Tracking decisions and milestones: how to stay on top of a construction project with more than two hundred choices

01Two things you are actually tracking

Most people treat everything that has to be done as one list. That's why it doesn't work.

A construction project asks you to keep two things separate.

Decisions: things that have been chosen or not chosen. Which floor tile. Which heating system. Whether the bathroom window opens or is fixed. The kitchen layout. Which door handle. Hundreds of them, big and small mixed together.

Milestones: moments at which a decision becomes irreversible or expensive to change. Plastering closes electrical changes. Screed closes the placement of underfloor heating. Kitchen installation locks in the worktop dimensions. Provisional acceptance opens the ten-year liability period.

The trap is to treat both as one to-do list. The result: you don't see the decision before you've passed the milestone that needed it.

The way out is to keep them separate, and link each decision to the milestone that determines its deadline. A decision then always has two dates that matter: when it was taken, and by when it must be taken. The distance between "now" and "must be decided by" is your action window.

Anyone who doesn't know the structure of those milestones can't set action windows either. For that, there is the article on the twelve phases of a Belgian construction project, which describes the underlying spine.

02The six states of a decision

A good decision isn't "yes or no". A good decision moves through six states, from the moment you know it exists to the moment it is physically in place.

• Open. You know the decision exists. You haven't investigated yet.
• Research. You're gathering options, prices, samples.
• Quoted. You have at least one written price quotation in hand.
• Decided. The choice is made and communicated to contractor or supplier.
• Ordered. The order has been placed, a delivery date confirmed.
• Installed. It is in place, checked.

One example to make concrete how long that journey runs. A heat pump.

In phase 2 your architect mentions you have to choose a heating system. The decision is open. In phases 2 and 3 you sit in research: you compare air-water, geothermal, hybrid. In phase 5 you ask three installers for a price, and the decision is quoted. In phase 5 or 6 you choose a system, and the decision is decided, with an impact on your EPB calculation. In phase 8 you order the unit with a confirmed delivery date, and the decision is ordered. In phase 9 it becomes installed.

One single decision, six states, five phases, six to eight months. Anyone who notes that as a single line in a notebook misses three times where it stands and three times when it needed its next step.

The second example is less dramatic, but shows how it goes wrong too. Bathroom floor tiles.

In phase 5 the decision first appears on your architect's list. Open. In phase 9 or 10 you walk into one showroom, find something you like, and think it is decided. No one has passed the tile dimensions to the contractor laying the screed, and no one has asked the supplier for the lead time. On site the screed is on the schedule, and the tiles you had in mind turn out to have a three-week lead time. You pick a different model one evening because the planning can't wait. Afterwards you no longer know whether that was your choice or time pressure's choice.

What was missing wasn't the taste. What was missing was that the decision never moved from research to quoted and ordered, while the physical milestone quietly approached.

03Linking decisions to milestones

A decision without an end date costs you money. Either you decide too late under pressure, or you decide too early and the world has moved on in the meantime.

The discipline: every decision in the system has a "must be decided by" date, linked to the milestone that depends on it.

A few concrete links.

The bathroom tile choice must be decided before the screed, because the layout of the underfloor heating circuits depends on it. Decide too late and you get either a sub-optimal circuit layout or additional work to undo it.

Socket positions must be decided before plastering. Cutting and chasing into a finished wall is additional work, often far more expensive than the original installation.

The kitchen layout must be decided weeks to months before installation, because Belgian kitchen suppliers work from delivery lists, not on demand. An early decision gives you time, a late decision gives you a compromise.

The milestones themselves follow the twelve phases. Each phase has one or more points at which something becomes irreversible. Early in the project these are legal and administrative moments, like submitting the environmental permit or signing the contractor agreements. In execution they are physical moments: screed, plastering, kitchen and bathroom installation. At the end it is the provisional acceptance, which contractually opens the ten-year liability period [1].

Your task as the homeowner is not to make every decision early. Your task is to take no decision too late.

