ChatGPTA team supervisor says, “Be more confident with customers.”
The team member nods. Next shift, nothing looks different.
That feedback wasn’t wrong. It was just too vague to practise.
If you work in L&D, People, or operations, you’ve probably seen this pattern a hundred times: managers give well-meaning labels (“be more confident”, “be more customer-focused”, “be more proactive”) and hope the label turns into behaviour.
It rarely does.
What changes behaviour is not the label — it’s a shared definition of what good looks like in one specific moment, plus a short, repeatable way of coaching it.
This post gives you one prompt supervisors can reuse on shift, without sounding awkward or making it personal.
Why coaching labels does not work
Labels are shortcuts. They’re also slippery.
“Be more confident” could mean:
- speak louder
- make eye contact
- use the opening line
- take control of the conversation
- stop asking for permission
- show the next step
When a supervisor gives a label, the team member has to guess which version they meant.
Most people default to what feels safe. On a busy shift, that usually looks like… doing the same thing again.
If you want consistency across sites, teams, and managers, labels are the wrong unit of coaching.
Coach the moment you observed. Agree the next-time behaviour.
That’s it.
The simple fix: SBI plus “next time”
A lot of managers are familiar with the SBI feedback model:
- Situation: when it happened
- Behaviour: what you saw and heard
- Impact: what it caused
It works because it keeps feedback specific and factual.
On the frontline, there’s one more ingredient that turns feedback into performance:
- Next time: the exact words or behaviour to try in the next relevant moment
Managers do not avoid coaching because they don’t care.
They avoid coaching because:
- they’re worried about saying the wrong thing
- they don’t want it to feel personal
- they can’t find the words quickly
- they do not trust it will lead to change
So we simplify the model into a one-line prompt they can remember.
Give supervisors one script: see, say, next
Here’s the script.
- (See) What I saw was…
- (Say) What good looks like is…
- (Next) Next time, say this…
Why this works on shift:
- it stays factual (no judgement)
- it focuses on one moment (no overwhelm)
- it gives the words (no guesswork)
Example: refunds
What I saw was a 10-second pause when the customer asked about refunds.
What good looks like is starting with “I can help with that” within a few seconds, then explaining the next step.
Next time, try: “I can help with that. Refunds are processed at the till. I’ll show you where.”
Two quick rules that keep this effective:
- One behaviour per conversation. If you coach three things, you coach nothing.
- Use a real moment. Hypotheticals do not stick like lived context.
Step 1: pick one common moment and define “good” as observable behaviours
You are not trying to teach supervisors how to coach from scratch.
You are giving them one shared way of doing it — and a shortlist of moments to coach.
Start with moments that happen every day. Adoption is easier when it’s frequent.
Pick one moment. Then write “good looks like” as what you can actually see and hear.
Example moment: customer asks about refunds
Good looks like:
- acknowledges within a few seconds
- uses the standard opening line
- gives the next step
Notice what’s missing:
- “shows confidence”
- “seems calm”
- “has the right attitude”
Those might be outcomes. They’re not coachable behaviours.
A simple way to choose the moment
If you’re stuck, ask:
- Which moment happens daily?
- Which moment drives customer experience or risk?
- Which moment is most inconsistent across sites?
If you can’t answer those quickly, you’re probably trying to coach too broadly.
Go smaller.
Step 2: coach the moment, then agree the next-time behaviour
This is where most frontline coaching falls down.
Managers notice the gap… then jump straight to a label.
The script stops that.
A quick checklist for supervisors
Before they speak, ask them to sanity-check their coaching with three questions:
- Can I point to a specific moment?
- Can I describe what I saw and heard, without guessing intent?
- Can I name one next-time behaviour, ideally with the words to use?
If the answer is “no” to any of these, the coaching will probably land as opinion.
More examples supervisors can copy
Moment: customer is waiting at the counter
- See: “What I saw was you finishing the stock task while the customer waited at the counter.”
- Say: “What good looks like is acknowledging within a few seconds, even if you’re mid-task.”
- Next: “Next time, try: ‘I’ll be with you in two seconds’ — then finish the action and step over.”
Moment: a complaint starts to escalate
- See: “What I saw was the customer raising their voice, and we stayed in problem-solving mode without acknowledging how they felt.”
- Say: “What good looks like is naming the feeling before the solution.”
- Next: “Next time, try: ‘I can see this is frustrating — let’s sort it now.’”
Moment: customer asks a question we don’t know
- See: “What I saw was we guessed the answer when the customer asked about delivery times.”
- Say: “What good looks like is being confident about the next step, not confident about the unknown.”
- Next: “Next time, try: ‘Let me check that so I don’t guess. It’ll take 20 seconds.’”
Step 3: build the habit with a five-minute huddle drill
Most managers know what to do.
