Reducing customer effort where it's actually created

What reducing customer effort actually means

Customer effort is the total work a customer has to do to get what they already paid for: the clicks, the messages, the waiting and the repeating. In e-commerce almost all of it is spent after checkout, and reducing customer effort means removing that work, not distributing it more politely. Every tactic in the standard advice starts counting effort when the customer opens a chat window, and by then it has already been spent. The conversation isn't where effort is reduced, it's where effort is measured, and the only number that goes to zero is the contact that never had to happen. I used to watch CX teams burn half a day hunting one order across five systems, copy-pasting tracking numbers and apologizing for a problem they hadn't caused, long after the customer had done their own chasing.

Dixon, Freeman and Toman set the idea out in Harvard Business Review in 2010, on a study of more than 75,000 people dealing with service representatives or self-service channels: customers want a simple, quick solution rather than delight.

Effort before checkout, and effort after it

Two different problems wear the same word. Effort before checkout is a design problem: too many fields, a missing payment method, copy that doesn't answer the question. It's real, it's well covered, and it belongs to somebody else's article. Effort after checkout is an operations problem, and it now rides on 16.9% of US retail sales, the e-commerce share in the first quarter of 2026. The parcel is stuck at a depot, the address failed validation after payment settled, the order shipped short, the return never got scanned in. Nothing about your help center changes any of that. The customer's effort here isn't friction in an interface, it's chasing, and chasing only exists because something broke and nobody acted on it. It's not getting better on its own either: the ACSI put online retail satisfaction down 1% to 79 in 2025, with more than two thirds of the companies it measures declining.

One in five orders hits an operational break after checkout. Brands hand those breaks to a helpdesk, and a helpdesk is a system for replying about problems, not for resolving them. Answering the customer and fixing the order are different jobs, and only one of them removes any work. Fixing the order is a different category of work: proactive e-commerce operations, where something is watching the order rather than the inbox.

The standard playbook, and where each tactic stops

Five tactics turn up on every page written about this, and they work. Run them. What none of them does is remove the break that created the contact, so be exact about where each stops.

• Self-service: removes the wait for an answer that's already written down. It leaves everything that needs an action rather than an answer, which is most of what breaks after checkout. Gartner found 14% of customer service issues fully resolved in self-service in 2024, and in 2019 found that only 9% of customers reported solving their issue completely on their own. That's the honest ceiling on the tactic everyone leads with.
• Omnichannel and no repeating: removes the cost of switching channels. It leaves the reason the customer had to switch. Gartner now has more than half of service journeys starting on third-party platforms, which is effort displaced rather than removed.
• First contact resolution: removes the second and third contact. It leaves the first one, which cost the customer the most. Getting ahead of it is what proactive customer service aims at, and what proactive operations actually does, because the contact is prevented upstream of the conversation rather than inside it.
• Arming the agent: removes hold time and escalation. It leaves a person hand-fixing an order that a system broke, at your cost instead of theirs.
• Feedback and surveys: removes nothing at all. It measures, after the fact, and only among the people who still care enough to answer.

Every tactic on that list makes the complaint cheaper. Not one of them makes the complaint unnecessary.

Measure effort as a count, not a score

The Customer Effort Score, like most customer satisfaction metrics, is a survey of effort that has already been spent, collected from the subset of customers who still answer surveys. It's not wrong, it's late. The evidence behind it is also thinner than the advice suggests: a peer-reviewed comparison across 93 firms in 18 industries found top-two-box satisfaction the better predictor of retention, found the best metric varies by industry, and found a combination of metrics beats any single one.

Effort is countable without asking anyone. Contacts per hundred orders. The share of those contacts that are order status rather than product questions. Days between the break and the first action taken on it. The number of times a customer has to touch you before the order is right. All four already sit in systems you run, all four fall when orders stop breaking, and not one of them needs a survey or a response rate to work.

The breaks that generate the most effort

If effort is created by breaks, the work is ordered by the breaks that create the most chasing per occurrence. Four of them account for most of the volume.

