says
Failure demand is rarely measured properly, which is why its true scale surprises most leaders when they first see it. Here's what the data shows when organisations actually look.
Failure Demand as a Share of Total Contacts — By Sector
Sources: Vanguard Consulting research across UK service organisations; 2024 Australian Contact Centre Industry Best Practice Report.
Reading this honestly
The Australian 31% figure comes from the 2024 Best Practice Report — the most recent large-scale Australian dataset. It's self-reported, and almost certainly understated. Most contact centre managers have never formally measured failure demand; they've estimated based on gut feel, or conflated it with "repeat contacts," which is a narrower definition. Organisations that actually undertake structured demand analysis consistently discover their real failure demand rate is considerably higher than they thought — often by a factor of two or more.
The 20–60% range for financial services and the 80–90% for public sector and utilities come from Vanguard Consulting's demand-analysis work across service organisations. These numbers have held up as the reference data because nobody else has done comparable measurement at scale — and organisations that repeat the exercise today keep finding the same patterns. The data isn't old; it's canonical.
The practical takeaway: whatever you think your failure demand rate is, it's probably higher. And even the conservative 31% figure means nearly one in three contact centre interactions in Australia is the organisation's failure showing up as cost.
The Commercial Cost — A Worked Example
Failure demand hides because it doesn't appear as a line on any cost report. It's absorbed into general contact centre operating costs, which are then treated as "normal." But the commercial impact becomes obvious the moment you actually calculate it.
A mid-size contact centre — the hidden cost of failure demand
Take a contact centre handling 500,000 voice interactions per year — a medium-sized operation by Australian standards. Industry-standard fully-loaded cost per contact sits around $8–$12 for voice in 2026. Use $10 as a working average.
That's $1.5 million being spent every year, in a mid-size operation, handling contacts that shouldn't have existed in the first place. And that's the conservative scenario.
Now apply the more realistic figure discovered by organisations that actually measure failure demand properly — often 50% or higher:
Why this isn't just a cost story
The commercial case alone is compelling, but it understates the opportunity. Failure demand also creates downstream costs the contact centre doesn't see directly: complaint handling workload, escalation time, regulatory reporting, agent attrition from handling frustrated customers, damage to brand reputation, and customer churn that cost-per-acquisition has to replace. The true cost of failure demand is typically two to three times the direct contact centre figure once these are accounted for.
In the other direction: every percentage point of failure demand removed is pure margin. You don't just save the cost of the call — you free up the agent time for value demand, reduce the workload on downstream functions, and stop creating the complaint events and churn that compound over time.
What Causes Failure Demand?
Failure demand is systemic. It isn't caused by individual agents making mistakes, or customers being unclear, or "the market changing." It's caused by the way services are designed and managed. Six causes account for the majority of failure demand in Australian contact centres.
Process failure — the organisation didn't deliver
Orders not shipped on time. Refunds not processed. Transfers that didn't go through. Applications sitting in a queue nobody owns. The customer calls because the process failed, and the contact centre is left absorbing the contact the process created.
Communication failure — the customer wasn't told
The customer wasn't informed about a change, delay, or requirement. They were told something different last time. The confirmation email never arrived. They're calling because they're operating on missing or wrong information the business never corrected.
Design failure — the service is too hard to use
The website doesn't work on mobile. The self-service portal requires information the customer doesn't have. The IVR routes to the wrong team. The form rejects the customer's postcode. The channel looks synchronous but operates asynchronously. Design choices that force the customer into the contact centre.
First-contact-resolution failure — it wasn't fixed
The previous agent closed the ticket without actually resolving the issue. The promised callback never happened. The fix didn't work. The customer's second, third or fourth contact chasing the same issue is entirely failure demand, driven by the contact centre's own handling practices.
Target-driven behaviour — AHT pressure closing issues prematurely
Agents measured on handle time are incentivised to close contacts quickly, which means closing them before they're properly resolved. The short-term productivity win creates the repeat contact that drives failure demand up. This is the classic case of activity targets creating the cost they were meant to reduce.