04A system that works

Underneath all the theory sits a simple system. Three components.

One place for all decisions

Every meaningful decision gets a line. Per line you track: what has to be decided, which of the six states it is in, who is involved, when it must be decided, and a link or reference to the quote, the sample, or the email thread that goes with it.

Two things make the difference here. You update the state weekly, not once every two months. And your partner sees what you see, because otherwise the decisions happen when you are not there.

A milestone calendar

Every irreversible moment in the project is on a calendar. Plastering start. Screed. Kitchen installation. Order moment for internal doors. Provisional acceptance. EPB declaration deadline.

The calendar doesn't have to be a work of art. A list with dates and the matching phases is enough. What matters is that the calendar and the decision list are linked to each other. Every decision points to the milestone that sets its deadline.

Some milestones are legally anchored and not negotiable. The EPB declaration must be filed within twelve months of occupancy [2]. Others are contractual. The majority are technical: plastering falls on a day, and what isn't decided by then is additional work.

A fixed cadence of review

Once a week, fixed slot, fifteen to thirty minutes. Two weeks is enough in quieter phases, weekly in phases 5, 9 and 10, where decision density is highest.

How such a session runs.

You look four to six weeks ahead on the calendar. For every milestone you see coming in that window, you check which decisions are still open or in research. For every open decision you set the next action: visit a showroom, request a quote, follow up a hanging quote, or hash it out with your partner. You update the system in the moment, not "later tonight". It's fourteen minutes of work in a quiet week, thirty in a busy one. The difference between a controlled project and a project that happens to you sits right there.

In practice you also learn to shorten your session. The first two weeks cost more, because you still have to learn the structure. After that it becomes routine. Not because the work shrinks, but because you no longer have to search for where things are.

The system itself is, practically speaking, unimportant. Excel with three tabs works. A paper notebook with a fixed template works. Notion with a database and a calendar view works. A dedicated tool works if it holds decisions, milestones, budget and documents together. The method is what makes the difference, not the instrument. How the budget actually attaches to the decisions is in the article on tracking your construction budget.

05Why this often falls over in practice

It isn't a failure of effort or intelligence. They are predictable failure patterns of a system that is too complex to keep in your head.

Spreadsheets go stale. Works in weeks one to four. By week ten the file hasn't been opened in two weeks. Decisions are happening in WhatsApp in the meantime.

Notebooks don't sync with your partner. One person takes the notebook to the site. The other is at the office. A decision is taken in the moment, in a conversation with the contractor, and never makes it back into the system.

Email threads die. A decision sits buried in a six-week-old email thread, and no one finds it back. How you check during contractor selection what a supplier actually tracks does, by the way, change how much you have to find back yourself.

Documents live in five places. Quotes in your mailbox. Photos in WhatsApp. Signed contracts in a folder on your laptop. Samples in a box in the garage. An approved variation-order quote in a message you've since swiped away.

The real cost isn't any one of those points. It is that they add up. You work full time. You have a family. You run a household. The mental energy to keep three tools open and reconcile them between sessions is what runs out first. Not the tools themselves.

Anyone who recognises themselves here, recognises themselves rightly. It is not an accusation. It is how most construction projects run.

06What this means for your project

A well-organised construction project isn't about being stricter. It's about holding less information in your head at once. A system doesn't make the choices for you. It ensures the choices you do make stay findable, the choices still waiting on you stay visible, and the legal and technical milestones don't show up as surprises.

Start small. One tab with decisions, one with milestones, one fixed moment per week. Scale up only when you notice this minimum holds. Many people start too ambitiously, fall back after six weeks, and conclude that it doesn't work. The problem wasn't the discipline. It was the amount of system they were trying to carry at once.

One last note. If you notice the system hasn't been kept for two weeks, don't go back to zero. Take one half-hour session, update the ten most urgent decisions, and plan again. Wanting to fully catch up is the second big failure pattern. The method also works with gaps in its history, as long as you keep the upcoming milestones sharp.