WISMO Reduction in 2026: Why Most of Your Contacts Come From Deliveries That Arrived On Time

WISMO reduction is the work of lowering “where is my order” contact volume, and it usually fails because the volume is treated as one thing. It is two. Some contacts come from customers whose delivery genuinely breached the promised window, and some come from customers whose delivery was running perfectly to plan but who had no way to know that. The two populations respond to completely different investments, they move in opposite directions as an operation improves, and almost no customer experience team measures them separately. Locus, the world’s first Decision-Intelligent, Agentic TMS, addresses both, predicting the breach early enough for a message to matter and holding one live order state that the customer can see.

Why WISMO Volume Resists the Obvious Fixes

Contact volume is expensive in a way that is easy to underweight, because it sits in a different budget from the delivery. McKinsey’s out-of-home delivery work puts the last mile at 60% to 70% of total parcel delivery cost, and the contact a delivery generates is paid for separately, by a team measured on handling time rather than on delivery performance. That separation is the reason the two halves of this problem are rarely looked at together.

The structural difficulty is that most European operations no longer control the vehicle. AlixPartners’ 2026 Home Delivery Survey found more than 90% of executives run a mix of last-mile carriers and 32% use four or more. Every one of those carriers reports on its own schedule and in its own vocabulary, so the customer-facing answer to “where is my order” is assembled from feeds of uneven quality, and the quality varies by which carrier happened to take the parcel.

Gartner’s customer service research points at the same gap from the other side, finding that 45% of customer service organisations report an inability to connect data across the systems they depend on. A team that cannot connect delivery execution data to contact data cannot tell which of its contacts came from a late delivery and which came from a silent one, which means it cannot tell which investment would reduce volume.

Conditions in Europe also make the promise harder to hold in the first place. INRIX’s 2025 UK Traffic Scorecard found London drivers lost 91 hours to congestion, and although that was a 10% improvement on the previous year, London still accounted for nearly half of the country’s total delays. Variance of that size is absorbed either by a wider promise or by a system that can re-decide during the day.

Also Read: Delivery Promise Accuracy Under Load: The 12x WISMO Math

Your Contacts Come From Two Different Populations

A customer contacts you about an order for one of two reasons. Either something went wrong and they noticed, or nothing went wrong and they could not establish that.

The first population is bounded by the breach rate. If 5% of orders miss the promised window, then at most 5% of customers have a genuine failure to ask about. The second population is bounded by everything else, which is the other 95%, and only a small fraction of them need to contact you for that fraction to be the larger number.

We modelled the arithmetic. The inputs are illustrative rather than measured: a share of breached customers who contact, a much smaller share of on-time customers who contact anyway, applied across 100,000 orders.

Breach rate	Contacts from breached deliveries	Contacts from on-time deliveries	Total	Share from on-time
2%	1,100	3,920	5,020	78.1%
5%	2,750	3,800	6,550	58.0%
7%	3,850	3,720	7,570	49.1%
10%	5,500	3,600	9,100	39.6%
20%	11,000	3,200	14,200	22.5%

Read the top row first. At a 2% breach rate, which is the performance most enterprise operations are trying to reach, roughly four in five contacts come from deliveries that arrived exactly as promised. The customers were not complaining. They were asking, because asking was the only way to find out.

The bottom row is the one that makes the point by contrast. A badly performing operation at a 20% breach rate genuinely does have a failure problem, and just over three quarters of its contacts are about real failures. Its contact volume is nearly three times higher, and its priorities are obvious.

What the table says about the middle is less obvious and more useful. As an operation improves, its total contact volume falls, but the composition of what remains shifts steadily towards customers who had a perfectly good delivery. The residual volume becomes progressively less addressable by anything the operations team can do.

The Crossover Point Most Operations Sit Below

There is a specific breach rate at which the two populations are equal in size. Below it, uncertainty generates more contacts than failure does. Above it, failure dominates.

Contact rate among on-time customers	Contact rate among breached customers	Crossover breach rate
2%	55%	3.5%
3%	55%	5.2%
4%	45%	8.2%
4%	55%	6.8%
5%	55%	8.3%
6%	45%	11.8%

Across a plausible range of inputs the crossover falls somewhere between 3% and 12%, and around 7% for the central case. That number is worth holding onto, because it divides two quite different businesses that use the same vocabulary.

An operation running above the crossover has a delivery problem that shows up as a contact problem, and fixing the delivery fixes both. An operation running below it has a communication problem wearing a delivery problem’s clothes, and no amount of operational improvement will resolve it, because the deliveries were already fine.

Where an operation sits relative to that line is something only its own data can answer, and the answer is unlikely to hold all year. An operation comfortably below the crossover on ordinary trading can cross it during a peak, which means the correct investment shifts seasonally, and that is not how notification and tracking programmes are usually scoped.