Digital-by-default pushing the wrong work to humans
Chatbots that can't actually resolve complex issues. IVR trees designed to deflect rather than route. Self-service flows optimised for the business, not the customer. The interactions that land in the contact centre are now systematically harder — and the "deflected" interactions that didn't get resolved come back later as failure demand anyway.
The systemic insight
None of these causes live inside the contact centre. They all live somewhere else in the business — operations, IT, marketing, policy, digital — and they arrive at the contact centre's door as work to be handled. This is why failure demand reduction is fundamentally a cross-functional exercise, not a contact centre optimisation exercise. Fixing failure demand means changing what happens upstream, not getting better at absorbing the consequences downstream.
Real-World Examples of Failure Demand
Failure demand is everywhere once you know how to see it. Here are patterns observed across Australian and international contact centres — all of them high-frequency, all of them symptoms of upstream design rather than customer behaviour.
The visible face of failure demand. Every frustrated contact is the business's failure to get it right the first time — and every one of them costs you twice: once in handling cost, once in customer goodwill.
"Where's my…?"
Customers chasing the status of an order, a refund, a claim, an application, a callback. If this category is significant in your contact volume, it's a signal that your status communication is broken — the customer shouldn't need to ask.
"I was told…"
Customer calling back because a previous interaction created an expectation the business then failed to meet. Each one represents a mismatch between what an agent promised and what actually happened — either the agent was wrong, or the agent was right and the process failed.
"I've explained this already"
Customer repeating their story because the previous contact wasn't resolved, wasn't logged, or wasn't accessible to the current agent. Every iteration of this is a sign that case management is failing — and a direct driver of frustration and escalation.
"Your website doesn't…"
Customer calling because self-service didn't work for them. Every one of these is a failure of the digital channel that's now being absorbed by the live channel, at many times the cost-per-interaction.
"I got a letter/email/SMS saying…"
Customer responding to an outbound communication that was unclear, poorly timed, or asked them to do something they shouldn't have had to do. The business created the inbound contact with its own outbound comms.
"This is the [3rd/5th/10th] time…"
The customer has already contacted you multiple times about the same unresolved issue. By this point the failure demand has compounded into a complaint, and the commercial cost has multiplied. For more on the handling side, see the ACXPA Complaints Hub.
How to Identify Failure Demand in Your Operation
Most contact centres don't measure failure demand because they don't have a structured way to identify it. The process is not technology-led — it's disciplined listening and categorisation. Here's the method that works.
Sample real interactions — don't start with reports
Listen to (or read) a representative sample of 100–300 actual customer contacts across channels. Call recordings, chat transcripts, email threads, social media exchanges. The work is qualitative — categorising what the customer actually wanted and why they contacted you. Starting with existing ACD reports or disposition codes will reproduce the categories the business already uses, which is exactly what's hiding the failure demand.
For each contact, ask: "Why is the customer here, from their point of view?"
This is the Vanguard test. Frame the question from the customer's perspective, not the business's. "The customer called to ask about their bill" is the business view. "The customer called because their bill was wrong and the website wouldn't let them dispute it" is the customer view. The second framing reveals failure demand.
Classify each contact as value demand or failure demand
Value demand is what the customer wanted to happen. Failure demand is what went wrong that the customer is now having to correct. A single contact can contain both — the customer wanted a quote (value) but had to chase it three times because the first two agents didn't respond (failure). Count the failure-demand element.
Categorise the failure demand by cause
Group the failure demand contacts by the underlying cause: process failure, communication failure, design failure, FCR failure, target-driven behaviour, digital-by-default failure. Most operations find 80% of failure demand concentrates in 20% of cause categories — giving you a priority list.
Trace each major cause back to its upstream origin
"Where's my refund?" contacts aren't solved inside the contact centre — they're solved by fixing the refund process in finance, or by adding proactive status communication. Each major failure demand category has an upstream owner. Identify them. Failure demand reduction is cross-functional from this point forward.