They freeze on the phrasing.
So make it normal to say the words out loud.
Here’s a simple huddle drill that takes five minutes:
- Ask supervisors to bring one real moment from yesterday.
- In pairs, they run the script once each.
- Swap feedback: did it stay specific? Was the “next time” behaviour clear?
This works because it turns coaching into a habit, not a heroic moment.
And it reduces the emotional load. If everyone practises the same script, it stops feeling like a confrontation and starts feeling like a standard.
Here’s what evidence-based coaching looks like in Ocasta’s Observation & Coaching Hub — move from opinion to structured observations with factual feedback:
Why it matters: evidence beats opinion
Opinion-led coaching varies by site.
One manager’s “confident” is another manager’s “pushy”. That inconsistency is frustrating for teams and impossible for head office to fix.
Evidence-led coaching is:
- easier to accept (it’s about what happened)
- easier to repeat (it’s a simple script)
- easier to check (you can re-observe the same moment)
It also helps new starters because “good” becomes visible.
When “good looks like” is written as behaviours, you remove a whole layer of guessing.
Close the loop this week
If you want this to stick, close the loop quickly.
Here’s the simplest way:
- Pick one moment.
- Coach it using the script.
- Re-observe the same moment within seven days.
In Ocasta, you can create actions during observations so follow-up doesn’t get lost. Clear ownership, due dates, and priorities ensure every coaching conversation leads to measurable progress:
A simple, no-fuss measure
Out of 10 relevant moments this week, how many matched your “good looks like” definition?
It’s not perfect.
It is good enough to:
- show improvement
- highlight inconsistency
- create a coaching rhythm
Ocasta’s real-time analytics show strengths alongside gaps. See which behaviours are improving, which moments need attention, and where coaching is having the biggest impact:
And ask supervisors one feedback question:
- “Did the script make coaching quicker?”
If it did, you’ve removed one of the biggest barriers to coaching happening at all.
Common pitfalls (and how to avoid them)
Pitfall 1: trying to fix everything
If a supervisor has a list, they will dump it all on the frontline employee, and nothing gets done.
Keep it to one behaviour. The goal is repeatability.
Pitfall 2: coaching personality instead of behaviour
Avoid:
- “You’re not confident.”
- “You’re too quiet.”
- “You’re not customer-focused.”
Replace with:
- “There was a 10-second pause after the customer asked.”
- “We didn’t use the opening line.”
- “We didn’t explain the next step.”
Pitfall 3: not giving the words
On the frontline, words matter. People need phrases they can try in the next moment.
If the “next time” line is missing, the coaching becomes a nice conversation that fades by tomorrow.
Pitfall 4: no follow-up
If nothing is re-observed, coaching becomes theatre.
Close the loop. It’s how teams learn that coaching leads somewhere.
A mini template you can drop into your supervisor pack
Pick one moment: __________________________________________
Good looks like (observable behaviours):
- 1:
- 2:
- 3:
See, say, next script:
- What I saw was… _________________________________________
- What good looks like is… _________________________________
- Next time, try… __________________________________________
Follow-up plan: Re-observe within 7 days.
Quick measure: Out of 10 relevant moments, how many matched?
How Ocasta supports evidence-based observations and coaching
The hard part is not knowing that coaching should be specific.
The hard part is making it consistent across managers, sites, and regions — and turning coaching into a loop you can see.
Ocasta replaces opinion-led coaching with evidence-led coaching, so supervisors have something concrete to work from on shift.
Ocasta makes capture fast with one-tap scoring and short notes, even on busy shifts. Time-stamped evidence and structured forms keep coaching factual:
Teams use Ocasta to:
- Evidence, not opinion: time-stamped notes and optional photos keep feedback factual.
- Fast capture: one-tap scoring and short notes make observations quick, even on busy weekends.
- Clear standards: ideal-journey templates define what good looks like for stores, showrooms, and front-of-house teams.
- Real-time views: spot patterns by person, site, or region, with strengths alongside gaps.
- The learning loop: turn themes into targeted content using the Knowledge and learning hub, so learning matches the gaps you’ve found.
- Action you can track: create actions during the observation so follow-up does not get lost.
The aim is simple: less guessing, more knowing. That’s how observations turn into measurable improvement, not just good intentions.
A good week-one plan
If you want to introduce this without a big change programme, here’s a lightweight plan that works.
- Choose one moment for the week (refunds, queues, complaints — pick what matters most).
- Share the “good looks like” behaviours.
- Ask supervisors to use the script once per shift.
- Run the five-minute huddle drill twice this week.
- Re-observe within seven days and count 10 moments.
Keep it simple.
The point is not perfection.
The point is creating a shared coaching language that shows up on shift.
If you want to see how Ocasta supports observation-led coaching across a region, you can learn more now:
