Feedback That Works: Core Work

Why this matters

Founders almost always know when someone on the team is underperforming. They feel it in their gut during the morning standup. They see it in client complaints. They notice it in the rework piling up on the shared drive.

The problem is not awareness. The problem is the gap between knowing and saying something. That gap costs you money every week it stays open. A team member delivering at 60% for three months is a feedback failure, not a people failure.

In the ARCAS diagnosis, this chapter maps directly to the Behaviour and Skills audits. When those scores are low, the cause is almost never that you hired the wrong people. It is that nobody told them clearly what good looks like, what they are doing that falls short, and what to change. That is people leakage in the Five Levels model, and it gets worse the longer it sits.

---

A founder you might recognise

Last year, the founder of a 35 person commercial signage and printing firm in Al Quoz was managing a team that mixed Indian, Pakistani, and Lebanese technicians, plus a South African operations manager and two Emirati client relations officers.

He knew the senior technician had been cutting corners on the preventive maintenance reports. Clients had noticed. Two had sent emails asking why the quarterly reports looked thinner than last year. He had not said anything directly. Instead, he told the operations manager to "keep an eye on quality," which changed nothing.

He also knew that the senior client relations officer had been taking on work outside her role because she did not trust the handoff from the technical team. She was now overloaded, and her own client response times had slipped. He had not acknowledged what she was doing, because he did not want to make the technicians feel blamed.

Three months of this, and he estimated the cost at roughly AED 40,000 (USD 10,890) per month in rework, unbilled hours, and one client who had started getting quotes from competitors on the side.

He did not have a people problem. He had a feedback vacuum.

---

The SBI model for small teams

SBI stands for Situation, Behaviour, Impact. It is the simplest feedback structure that consistently changes behaviour in mixed-nationality teams. Here is why it works: it removes opinion and replaces it with observable fact.

Situation: When and where did this happen? Be specific. Not "lately" or "in general."

Behaviour: What did the person actually do or not do? Describe the action, not the character.

Impact: What was the result? On the client, the team, the numbers, or you.

Here is what this sounds like in practice:

"Last Tuesday, on the Al Reem Tower quarterly report, the preventive maintenance section had three line items with no completion dates. The client emailed asking if the work was actually done. That puts the renewal conversation at risk."

Compare that to what most founders say: "Your reports need to be better." The second version gives the team member nothing to act on. The first version gives him the exact gap and why it matters.

Adapting SBI for cultural context

In a UAE service business, you are giving feedback across cultures with very different norms around hierarchy, directness, and face.

Indian and Pakistani team members often come from workplaces where feedback from a senior is received formally. Deliver SBI calmly and privately. Give the person a moment to respond. Do not rush to fill the silence. Many will need a few seconds to process before they speak.

Lebanese and other Arab team members may respond better when you start with a brief acknowledgement of their overall contribution before moving to the SBI point. This is one honest sentence, not a compliment sandwich: "Your client relationships are strong, and I need to talk about something specific."

British and South African team members tend to expect direct feedback but can interpret vague language as more negative than intended. Be precise about what needs to change. "I need you to do X differently" carries more weight than "I'd like you to think about how you approach X."

Emirati team members in client-facing roles often respond well to feedback framed around the company's reputation and the national context. Connect the behaviour to what it means for the brand alongside the immediate task.

The constant across every culture: privacy, specificity, and respect. Never give corrective feedback in front of others. Never make it about personality. Always make it about the work.

---

The feedback loop that changes behaviour

Giving feedback once does not change anything. A single SBI conversation is like sending a single invoice and hoping the money arrives. You need a loop.

Step 1: Deliver the SBI. State the situation, behaviour, and impact. Be calm and specific.

Step 2: Ask for their view. "What is your take on this?" You may learn something. The technician might tell you the maintenance data is not being entered by his team because the Excel tracker keeps corrupting. That changes the problem entirely.

Step 3: Agree on the change. "What will you do differently starting this week?" Get them to commit to the change themselves rather than have you prescribe it.

Step 4: Set a check-in. "Let us look at this together next Thursday." This is the step everyone skips. Without it, the conversation evaporates.

Step 5: Follow through. At the check-in, either acknowledge the improvement or repeat the SBI with the new data. This is where behaviour actually shifts.

---

Weekly 1:1 structure

If you manage more than five people directly, you need a weekly 1:1 with each of them. Fifteen minutes is enough. Thirty minutes is better if you manage managers. Here is a structure that works.

First 5 minutes: Their update. Let them talk. What happened this week, what is stuck, what do they need from you. Do not interrupt.

Next 5 minutes: Your feedback. One SBI if there is corrective feedback to give. One specific acknowledgement of something they did well. Not a generic "good job" but a real observation: "The way you handled the Al Barsha client complaint on Wednesday, calling them back within the hour, kept that relationship intact."

Last 5 minutes: Forward look. What matters most next week. Any decisions you need to make together. Any blockers you need to remove.

Output: After the 1:1, send a three-line WhatsApp message summarising what you agreed. This takes 30 seconds and creates accountability without a system. The message is the system.

What kills 1:1s

Cancelling them. The moment you cancel two in a row, your team reads it as "this is not important." Protect these the way you protect client meetings. If you must reschedule, say when.