Report the financial impact in cost terms, not volume terms
"15% of our volume is refund-chasing contacts" doesn't get executive attention. "We spent $450,000 last year handling refund-chasing contacts that shouldn't have existed" does. The cost framing is what unlocks cross-functional investment in upstream fixes — because the upstream function now has skin in the game.
How to Reduce Failure Demand
Reducing failure demand is not about handling it faster, routing it better, or deflecting it to cheaper channels. All three approaches leave the cause in place and just shuffle the cost around. The only sustainable approach is to design the failure demand out — stop creating the contacts in the first place.
Fix the upstream process
If 15% of your contacts are "where's my refund?", the answer isn't a better refund-enquiry script. It's reducing the refund processing time and adding proactive status notifications so the customer doesn't need to chase. The contact centre should be escalating the upstream problem, not absorbing it.
Proactively communicate
Most "where's my…?" failure demand can be eliminated by telling the customer what's happening before they ask. Status updates via SMS, email, or in-app notification cost fractions of a cent per customer. Each one prevents a multi-dollar contact centre interaction.
Prioritise first-contact resolution over handle time
AHT-optimised operations systematically generate failure demand by closing contacts before they're properly resolved. FCR-optimised operations take longer per contact but generate fewer of them overall. Net cost is lower, customer effort is lower, and failure demand drops.
Redesign self-service for actual customer needs
Most self-service channels are designed around the business's categorisation, not the customer's task. Rebuild around what customers actually try to do — which you learn from the demand analysis above — and a significant share of the "the website doesn't…" failure demand disappears.
Make failure demand visible cross-functionally
Share the failure demand cost data with upstream functions monthly. When operations sees they're generating $400k/year of contact centre cost through refund delays, they have a reason to fix it. Without the data, the contact centre absorbs the cost silently and the upstream function has no pressure to change.
Measure failure demand as a standing KPI
If you only measure it once during a demand-analysis exercise, it creeps back up. Establish failure demand rate as a reported KPI at the same level as service level and CSAT — and trend it monthly. What gets measured gets managed; what's measured as a percentage of total volume is what actually moves over time.
Common Pitfalls in Tackling Failure Demand
Treating failure demand as a contact centre problem
Failure demand is generated upstream and arrives at the contact centre's door. Framing it as a contact centre KPI to improve (rather than a cross-functional outcome to redesign for) means you'll spend money on contact centre tools and see no reduction.
Buying technology to "manage" failure demand
Plenty of vendors will sell you software to track, categorise, and report failure demand. None of them reduce it. The cause is systemic; the solution is cross-functional redesign. Technology that logs failure demand more efficiently doesn't make it go away — it just measures the same waste more accurately.
Deflecting to cheaper channels
Pushing failure demand from voice to chat, or from chat to chatbot, reduces per-contact cost but doesn't remove the contact. In many cases it increases failure demand further, because the cheaper channel resolves the issue less effectively, generating repeat contacts. Net volume rises while per-contact cost falls.
Optimising AHT while failure demand rises
The classic false economy. Shorter handle time, more contacts closed per hour, less time per customer — and failure demand rises because issues aren't being properly resolved. The handle time savings are wiped out by the repeat-contact cost, but the KPI dashboard shows green because AHT is the metric being watched.
Reporting failure demand as a single number
"Our failure demand is 28%" gives executives nothing to act on. "Refund chasing is 8%, service outage complaints are 6%, website error calls are 5%" gives specific upstream owners specific problems to fix. Aggregate reporting hides the actionable categories.
Measuring once and declaring victory
A one-off demand analysis produces a one-off number. Without sustained measurement and reporting, failure demand creeps back as processes drift, new products launch, and digital channels change. Failure demand reduction requires the discipline of ongoing measurement, not a single project.
Frequently Asked Questions
What's the simplest way to explain failure demand to my executive team?
