You Manage Projects in Four Countries. You’ve Visited Each Site Twice.
Written and narrated by Serguei Poppeleer.
Retail brands manage shopfronts and back-of-house facilities across dozens of countries from a single European headquarters.
Most of those portfolios are running on two completely different plans at once.
The steering committee sees one. The contractor runs on another. The gap between them widens every week. Nobody at head office can see it — because every governance signal still reads green.
Here’s what nobody says in the project review:
The pattern I see across every distributed store portfolio:
But the contractor’s team still runs off a WhatsApp group and a notebook on site. By week two the two plans have diverged. Nobody at headquarters can see it.
The site re-sequenced three critical activities last week. Nobody updated the steering deck. A milestone that was showing 60% complete has a critical-path activity already three weeks behind it.
Weekly email from the site supervisor. WhatsApp update from the local project manager. Monthly steering report from the lead contractor. All describing different versions of the same project. None of them wrong. None of them the truth.
The project that looks fine in the steering report and fails at handover had two plans running in parallel for months. The dashboard stayed green. The project drifted.
The fundamentals that keep getting missed:
Not the governance version. The operating version. The one that reflects how work is actually being sequenced this week.
If the answer requires thought, they have already diverged.
If the answer is nobody, there is no single operating truth. That is where overruns and disputes are created.
What cross-border retail management needs is not better reporting software. It is one operating plan — built at the granularity the site actually runs on, reconciled every week.
This isn’t about the project managers managing this gap every day. It is about organisations that add reporting layers when the actual problem is that no single operating truth exists.
When there are two plans, the contractor interprets the gap between them. That interpretation arrives in the final account.
The assumption that creates the gap
Most project leads on distributed portfolios operate on a version of the same belief: “If I have weekly calls and a programme manager on it, the project is under control.”
A weekly call that reports status is not control. It is a snapshot of the governance version of the project — which may or may not match what is actually happening on site.
Control is a single operating plan that all parties recognise as reality, updated weekly, reconciled against actual site progress. It is not assembled for governance meetings. It reflects how work is actually being sequenced on the ground.
Why two plans always emerge
The master schedule is structured around milestones the client cares about, formatted for steering decks, updated at the frequency the reporting cycle requires. It is not built for how a construction site actually sequences work.
The site team does not update it in real time because the granularity it requires is not how they think about delivery. They work in days, trades and material deliveries. The master schedule works in months, milestones and percentage completions. These are different operating models.
The gap between them is where slippage hides. A milestone can show 60% complete while the critical-path activity that determines whether it will finish on time is already three weeks behind. The percentage looks fine. The sequencing has already determined the outcome.
Most projects never reconcile the steering version against the site reality. The gap widens every week until a crisis forces an honest reckoning — usually when recovery is expensive and time has already been lost.
Part of a store expansion across several African countries, French- and English-speaking, each with its own legal system and customs regime. Three separate reporting streams were feeding the head office project team simultaneously: a weekly email from the site supervisor, a WhatsApp update from the local project manager, and a monthly steering report from the lead contractor. All three were accurate. All three described different versions of progress. None of them was the operating truth.
The switch to a single operating plan, updated every Monday and shared with all parties, removed the discrepancy in two weeks. When the plan showed a milestone at risk, it was at risk. Reporting time per project dropped from 90 minutes to under 20 minutes per week. The site and the steering committee were working from the same reality for the first time in the project.
Five questions that tell you where your plan coherence actually stands
Including WhatsApp threads, site office whiteboards, informal supervisor sequences. Which one does the contractor actually use on Monday morning?
Was it ever formally documented?
Including the site supervisor and the lead contractor. If not, there is no single operating plan.
That number is the cost of running two plans simultaneously.
If the answer is the second, there are two plans. The contractor is using the other one.
The objections this always raises
“Our project manager is on top of it — we have weekly calls.”
Weekly calls that report against the governance version of the programme are not control. They are reporting. The question is whether the plan being discussed matches what is actually happening on site. If your project manager cannot produce a single document that the contractor also uses as their operating plan, there are two plans.
“We have a master schedule — everyone can see it.”
A master schedule everyone can see is not the same as a plan everyone uses. The schedule shows what was planned. The operating plan shows what is actually being executed this week. If the contractor is not using the same document as the client, the visibility is an illusion that will be corrected at handover.
“The contractor is responsible for their own programme.”
The contractor is responsible for executing. The client is responsible for the governance architecture that connects execution to what was agreed. When those two things diverge — and they always diverge without a single operating plan — the financial risk sits with the client at final account.
AI-powered project management tools can now process multiple reporting streams and surface discrepancies in real time. Some platforms use pattern recognition to detect when site progress has diverged from the programme and flag it before the steering meeting.
This is valuable when there is one authoritative plan to diverge from. If the project is running on two parallel plans, AI will surface two versions of reality more efficiently. It will not decide which one is true. It will not create the single operating truth that the governance architecture was never designed to produce.
AI can detect when plans diverge. It cannot replace the discipline of running one plan that all parties use as the operating truth.
What one real plan actually looks like in operation
It is updated weekly. It reflects actual sequencing at the level the contractor manages — trades, materials, handoffs — not the level the client monitors. It is the same document the site supervisor uses to plan next week and the client uses to assess programme risk.
When it shows a milestone at risk, it is at risk. When it shows progress, that progress has been verified against what has actually been built — not estimated against a planned percentage.
The site visit becomes a verification exercise at defined milestones rather than the mechanism by which the client discovers what is happening. The weekly call becomes a decision meeting rather than a status update, because everyone in the room is working from the same reality.
The project that closes with a dispute usually had two plans running in parallel for months.
The disagreement at handover was not new. It was the moment the two versions of reality finally met in the same room.
Where does your delivery system actually stand?
The Delivery Score measures your governance architecture across the five dimensions of The 5Cs Framework™. Fifteen Yes/No questions. About four minutes. Instant result.
Get my Delivery Score — free → Read the Special Report →