Also Read: European Grocery Delivery: 5 Experience Strategies That Fail

What Each Lever Can Actually Achieve

Once the population is split, every intervention acquires a ceiling, because each one only addresses part of the volume. Taking the 5% breach case above as the baseline, at 6,550 contacts split 42% breach-driven and 58% uncertainty-driven:

Investment	Total contacts	Change
Baseline	6,550	n/a
Predictive notification alone, deflecting 70% of breach contacts	4,625	?29.4%
Visible live tracking alone, removing 60% of uncertainty contacts	4,270	?34.8%
Operational improvement alone, halving the breach rate	5,275	?19.5%
Notification and tracking together	2,345	?64.2%
All three	1,972	?69.9%

Three things in that table are worth arguing about internally.

The first is that halving your breach rate, which is a substantial and expensive operational achievement, reduced contacts by under 20%. That result reliably surprises the people who funded it, and it is not a failure of execution. It is the arithmetic of improving the smaller of two populations.

The second is that predictive notification, the standard answer in this category, caps out at roughly 29% even when it works well, because it can only reach customers who are going to have a problem. It is a good investment and it cannot be the whole programme.

The third is that the two communication levers together achieve more than either plus operational improvement, because they are the only pair that addresses both populations. That is the case for treating notification and tracking as one programme rather than as separate line items, which is how they are usually budgeted.

How to Split Your Own Contact Population

1 Join contact records to delivery outcomes at order level

Take every contact from a defined period and attach the promised window and the actual delivery time for that order. This is the whole exercise, and the reason it is rarely done is that the two datasets usually live in different systems owned by different functions.

2 Classify each contact against the promise, not against the plan

A contact is breach-driven only if the delivery missed the window the customer was shown at checkout. Measuring against an internally revised window will classify breaches as on-time deliveries and understate the failure population.

3 Timestamp the contact against the delivery, not the day

A contact placed before the window opened is a different thing from one placed after it closed. The first is almost always uncertainty. The second is usually a breach. The distribution of contact timing relative to the promised window is the cleanest single diagnostic available.

4 Compute the two contact rates separately

Divide breach-driven contacts by breached orders, and on-time contacts by on-time orders. These two percentages are the model inputs for your own operation, and they are far more useful than a blended contact rate, which moves whenever the mix moves.

5 Re-run the split by carrier and by market

Contact rates among on-time customers vary sharply with how visible the journey is, which in a multi-carrier European network varies by carrier and by country. The markets with the highest uncertainty contacts are usually the ones with the sparsest carrier reporting, not the worst delivery performance.

6 Set each lever a target against its own population

Give the notification programme a target on breach-driven contacts and the tracking programme a target on uncertainty-driven contacts. A shared target on total volume lets each hide behind the other, and neither can move the whole number.

Also Read: How Many Messages Does One Delivery Deserve?

Breach-Driven and Uncertainty-Driven Contacts Compared

Dimension	Breach-driven contact	Uncertainty-driven contact
Trigger	The delivery missed the promised window	The customer could not establish the order’s status
Timing relative to the window	Usually after it closes	Usually before or during it
Population size	Bounded by the breach rate	Bounded by everything else
Behaviour as operations improve	Falls with the breach rate	Largely unchanged
What reduces it	Lead time on a predicted breach	A visible, current and trusted order state
Customer sentiment	Complaint or recovery request	Information request
Where it shows up in reporting	Correlates with on-time performance	Correlates with nothing operations measures

The last row explains why this split stays invisible. Uncertainty-driven contacts do not correlate with any operational metric, because operationally nothing happened. They look like unexplained baseline noise on a contact dashboard, and unexplained baseline noise is exactly what gets attributed to seasonality and left alone.

What to Look For in Delivery Experience Software

A single order state rather than a carrier status feed. The customer needs one current answer, not a list of scan events from whichever carrier holds the parcel. Ask how the platform harmonises statuses across carriers into one vocabulary, because that harmonisation is what makes a tracking page answer the question rather than display data.

Breach prediction with usable lead time. A notification sent when a breach is confirmed deflects nothing, because the customer has already noticed. What matters is how far ahead the platform detects that a promise is going to fail, and whether that detection fires a message automatically.

Freshness carried as data. A tracking page that shows a position without showing how old it is will confidently display stale information, which generates contacts rather than preventing them. The platform should distinguish “confirmed moving two minutes ago” from “no information for four hours”.

Contact attribution built in. The platform should be able to tell you which contacts came from breached orders and which did not, without a data engineering project. Gartner’s finding that 45% of service organisations cannot connect their data is the default condition this has to overcome.

Re-decisioning, not just re-reporting. Detecting a breach only helps if something happens next. The platform should be able to re-plan, re-sequence or re-promise on the strength of its own detection, because a faster alert into the same manual queue produces the same outcome slightly earlier.

Also Read: Predictive Exception Management for European Grocers

Contact Reduction in Practice

A leading ASEAN apparel retailer running a large store network alongside a global e-commerce business, with last mile almost entirely through carriers, cut WISMO and returns queries by more than 40% through multi-carrier parcel management. The mechanism is the one this article argues for: every carrier’s status codes were harmonised into a single standard set synced to the order systems, and a network-aware delivery date was computed at checkout. Before the change the operation had handled hundreds of thousands of delivery and returns complaints in a single half-year while each carrier reported its own way. The reduction came from making the answer available, not from delivering faster.