Failure demand is contact you created. Value demand is contact you wanted. If the customer is calling because your business did something wrong, didn't tell them something, or failed to follow through — that's failure demand, and it's pure cost. The commercial question is: what percentage of your contact centre budget is spent handling contacts that shouldn't have existed?
How is failure demand different from "repeat contacts"?
Repeat contacts are one subset of failure demand — customers calling back because a previous interaction didn't resolve their issue. Failure demand is broader: it includes any contact caused by the business failing to get it right the first time, even if it's the customer's first contact on that specific issue. A customer calling for the first time about a delayed refund is failure demand, not a repeat contact.
What failure demand rate should we aim for?
There's no universal target rate, and treating failure demand as a KPI to hit creates perverse incentives (teams learn to classify contacts as value demand to meet the number). The better question is: what's the trend? Failure demand should be falling year-on-year as upstream causes are fixed. If it's stable or rising, the design changes aren't working.
How do I sell this internally when failure demand fixes live outside the contact centre?
Lead with cost, not with process. Executives respond to "we spent $1.8M last year handling contacts that shouldn't have existed" — they don't respond to "our refund process needs improvement." Show the money first, then present the upstream fixes as the investment case. The finance function and the upstream operational owner become allies once they understand the scale of the hidden cost.
Can AI and chatbots reduce failure demand?
AI can reduce the cost of handling failure demand — but it doesn't reduce the failure demand itself, and often increases it. A chatbot that incompletely resolves an issue creates a second contact when the customer calls back. The cause of the failure demand is still upstream; putting AI on the contact centre side is optimising the wrong layer.
Is failure demand the same as an "avoidable contact"?
The terms overlap significantly but aren't identical. "Avoidable contact" tends to be used in efficiency language — a contact the business could have avoided having — whereas "failure demand" is specifically contact caused by the business's failure to do its job. Most avoidable contacts are failure demand; the language of failure demand is more precise and pushes toward cross-functional accountability rather than contact centre efficiency.
How long does a failure demand reduction programme take?
Initial demand analysis takes 4–8 weeks. First wave of upstream fixes typically delivers visible reduction within 3–6 months. Structural reduction (rebuilding self-service around customer tasks, rewiring cross-functional accountability, changing KPIs) takes 12–18 months. Programmes that promise failure demand reduction in 30 days are selling you technology, not results.
Where does failure demand fit alongside other contact centre metrics?
Failure demand sits above other operational metrics — it measures whether the demand hitting your contact centre should be there at all. Once that's understood, service level tells you how well you're responding to the demand that arrives, and abandonment rate tells you how many customers gave up on it. You want good performance on all three, but if failure demand is high, the other two are measuring the wrong thing well.
Where to Next
Summary
Failure demand is customer contact caused by the business failing to do something, or failing to do something right, the first time. It's the calls that shouldn't exist — chasing refunds, correcting errors, re-explaining issues, following up on promises that weren't kept. Australian contact centres self-report around 31% of contacts as failure demand on average, but the true figure is almost certainly higher because most organisations have never measured it properly. International research suggests 20–60% in financial services and up to 80% in public sector and utilities.
Commercially, failure demand is the single biggest lever most contact centres have. A mid-size operation handling 500,000 contacts per year at $10 per contact spends $1.5–$2.5 million annually handling contacts that shouldn't have existed — plus the downstream costs in complaints, escalations, attrition and churn. Every percentage point of failure demand removed is pure margin, plus freed-up capacity for value demand.
Reducing failure demand isn't a contact centre optimisation exercise — the causes live upstream, in process failure, communication failure, design failure, first-contact-resolution failure, target-driven behaviour, and digital-by-default mis-design. The approach that works is structured demand analysis (listen to real customer contacts, classify them, categorise the failure demand by cause, trace each cause to its upstream owner, and report the cost cross-functionally). Technology that logs failure demand more efficiently doesn't remove it. Deflection to cheaper channels doesn't remove it. Only upstream redesign does.
