Borrow Ten Minutes from Debate: Let the Customer Choose the Test

Workplace Growth Daily · 2026-09-22

22 September 2026

Illustrative scenario: A customer writes, “I could not tell which option suited me.” Support calls it a wording problem. Sales suspects pricing. Product proposes a redesign. By the time everyone has explained why everyone else is wrong, the customer could have learned pottery.

The responsible response is neither a sweeping project nor the loudest opinion. Listen for the customer’s actual difficulty, identify the smallest plausible change and test it safely. Because none of today’s supplied evidence was published in the last seven days, this is a research-backed practical briefing, not breaking news.

The findings below provide context. The exercises and experiment are LUMARA’s practical suggestions, not claims made by the sources.

What is changing

1. Clarity is improving, but voice still matters. On 21 July 2026, Gallup reported that 49% of surveyed US employees strongly agreed they knew what was expected at work, up two percentage points from
2. Yet measures of whether opinions count and connection to organisational purpose declined slightly. For a real team, clear tasks are necessary but insufficient: people also need a credible route for influencing how work is done. (Gallup, 21 July 2026)
3. Managers themselves may need support. Gallup’s global report, published 12 June 2026, said manager engagement fell from 27% in 2024 to 22% in 2025, while 79% of managers were engaged in the best-practice organisations it studied. These figures show an association, not a guarantee. The practical implication is to give managers time, training and manageable spans of work rather than expecting them to absorb every customer issue unaided. (Gallup, 12 June 2026)
4. Hiring is not an easy escape from capability gaps. The UK-focused CIPD survey published 16 August 2026 found that 62% of more than 2,000 employers planned to recruit in the next three months; in the private sector, the figure was 57%, the lowest since that measure began in 2016 outside the pandemic. Teams should therefore make experiments teach existing colleagues something useful—not assume more people will arrive to repair a poorly understood process. (CIPD, 16 August 2026)

Build a skill in 10 minutes

This is the ability to convert what a customer said into a neutral problem statement and a testable idea—without pretending one comment represents everyone.

Ten-minute exercise

Choose one recent, anonymised customer comment.

1. Minutes 0–2: Copy the customer’s exact words. Remove names and sensitive details.
2. Minutes 2–4: Write: “The customer was trying to ___ but experienced ___.” Mark assumptions with a question mark.
3. Minutes 4–6: List two possible explanations. Do not select a favourite yet.
4. Minutes 6–8: Create the smallest reversible test of one explanation.
5. Minutes 8–10: Define one result measure and one safeguard.

Visible success criteria: The page contains the original signal, a neutral problem statement, at least two explanations, one reversible test, one measure, one safeguard and no claim that a single customer speaks for the whole market.

Staff and management: respect in both directions

Managers should treat staff with respect, apply clear and consistent expectations, recognise useful contributions, listen before deciding and keep workloads manageable. Invite disagreement without punishment. Never dismiss bullying as “passion” or ask someone to tolerate it; follow the organisation’s reporting and safeguarding process.

Manager—say this: “Thank you for challenging the idea. Please show me the customer evidence and the workload risk you see. We will agree what success means, and if this test is added, we will decide what pauses. I will credit the people who contributed.”

Staff should communicate early, distinguish evidence from interpretation and explain the effect on customers or delivery. Raise concerns through the manager, HR, ethics or safeguarding route appropriate to the issue. Disagreement can be direct and respectful; personal attacks cannot.

Staff member—say this: “I support learning from this signal, but I disagree with launching the full change. We have one comment, not a pattern, and the current deadline leaves no review time. Could we test the wording with a small group and reassess in seven days?”

Work better as a team

Run this 15-minute exercise with four roles: facilitator keeps time; customer-voice reader presents exact, anonymised evidence; experiment designer proposes a minimal test; risk checker examines workload, fairness and unintended effects.

1. 0–3 minutes: The reader shares one customer signal and its context. The team may ask factual questions only.
2. 3–6 minutes: Everyone silently writes one possible cause. The facilitator groups similar ideas without forcing agreement.
3. 6–10 minutes: The designer proposes one reversible change that can run for seven days with existing tools.
4. 10–13 minutes: The risk checker asks: Who might be excluded? What work must pause? Could quality, privacy or accessibility worsen?
5. 13–15 minutes: The team selects or rejects the test and records the reason.

Shared output: One completed card containing customer evidence, hypothesis, audience, change, owner, baseline, target direction, review date, stopping rule and workload safeguard. “We need more evidence” is a valid output.

Turn teamwork into measurable growth

LUMARA practical experiment: Where customers commonly hesitate—such as a form, quotation or service-choice page—test one clearer sentence with a small eligible customer group. Keep the current version available so the change is easy to reverse. Do not add an AI tool unless it solves a defined need.

Owner: Customer Experience Lead.

Baseline: Before changing anything, record the previous seven days’ completion rate and the number of contacts asking for clarification at that step. Note volume so raw counts are not misleading.

Test: For seven days, show the revised sentence to a limited group. Keep every other practical element as stable as possible. Log completion and clarification contacts separately. Invite a short optional comment: “What, if anything, was unclear?”

Review: On day seven, compare rates and read the comments. Look for direction and useful learning, not proof of universal impact. Decide to retain, revise, expand cautiously or stop.

Balancing measures: Track staff minutes spent administering the test, unresolved customer issues and any accessibility or quality complaints. Set a time ceiling in advance. If workload or harm exceeds it, pause the experiment. A positive result may justify another test; it does not guarantee retention or revenue.

Try it today

• Select one recent, anonymised customer signal.
• Separate the customer’s words from your assumptions.
• Write two plausible explanations.
• Define one reversible seven-day test.
• Name an owner, baseline and review date.
• Agree what work pauses to create capacity.
• Add a workload and customer-harm stopping rule.
• Thank and credit the colleague who surfaced the signal.

Discussion question: What is the smallest change that would help us learn whether we understood the customer correctly—without creating hidden work for the team?