The call lasted eleven minutes.

I remember the exact number because I checked my phone afterwards, trying to work out how a 14 month programme could be ended in less time than it took me to make my morning coffee.

The client's operations director was polite. Professional. He thanked me for our work, said the delivery had been strong and acknowledged the team had hit every milestone we'd committed to. Then he said they were going in a different direction.

I asked what we could have done differently. He paused - the kind of pause that tells you the person already has the answer but isn't sure you're ready to hear it and then he said: "Aman, we never really felt like you were on our side."

I had 47 status reports. 14 months of green dashboards. A signed acceptance certificate from their own technical team.

I had delivered the project. I had not delivered the relationship. And in enterprise technology, those are not the same thing.

Today's menu

  • The problem: How can a programme succeed on every visible metric and still destroy the client relationship

  • What it cost: The six signals that were present the entire time, all of them visible in hindsight, none of them on my RAID log

  • The fix: The three things I now track on every engagement that have nothing to do with delivery and everything to do with keeping the client

THE PROBLEM

⚠️ You have been measuring the wrong things

Let me tell you what the programme looked like from the inside.

We had a 22-person team split across three locations. The scope was a £1.1M platform migration for a financial services client - complex, regulated, politically sensitive. We ran 28 sprints over 14 months. We closed everyone on time or early. Defect rate at go-live: 0.3%, against an industry average of around 4%. The client's own QA team signed off on every release. We finished four days before the contractual deadline.

By any objective delivery measure, this was a success. The project management certification bodies would have used it as a case study.

The problem was not what we built. The problem was how it felt to be our client for 14 months.

After the call ended, I did something I had never done before. I went back through 14 months of client communication, every email thread, every meeting note, every informal message and I read it as if I were the client.

I found six things that should have been red flags. None of them were on my RAID log. All of them were visible in real time. I had looked at all of them and seen nothing.

The 6 signals I missed

01 — The decision latency pattern
From month 8 onwards, the client's responses to our questions started taking longer. Not a lot a day, sometimes two. We noted it once in a retrospective and moved on. In hindsight, they had started routing our questions through a new internal stakeholder who was not in our governance structure. Someone was already building a case


02 — The meeting attendance shift
In month 9, the operations director stopped attending sprint reviews. He sent a delegate. Then the delegate started sending a delegate. We took it as a sign that things were running smoothly and they trusted us. The actual signal: senior client attention had moved elsewhere, and not because we had earned their confidence, because they had mentally deprioritised the programme


03 — The language change
Client emails in months 1–6 said 'our platform' and 'we'. By month 10, they were saying 'the platform' and 'the project'. A small shift in pronoun. A significant shift in ownership. They had emotionally detached from the thing we were building for them



04 — The celebration gap
We hit a major milestone in month 10, the most technically complex phase of the migration, delivered it a week early. We sent a status report. The client replied with one sentence: 'Noted, thank you.' No call. No acknowledgement of their own team. A client who is brought in celebrates with you. A client who has mentally left thanks you for your service


05 — The scope silence
From month 11, the client stopped requesting changes. Not because the product was perfect. Because they had stopped investing in it. When a client goes quiet on scope, it usually means one of 2 things: they're completely satisfied, or they've decided not to throw good money after bad. Given the other signals, I should have known which one this was

06 — The reference deflection
In month 12, I asked the client whether they'd be willing to serve as a reference for a prospect we were courting. They said they needed to check internally. They never came back to me. I followed up once and got a non-committal reply about timing. A client who believes in what you've built will take that call with your prospect. The hesitation told me everything, I just didn't let myself hear it internally. They never came back to me. I followed up once and got a non-committal reply about timing. A client who believes in what you've built will take that call with your prospect. The hesitation told me everything; I just didn't let myself hear it



🤔 QUIZ

Q: You are running a programme. Delivery is on track. The client has never formally complained. But over the past 6 weeks, you have noticed the following:

  • Senior client stakeholders have stopped attending sprint reviews

  • Email response times have increased from same-day to 2–3 days

  • The client has not requested a single scope change in 8 weeks

What is the most accurate interpretation of these three signals together?

A)  Everything is fine — no news is good news
B)  The client is busy internally, it has nothing to do with you
C)  One of these signals is a warning, the other two are probably coincidence
D)  All 3 together form a specific pattern that typically precedes a client relationship breakdown — and you have a narrow window to act

👉  Take a moment. The answer is at the end of this issue

💡The fix: the three things I now track that have nothing to do with delivery

After losing this client, I made a decision. I would keep tracking delivery metrics, but I would add a second layer of measurement that had nothing to do with scope, budget, or timeline. I called it the Relationship Health Check. It takes 15 minutes every two weeks and has flagged a relationship risk on every single programme I've run since.

  Fix 1: The Client Engagement Audit

Every two weeks, I answer five questions about the client relationship, not the delivery:

  • increased, held steady, or declined?

  • changed in either direction?

  • to 'the project/the vendor'?

  • (a call about their business, their team, their challenges) in the past two weeks?

  • would this client say yes without hesitation?

If I cannot answer 'yes' or 'positive' to at least 4 of those 5, I schedule a relationship call within the week not a status call. A conversation about how they are experiencing the engagement. Not the deliverables. The experience.

