Who killed the project’s budget?

It was found on a Tuesday morning, on the cold floor of a meeting room. There was no blood. There were numbers. And there was that particular silence that appears when someone realises the “Contingency” line had been more optimistic than it looked.

I was called in urgently, as one often is when a budget stops breathing.

“We need to understand what happened. Yesterday it was still under control. Today it’s… like this.”

The sentence remained unfinished. Nobody likes to say “dead” when the investors are still on the other end of the line.

I looked at the crime scene. An open dossier, nervous post-its, and a spreadsheet with a final total that no longer fitted inside the cell. I knew the pattern. A budget rarely dies suddenly. It dies by accumulation. It dies from small decisions that do not look like decisions. It dies from haste, omissions, and sentences delivered with excessive confidence: “we’ll adjust that later.”

It was the perfect crime. Almost.

I gathered the evidence. Nothing looked dangerous. Everything looked as if it belonged to the process.

Suspect number one: the investor flyer

It was found in a folder called “Final ideas v3 FINAL REALLY”, with a more international subtitle: “Investor Briefing”. It presented an appealing vision for the development: “premium experience”, “premium finishes”, “sea view”, “construction cost 800 €/m²”.

Its alibi was simple: “I was only presenting the idea.”

And that was true. The problem was not ambition, nor vision. It was the distance between what was an intention, what was an assumption, and what had already become a commitment. When those boundaries are not clear from the beginning, the budget is left to fill in the blanks. And filling in blanks is rarely cheap.

Suspect number two: the seductive render

It arrived early, charmed everyone, and left the meeting without a single accusation.

“I am only an image,” it declared.

Technically, that was true. But an image can create expectations that the structure, the building envelope, the MEP systems and the schedule may struggle to accommodate without cost impact. The render rarely kills on its own. It has more class than that. It simply pushes expectations in a direction without asking how much they weigh.

Suspect number three: the material sample “only as a reference”

Small, elegant, placed on the table as if it carried no responsibility at all.

“It was just to understand the tone,” someone said.

The sample has that talent: it enters as a reference and leaves as an obligation. When a material enters the project as a desire rather than an informed choice, the budget pays the difference between “similar” and “the same”. And that difference is rarely poetic.

Suspect number four: the rush to start on site

It entered the room before everyone else, with an argument that was difficult to oppose: “we cannot lose time.”

Its alibi was convincing: “the design is practically ready.”

Practically. The favourite word of those who believe details can be resolved during execution. But starting construction without coordinated design information is like beginning an investigation before collecting the evidence. Incompatibilities do not disappear when the works begin. They simply change location. They move from paper to site, where every conflict between disciplines demands urgent decisions, rework, changes and additional time. They look like small adjustments. Until they begin to appear in measurements and invoices.

Suspect number five: the Bill of Quantities with gaps

It looked respectable, well formatted, full of lines, units and subtotals. But it hid a few grey areas. Discreet notes such as “to be defined”, “to be confirmed”, “included in another chapter”.

It defended itself naturally: “some definitions were still missing.”

Perhaps they were. But what remains undefined does not disappear. It simply waits for a less convenient moment to return. And when it returns, it no longer comes as a forecast. It comes as an extra.

Suspect number six: the thin specification

It had few pages, many generalities, and admirable confidence in everyone’s goodwill. It spoke about deadlines, but not much about acceptance criteria. It spoke about execution, but not much about testing, warranties, responsibilities and exclusions.

“It was meant to simplify things,” it explained.

The problem is that simplifying without defining does not simplify. It transfers the discussion to later. And later is almost always a more expensive place. A weak specification does not protect the budget. It leaves it to negotiate alone.

Suspect number seven: the WhatsApp “just a small change”

It entered quietly, with no attachments, in the friendly tone of someone only trying to help.

“It’s simple. We just need to change…”

Its alibi was modern: “it was meant to make communication easier.”

Changes are legitimate. What kills the budget is change without control, without acknowledged impact, without a clear decision and without an owner. WhatsApp is not guilty for existing. It becomes a problem when it serves simultaneously as meeting record, site instruction, approval mechanism and project archive.

Suspect number eight: the optimistic schedule

It was attractive. It used sober colours, elegant arrows, and spoke English with ease: “waterfall”, “critical path”, “milestone”. It did, however, have one worrying feature: no float. It always pointed towards a clean future, with no difficult dependencies, no lengthy reviews, no external authorities, no doubts.

“It was the objective,” it said.

An objective is not a plan. A schedule without realistic dependencies turns technical delays into emotional urgency. And emotional urgency has a known effect: it accelerates decisions that still needed confirmation.

When I finished hearing the suspects, the room fell silent. Everyone wanted a single culprit. A name. A simple explanation that would allow the case to be closed.

That was when I said, with the irritating calm of someone who has seen too many budgets fall:

“The budget was not killed by one suspect. It was killed by all of them.

The investor flyer confused vision with commitment. The seductive render made unresolved issues look simple. The material sample turned a reference into a requirement. The rush to start on site pushed decisions forward before the conditions existed to make them properly. The Bill of Quantities with gaps left too many questions unanswered. The thin specification placed too much trust in each party’s interpretation. The WhatsApp “just a small change” accumulated decisions without formal record. And the optimistic schedule created an urgency the project was not yet ready to sustain.

Each one delivered a small blow. None seemed fatal. But a budget rarely survives repeated small blows.

The revelation is not meant to punish. It is meant to change the method

If you want to protect the budget, do not rely on luck. Rely on discipline. A brief that distinguishes intention, assumption and decision. A project with a single state. A Bill of Quantities that identifies unknowns instead of hiding them. Comparable proposals, with clear inclusions and exclusions. Decisions made at the right moment, with cost and time impact made explicit. A simple change control mechanism, where every change has a reason, an owner and a consequence.

The budget should not be treated as an inevitable victim. It needs to be protected before construction, before procurement, before haste.

I closed the dossier. The room was still cold, but it no longer felt like a postmortem.

It was a beginning. The kind of beginning that makes no promises. It makes choices.