• The parcel that stops moving: tracking stops updating, so the customer checks it, checks it again, then writes in. Nobody at the brand was reading the same scan feed, which is why a delivery exception usually arrives as an email rather than as an alert.
• The address that fails after payment: settled, accepted, undeliverable. The customer hears it from the carrier rather than from the brand that took the money, and then has to prove what they ordered and where it should go.
• The order that ships short or split: two parcels, one shipping email. The customer has to work out for themselves whether the rest of it is coming or missing, and the only way to find out is to ask.
• The return that stalls: received, unscanned, unrefunded. NRF puts returns at 19.3% of online sales in 2025, and finds 71% less likely to shop again after a poor returns experience, up from 67%. The RMA step is where it stalls.

Reduce effort at the source

The mechanism has three parts, in order. Detect the break from the signals your systems already emit, the same scan feeds and status webhooks the customer is refreshing, rather than from the customer's email. Decide what should happen by measuring that order against the promise made at checkout, not against a queue position. Then act: reship, refund, re-route, or tell the customer before they ask. The last one is the cheapest and the most often skipped.

In the US, acting before the customer asks isn't only good practice, it's partly the law. Under the FTC's Mail, Internet, or Telephone Order Merchandise Rule a seller must have a reasonable basis to expect it can ship within the window it advertised, or within 30 days where it advertised none. When it cannot, 16 CFR Part 435 requires a delay option notice and either the buyer's consent or a refund. Read that as an operator: the rule already treats the shipping date as a promise, and already says the seller has to speak first. Most brands find out the promise broke when the customer tells them, days after the moment they were supposed to have acted.

What this does not mean

Keeyu isn't a helpdesk, not a chatbot, not a survey tool, not a carrier and not an OMS. We won't make your help center easier to search, we won't write your CES survey, and we don't replace the team that talks to your customers. We remove the breaks they're talking about. The argument has an honest limit too. The effort of choosing, comparing and buying belongs to the customer and isn't yours to take away. The effort of chasing an order that broke on your side is entirely yours.

Reduce the effort you created

If your effort number is bad, the lever is upstream of the conversation you're trying to improve. Keeyu is proactive e-commerce operations: we watch every order against what it was promised, detect the break, decide what should happen and act on it, usually before the customer knows there's anything to chase. Every order is a promise, and Keeyu keeps the promise. Book a Keeyu demo and bring the orders your team hand-fixed last week.

• Dixon, M., Freeman, K. and Toman, N. Stop Trying to Delight Your Customers. Harvard Business Review, July to August 2010. The origin of customer effort as a concept, on a study of more than 75,000 people interacting with service representatives or self-service channels.
• de Haan, E., Verhoef, P. C. and Wiesel, T. The predictive ability of different customer feedback metrics for retention. International Journal of Research in Marketing 32(2), 2015. Ninety-three firms across 18 industries: top-two-box satisfaction predicts retention best, the best metric varies by industry, and a combination beats any single one.
• Gartner. Only 14% of issues fully resolved in self-service. Press release, 19 August 2024.
• Gartner. Only 9% of customers solve issues completely via self-service. Press release, 25 September 2019.
• Gartner. More than half of service journeys now start on third-party platforms. Press release, 23 July 2025. Effort displaced across channels rather than removed.
• Federal Trade Commission. Business Guide to the Mail Order Rule. The reasonable basis to ship in the advertised window, 30 days where none is advertised, and the delay option notice.
• Federal Trade Commission. Mail, Internet, or Telephone Order Merchandise Rule, 16 CFR Part 435. The rule itself, including consent to a delay or a refund.
• National Retail Federation and Happy Returns. 2025 Retail Returns Landscape. An estimated 19.3% of online sales returned in 2025.
• National Retail Federation. Consumers expected to return $850 billion in merchandise in 2025. Press release, 15 October 2025. Seventy-one percent of consumers are less likely to shop with a retailer again after a poor returns experience, up from 67%.
• American Customer Satisfaction Index. Retail and Consumer Shipping Study 2025. Online retail satisfaction down 1% to a score of 79, with more than two thirds of measured companies declining.
• US Census Bureau. Quarterly Retail E-Commerce Sales. E-commerce at 16.9% of total US retail sales in the first quarter of 2026, the share of trade that runs on a post-checkout promise.