A Canadian grocery brand delivering fresh and perishable orders to homes across more than 30 cities through contracted third-party fleets saw customer support resolution run 10 to 20 times faster after carrier orchestration, alongside 33% faster deliveries and 15% lower fulfilment cost. Resolution speed is the uncertainty population’s metric rather than the breach population’s, because it measures how quickly an agent can establish what is actually happening to an order that may be entirely fine.

Both cases share a shape worth noting. Neither operation reduced contacts primarily by improving delivery performance. Both reduced them by making the order’s true state legible, first to the agent and then to the customer.

Common Mistakes in WISMO Reduction Programmes

Treating contact volume as a single number. A blended contact rate moves whenever the mix between the two populations moves, which makes it impossible to tell whether an intervention worked or the composition shifted underneath it.

Assuming better delivery performance will fix it. Halving the breach rate reduced contacts by under 20% in our model, because the larger population was never about failed deliveries.

Adding messages instead of adding lead time. More notifications against an unreliable estimate re-anchor the customer on successive wrong times, which generates the contacts the messages were bought to prevent.

Measuring tracking success by page views. Page views rise when customers are uncertain, so a tracking page that is heavily used may be evidence of the problem rather than of the solution. Contact volume among on-time deliveries is the honest measure.

How Locus Reduces Both Contact Populations

Locus, the world’s first Decision-Intelligent, Agentic TMS, works on both populations because it holds the promise, the plan and the live execution state in one decision layer rather than in three connected systems. The Control Tower presents one current state per order across every vehicle and carrier, with statuses harmonised into a single vocabulary, so the answer a customer sees is the same answer the operation holds. That addresses the uncertainty population directly, because uncertainty is a property of the information rather than of the delivery.

For the breach population the relevant capability is lead time. Because the route planning engine evaluates the plan continuously against more than 250 real-world operating constraints rather than only at dispatch, a promise that is going to fail becomes detectable while the delivery is still in progress, which is the window in which a message deflects a contact instead of confirming one. The Customer agent runs that communication and the Dispatch agent holds the reallocation, so detection can trigger a re-plan rather than only a notification. DiSCO governance mechanisms, including Explainability and Autonomy Levels, determine which of those actions the system takes without a human.

Locus has been recognised by Gartner for seven consecutive years across multiple research categories, including Representative Vendor status in the 2026 Gartner Hype Cycle for Supply Chain Execution and Logistics Technologies and the 2026 Gartner Market Guide for Multicarrier Parcel Management Solutions, where ShipFlex is featured as a Representative Vendor. QKS Group positions Locus as the Leader in its SPARK Matrix for Transportation Management Systems 2025, and G2 ranked Locus number one in Route Planning in its 2026 Best Software Awards. The platform has run more than 1.5 billion deliveries for 360+ enterprise customers across 30+ countries at 99.99% uptime.

In October 2025, Ingka Investments, the investment arm of Ingka Group, the world’s largest IKEA retailer, acquired Locus. Locus continues to operate independently.

The change worth making first costs nothing and takes a morning. Join last month’s contacts to last month’s delivery outcomes and work out what share of them came from orders that arrived inside the promised window. If that share is above half, and at ordinary breach rates it usually is, then the programme you are funding to improve delivery performance is aimed at the smaller half of your contact volume, and the larger half is waiting on something you have not bought yet. Locus holds one live order state and predicts the breach early enough for a message to matter. Speak to a Locus specialist about reducing contact volume across your European network.

Also Read: Delivery Experience Platform: Lift NPS, Cut WISMO

Frequently Asked Questions

What causes WISMO contacts? Two separate things. Some customers contact you because their delivery missed the window they were promised, and some contact you because their delivery was proceeding normally but they had no reliable way to confirm that. The second group is usually the larger one in a well-run operation, because on-time deliveries vastly outnumber late ones.

How do you reduce WISMO calls? By working on both populations rather than one. Lead time on a predicted breach reduces contacts from customers whose delivery is failing, and a visible, current and trusted order state reduces contacts from customers whose delivery is fine. In our illustrative model each lever alone reached roughly 30% to 35%, while the two together reached 64%.

Does improving on-time delivery reduce WISMO volume? Less than most teams expect. In our model, halving the breach rate from 5% to 2.5% reduced total contacts by 19.5%, because the majority of contacts were coming from deliveries that were already arriving on time. Operational improvement works on the smaller population.

What proportion of WISMO contacts come from on-time deliveries? It depends on the breach rate, and the relationship is steep. In our illustrative model, an operation at a 5% breach rate saw 58% of contacts come from on-time deliveries, and an operation at 2% saw 78%. The better the operation performs, the more of its remaining contact volume is about uncertainty rather than failure.

Why do proactive delivery notifications have limited effect on their own? Because they can only reach the population that is going to have a problem. Even a notification programme that deflects 70% of breach-driven contacts reduced total volume by under 30% in our model, since it never touches the customers whose delivery is running normally and who simply cannot see it.

How should a CX team measure delivery experience software? On two numbers rather than one: contact volume among breached orders and contact volume among on-time orders. A blended contact rate moves whenever the mix shifts, which makes it impossible to attribute a change to a specific investment, and it hides the population that most tooling never addresses.