Lubinpla

How to run a win-loss review after a customer switches

Summary

A technical review of a lost account takes 5 steps: open inquiries, repeat problems, promises against delivery, rep handovers, and what the records cannot confirm. Work from the account's work logs, inquiry records, projects and sent documents for the last 12 to 24 months (illustrative window), and label every suspected reason confirmed by a record or unconfirmed.

  • ISO 10002:2018, edition 3, gives guidelines for complaints handling that include analysing and evaluating complaints to improve products and services; ISO confirmed it as current in 2023 (ISO, 2018).

  • ISO 10004:2018, edition 2, gives guidelines for monitoring and measuring the satisfaction of customers external to the organization (ISO, 2018).

  • ISO 9001:2026, the sixth edition, was published on 16 September 2026 and includes clause 9.1.2, Customer satisfaction (ISO, 2026).

  • A 2014 Westfair column on lost accounts recommends a debrief by a team from sales, marketing, operations and finance (Westfair, 2014).

This review covers the technical side of a lost industrial chemical account: inquiries, site problems, commitments and the people who served it. Contract terms, pricing decisions and any dispute with the customer are outside it. So is contacting the customer, which your sales manager decides.

Which technical records should you pull after a customer switches suppliers?

After a customer switches suppliers, pull every technical record for the account from one fixed review window. The set covers work logs and visit reports, technical inquiries and complaints, projects and trials, sent documents, and the list of reps who served the account. Fix the window first, for example the 12 to 24 months before the switch (illustrative), so that every suspected reason is tested against the same period.

The CRM holds only part of the story. Technical history often sits in sent mail with attachments, shared folders, lab reports and the notes of a rep who has since moved on. ISO 15489-1:2016, confirmed as current in 2021, applies to the creation, capture and management of records regardless of structure or form (ISO, 2016). Under that scope, an email thread with an agreed dosing rate is a record too.

Build the inventory as one table per record type. Give each row a date, the customer contact, the product or system it concerns, and a pointer to the original. Internal opinions do not belong in the inventory.

Write down the reasons people suspect for the loss on a separate list before the review starts. The records then confirm or reject each reason, instead of being read to fit a story everyone has already told.

How do you run the win-loss review of the account history step by step?

The review has five steps, and each one produces a list that feeds the check table at the end.

  1. Open inquiries. List every technical inquiry and complaint in the window with the date received, date answered and date closed. Mark any that got no answer, or got one later than the time you committed to.

  2. Repeat problems. Group work log entries and complaints by symptom and system, and mark any symptom recorded two or more times in the window (illustrative threshold).

  3. Promises against delivery. List each promised action from visit reports, meeting notes and sent mail, then find the record that shows it was delivered, and when.

  4. Rep handovers. Draw a timeline of who served the account, with handover dates, and place the first three lists on it.

  5. What records cannot confirm. Go back to the suspected reasons written before the review and mark each one confirmed, contradicted or unconfirmed.

Westfair's 2014 column on lost accounts recommends a debrief by a team from sales, marketing, operations and finance, and asks who was looking after the account and what they knew, when (Westfair, 2014). The five steps hand that debrief a technical record in place of recollections. Step 5 keeps any explanation that nothing supports visible as an open question.

How do you find repeat problems and unresolved inquiries in the account history?

Sort every inquiry and work log entry by system and symptom, then read each group in date order. Look for two things: a problem that shows up in several entries with a different fix each time, and an inquiry with no recorded close.

For each inquiry, record four dates where they exist: received, first response, final answer and closed. If the customer never confirmed an answered inquiry as solved, it counts as open. ISO 10002:2018 gives guidelines that include analysing and evaluating complaints in order to improve the quality of products and services (ISO, 2018). Treat technical inquiries the same way, since they often carry complaints that nobody logged as complaints.

For repeat problems, work from the customer's words and the measured values, not from the rep's diagnosis. Suppose three visits record foaming in the same sump and each closes with a different adjustment (illustrative): the problem was handled three times and never solved. For each group, note whether anyone recorded a root cause, whether a corrective action was agreed, and whether a later reading showed it worked. ISO 9001, in the 2015 edition and the 2026 edition alike, has clause 10.2, Nonconformity and corrective action (ISO, 2026; Advisera, 2026). So a customer certified to ISO 9001 may have tracked the same repeat problem on its own side.

How do you check what was promised to the customer against what was delivered?