BEFORE

Six warning signals across 14 months. Identified zero in real time. Client departed 4 days after delivery

AFTER

Same audit introduced on next programme. Signal detected in month 4. Relationship call scheduled. Client raised a concern that we had no idea existed. Resolved. Renewed for a second year

  Fix 2: The Invisible Stakeholder Map

On the programme we lost, the operations director who called to fire us was not the person we had been managing. We had been managing his deputy. The director, the person with actual authority over the renewal decision, had formed an opinion of us almost entirely from second-hand information.

I now build two stakeholder maps at the start of every engagement:

  • The visible stakeholders — the people in the meetings, on the emails, in the governance structure

  • The invisible stakeholders — the people who influence the renewal decision but are never in the room

The second list is the one that matters. I built a specific communication plan for those people — a cadence of brief, business-level updates that reach them directly rather than being filtered through their deputies. Not delivery metrics. Business outcomes. What we have changed in their organisation. What problems we have solved that they care about?

The person who signs the renewal is almost never the person who attended your sprint reviews

BEFORE

Stakeholder map had 9 people. All in regular governance meetings. Decision-maker had not attended a single review in 5 months. Had no direct communication from us

AFTER

Invisible stakeholder map on next engagement identified 3 people outside governance. Monthly one-paragraph business update sent directly to each. Renewal conversation started by them, not us

  Fix 3: The 90-Day Relationship Conversation

Every 90 days, I now have one conversation with the most senior client contact that is explicitly not a status call. I call it the 'bigger picture' conversation and I open it the same way every time:

"I want to take 20 minutes away from the project itself. Not to talk about delivery, but ask you whether this engagement is giving you what you actually need. Not the scope. The thing underneath the scope."

Then I ask one question and I do not speak for at least 90 seconds:

"If you had to explain to your board in two sentences what this programme has done for your business — not technically, commercially — what would you say?"

The answer to that question tells you three things simultaneously: whether the client can articulate the value you are delivering, whether they are invested enough in the outcome to have thought about it, and whether your framing of the work matches theirs.

On the programme we lost, I know with certainty that the operations director could not have answered that question. That is not his failure. That is mine. I never gave him the language.

BEFORE

14 months of delivery. Zero conversations about business outcome, only delivery output. Client couldn't articulate the value at the point of renewal decision

AFTER

90-day conversation at month 3 on next programme. Client said they were struggling to explain the value internally. We built a one-page business case for them. They presented it to their own board. Renewal was approved 3 months before contract end

🎯 What You Should Do This Week

Pick your longest-running active client engagement. The one where delivery feels smooth and you haven't worried about the relationship in a while. Now answer these three questions honestly:

  • Can the most senior person on the client side articulate, in two sentences, the business value of what you are building — without referring to the technical scope?

  • When did you last have a conversation with that person that was not about delivery, timeline, or budget?

  • If they left their role tomorrow, does anyone else at that client know your name, what you stand for, and why they should keep you?

If you hesitated on any of those 3: the relationship is more fragile than your dashboard suggests. Not because delivery is failing. Because delivery is all you've been managing.

Want the Relationship Health Check as a one-page template?

I've turned the 5-question audit and the 90-day conversation framework into a single sheet my teams use at every client review. Reply "relationship" to this email and I'll send it directly to you


🌐 Around the web this week

⚡  1 tool:  Grain.xyz — records and transcribes client calls automatically. I use it specifically to re-read client language patterns over time. The pronoun shift from 'we' to 'the project' becomes visible in black and white

📊  1 number:  According to Bain & Company research, a 5% increase in client retention increases profits by 25–95%. The industry spends the vast majority of its management attention on acquisition and almost none on the relationship mechanics that drive retention.

💬  1 quote:  "People will forget what you said, people will forget what you did, but people will never forget how you made them feel." — Maya Angelou. I have never read a sentence that applies more directly to client delivery than this one.

📣 SHARE THIS NEWSLETTER

If you’ve found the above information valuable, then hit the forward button on this email and share it with your network to help me grow.

👉 HERE IS THE ANSWER TO THE QUIZ

D — all 3 signals together

Any one of them in isolation is inconclusive. But all three together declining senior attendance, slower responses, and zero scope requests form a specific pattern that I now recognise immediately. It means the client has psychologically moved from 'invested partner' to 'managing an existing obligation'. They are not angry. They are not about to escalate. They are quietly preparing not to renew. That is the most dangerous client state because it generates no complaints for you to respond to.

The window to recover from this state is approximately 6–8 weeks. After that, the internal narrative has solidified and the decision is effectively made before the contract end date conversation even begins. The prompt fix is not a delivery action. It is a relationship call senior to senior, off the record, about how they are experiencing the engagement.

🧞‍♂️Your wish is my command

The operations director who called to fire us was not wrong to leave. We had delivered what was in the contract. We had not delivered what was underneath the contract, the confidence that we understood their business, not just their backlog. That we were on their side, not just on their payroll.

Most delivery leaders I know are excellent at managing projects. Far fewer are excellent at managing the human being at the other end of the engagement, who has to justify your existence to their own board every quarter.

The certification will never teach you that. The only thing that teaches you that is losing a client you thought you had earned.

One question before you go:

Think of your most important current client relationship. When did you last have a conversation with the most senior person on their side that was genuinely about their business — not your delivery? Hit reply and tell me. I read everything and I'll respond to every one

Until next Wednesday
Aman
www.amansingh.pro

Keep Reading