---

Difficult conversation templates

Some conversations go beyond weekly SBI. The first handoff tool gives you a ready-made conversation script for your first delegation, and the templates below cover the harder situations.

Template 1: Repeated performance gap

"[Name], we talked about [specific issue] on [date]. I said I would check in, and here is what I see now: [current observation]. The impact is [specific cost or risk]. I need to see [specific change] by [date]. If that does not happen, we need to have a different conversation about your role here. What support do you need to make this happen?"

Template 2: Attitude or behaviour issue

"[Name], I want to talk about something I have observed in the last [timeframe]. In [specific situation], you [specific behaviour], for example, [concrete instance]. The effect on the team is [specific impact]. I value what you bring, and I need this to change. What is going on from your side?"

Template 3: Acknowledging someone carrying too much

"[Name], I have noticed you are picking up work that is not in your role, specifically [examples]. I want you to know I see it and I appreciate it. It is also not sustainable, and it is starting to affect [specific impact on their core work]. Here is what I am going to change on my end: [specific action]. I need you to stop taking on [specific tasks] starting next week."

That third template is the one this founder needs for the senior client relations officer. It is also the one most founders never deliver because it feels like punishing someone for being good.

---

Common mistakes

Stacking feedback. Saving three months of observations for one overwhelming conversation. Give feedback within 48 hours or not at all.

The compliment sandwich. Everyone knows the pattern. It makes the compliments feel fake and the criticism feel hidden. Just say the thing.

Feedback by WhatsApp. Corrective feedback is face-to-face or video call. Acknowledgement can be written. Never reverse these.

Making it personal. "You are careless" is not feedback. "The report was missing three data points" is feedback.

Avoiding feedback with high performers. Your best people are the ones who most need to hear what they are doing well, specifically. They are also the ones most likely to leave if they feel invisible.

Applying one style to every culture. What feels respectful to a British operations manager may feel cold to an Indian team lead. Learn how each person receives feedback best by asking them directly.

---

When to move on

Move to the next chapter when you can answer yes to these:

• You have a weekly 1:1 with every direct report, and it has run for at least three consecutive weeks.
• You have delivered at least one SBI feedback point to every direct report in the last 14 days.
• You have had one difficult conversation you were previously avoiding.
• Your team knows what "good" looks like in their role because you have told them directly.

---

Where to focus by team size

• 10 to 19 people: Start weekly 1:1s with your direct reports. Even 15 minutes changes the dynamic.
• 20 to 34 people: Your team leads need to run their own 1:1s. Train them on the SBI model.
• 35 to 50 people: Feedback should flow in three directions: founder to leads, leads to teams, teams to leads. If it only flows down, you have a culture problem.

---

Working prompts

People prompts

• Who on your team has not received specific feedback from you in the last 30 days?
• Which underperformance have you noticed but not addressed? What is the monthly cost of that gap?
• Who is carrying weight that is not theirs, and what happens when they stop?

System prompts

• Do you have a recurring 1:1 schedule, or does it happen when you remember?
• Is there a written standard for what "good" looks like in each role, or does the team guess?
• When you give feedback, is there a follow-up step, or does the conversation end and dissolve?

AI prompts

• Could a shared Excel tracker surface performance patterns before they become problems?
• Where could Zoho or your CRM flag missed SLAs so the data comes to you instead of you hunting for it?
• What repetitive quality checks could be automated so your feedback focuses on judgment instead of data entry?

---

Founder exercise

Set aside 45 to 60 minutes. Do this before your next round of 1:1s.

Part A: The feedback audit (15 minutes)

List every person who reports to you directly. Next to each name, write:

1. The last time you gave them specific corrective feedback (date and topic).
2. The last time you gave them specific positive feedback (date and topic).
3. One thing you have noticed but not said.

If you cannot remember dates, that is your answer.

Part B: Three SBI scripts (15 minutes)

Pick three items from the "not said" column in Part A. Write a full SBI for each one:

• Situation: when and where
• Behaviour: what they did or did not do
• Impact: what it cost the business, the client, or the team

Write these out. Do not leave them as bullet points in your head. The act of writing forces precision.

Part C: The 1:1 calendar (10 minutes)

Open your calendar. Block a recurring 15-minute slot with every direct report for the next four weeks. Send the invite now, before you close this page. Put the first SBI from Part B in your notes for the relevant 1:1.

---

ARCAS lens

Feedback is the bridge between having good people and getting good performance from them. Without it, even a well-designed role architecture and a strong hiring process decay over time. People drift. Standards blur. Small problems become expensive ones.

People build the feedback habit. Systems make it consistent, through 1:1 schedules, written follow-ups, and clear standards. AI can surface the data that triggers feedback, flagging missed SLAs or quality drops before you notice them. But the conversation itself stays human. That is the one part you do not automate.

---

Start now: Quick self-assessment

Rate each statement from 1 (never true) to 5 (always true):

Score 24 or above: Your feedback system is working. Move to the next chapter. Score 15 to 23: The structure is partially there. Work through the founder exercise to close the gaps. Score below 15: Feedback avoidance is likely costing you more than you realise. Do the full working session before moving on.