List every commitment the account team made to the lost customer, then find the record, with its date, that shows each one was delivered. For the review, a commitment with no delivery record counts as missed, even if someone remembers doing it.

Commitments turn up in visit reports, meeting notes, quotations, trial plans and sent mail. Typical ones are a sample, a trial, a lab analysis, a report, a revised technical data sheet or a follow-up visit by a set date (illustrative examples). Write each one as a row: what was promised, to whom, by when, and where the promise was made.

Then look for the delivery: the sent mail with the report attached, the work log of the visit, the shipment of the sample or the project record of the trial. Record the delivery date and how late it came against the promised date.

Some deliveries happened but did not give the customer what they expected, such as a trial that failed its success criteria. Note those separately, because a delivered trial that failed and a trial that never ran point to different reasons for the switch.

Limit this list to the lost account. Tracking commitments across all accounts belongs to day-to-day account management, not to the loss review.

Did a change of technical rep play a part in the customer switching?

A change of technical rep played a part in the switch if the records show gaps that begin at the handover date: inquiries answered later, the previous rep's promises left undelivered, repeat problems treated as new, or fewer visits. The rep timeline from step 4 of the review is where these gaps show.

Place every handover on the timeline and compare the period before it with the period after on the same measures: response time to inquiries, number of visits, open commitments carried over, and problems that reappeared. Then check whether the incoming rep received the account history in any form, whether a handover document, a joint visit, or access to the previous rep's files and sent mail.

Westfair's 2014 column asks of a lost account who was looking after it and what they knew, when (Westfair, 2014). The timeline answers that question with dates. Timing alone settles nothing, though: a switch that follows a handover is not proof that the handover caused it.

Record the handover as confirmed only when a specific gap, such as an unanswered inquiry or an undelivered promise made before the handover, sits between the handover date and the switch. Otherwise, label the rep change a possible factor and leave it unconfirmed.

Which reasons for the customer switching suppliers can the records confirm?

The records can confirm the reasons for a switch that leave a trace, such as an unanswered inquiry, a repeat problem or a missed promise. They cannot confirm price, relationship or a competitor's offer unless the customer said so in writing. The check table below sorts each suspected reason into one the technical history supports or one that stays in the internal report as an unconfirmed opinion.

Possible reason for the switch

Record that confirms it

Confirmed

Not confirmed and what to do

Unresolved inquiry

Inquiry with no answer, or answer with no close

Inquiry log shows it open at the switch date

Keep as possible; ask the customer in the exit conversation

Repeat problem

Same symptom logged two or more times with no root cause (illustrative threshold)

Work logs show the pattern

Check whether the customer logged it elsewhere

Missed promise

Promise with no delivery record, or a late one

Promise and missing delivery both on record

Ask the rep who made it; mark unconfirmed if no source

Rep handover gap

Gap that starts at a handover date

Timeline shows the gap after the handover

Label as possible factor only

Price

Written customer statement or tender result

Customer wrote that price decided it

Leave unconfirmed; do not infer from silence

The price row matters most, because price is the reason most often assumed and least often recorded.

Write each reason into the internal report with its label and the record behind it, so a reader can open the inquiry, work log or sent mail and judge it. A reason the records contradict, such as a repeat problem that the work logs show was solved months before the switch, stays in the report with that finding. The same assumption will come up at the next lost account.

Regular reviews can surface these signals before a customer leaves; preparing a quarterly business review for an industrial customer covers how. In Lubinpla, the account technical support status task collects the customer's work logs, projects and inquiry records for the period you set into an internal report for you to review.

Frequently asked questions

Should you ask the customer why they switched?

Yes, if your sales manager agrees, and after the records review, so your questions are about specific items rather than general. ISO 10004:2018 gives guidelines for monitoring and measuring the satisfaction of customers external to the organization (ISO, 2018). A short exit conversation is one way to test the unconfirmed reasons.

Who should take part in the lost account review?

Include the technical rep or reps who served the account and their manager, plus someone from sales, operations and finance who was not involved. That way the records are read by people with no stake in any particular reason. The 2014 Westfair column recommends a team from sales, marketing, operations and finance for the same purpose (Westfair, 2014).

What should the internal report on a lost account contain?

The report holds the review window, the record inventory, the five lists from the review, and the check table with each suspected reason marked confirmed, contradicted or unconfirmed. Close it with the corrective actions for other accounts at risk of the same gaps, each with an owner and a date.

Related posts