The hole was never the problem. Standing in it was.

Start with the hole and start filling it.

Fill the hole, lever by lever

Front-load the schedule of values. You already know how from Week 9 — weight the early activities, bring the money forward. On this job that pulled about $66,000 of income six months early, and dropped into the worst month it takes the hole from $60,462 to roughly twenty thousand. The cost: the client's QS is not stupid, and an aggressive front-load sours the relationship and gets scrutinised on every valuation thereafter. Used with judgement, it is the biggest single lever. Used greedily, it is a fight you have every month.

Certify faster. Week 17's cycle had a week of valuation before the twenty-eight-day clock even started. Get your applications in clean, agreed, and early — no disputes for the engineer to sit on — and you can pull the payment lag from thirty-five days towards twenty-one. That alone frees around thirty thousand of working capital. The cost here is discipline: it means immaculate applications, every month, with the measure agreed before you submit.

Take the mobilisation advance. Many contracts offer an advance at the start — ten percent is common — to fund your setup. On a million-dollar job that is $100,000 landing in month one, and on its own it would bury the entire $60,462 hole with room to spare. It is not free money; it is repaid out of your later certificates, and it usually needs a bank guarantee. But as a timing tool it is decisive, and the contractors who skip it are often the ones who later hit the wall.

The lever that costs a phone call

The three levers so far all pull money in sooner. The fourth works the other end — when your own money goes out — and it is quietly the most powerful of all.

Think about the order the money moves in. On the job that sinks you, you pay your steel supplier at day thirty and the client pays you at day sixty. For thirty days you are out of pocket on every pound of material, funding the gap yourself, on every job, forever.

Now reverse it. Negotiate sixty-day terms with your supplier — the $197,400 steel order from Week 10 is a large enough prize that a supplier will often agree — while getting your client to certify and pay in thirty-five. Now you are holding the supplier's money for twenty-five days after the client has paid you. The job is funding itself. You are running on other people's cash instead of your own.

That is the cash conversion cycle: the gap between paying out and being paid. Positive, and the project is a source of cash. Negative, and it is a drain that gets deeper every time you win more work — which is exactly the overtrading trap from last week, now expressed as something you can fix. And fixing it costs nothing but a negotiation you were going to have anyway. Most contractors never think to have it, and finance every job out of their own overdraft as a result.

Forecast the cash, not the profit

Every lever so far assumes you can see the hole coming. That is the fifth discipline, and it is the one Week 18's contractors did not have.

They had a profit forecast. It said all three jobs made money, and it was right. A profit forecast is updated monthly, warns you about a bad job, and is completely blind to the wall, because the wall is not made of losses — it is made of timing, and profit forecasts do not model timing.

A cash forecast is a different instrument. It is the two curves from Week 16, projected forward, month by month, for every job stacked together — and the number it exists to find is the single deepest point of the combined gap. That number is how much cash the whole company must be able to lay its hands on in its worst month. For one job it was $60,462. For Week 18's three, stacked, it was $181,386 — and if they had run this forecast, they would have seen that number form months before it opened, in time to pace the wins, phase the starts, or arrange the finance. They did not run it, so the first they knew of it was a bounced cheque.

Update it weekly, not monthly. Cash moves faster than profit, and by the time a monthly report shows a cash problem, you are already in it.

Every lever is a trade

One honest word before you go and pull all of these at once. None of them is free.

Front-loading strains the client relationship. The mobilisation advance carries interest and ties up a guarantee. Stretching your suppliers can cost you goodwill, or a better price you would have got for paying early. Even certifying faster costs real administrative effort, every month, forever. Cash management is not a set of tricks that beat the system — it is a set of deliberate trades, each one buying you timing at some price, and the skill is knowing which to pull, how hard, and when to stop before the relationship that gives you your next job is the thing you spent.

But pull them with judgement and the arithmetic is remarkable: the same project, earning the identical $172,992, can either drain sixty thousand of your own money for eight months or fund itself from the client's. Same profit. Completely different business.

Practical insight

Work out your project's cash conversion cycle with two numbers: the average days until your client pays you, and the average days until you pay your suppliers. Subtract the first from the second.

If it is negative — you pay out before you collect — then every pound of growth is digging the hole deeper, and you should treat winning more work as a cash event before you treat it as good news. If it is positive, your projects fund themselves, and you can grow as fast as you can build.

Then ask whether anyone on your project updates a cash forecast weekly and knows its deepest point. If the honest answer is that you have a monthly profit report and nothing else, you are managing the number that cannot kill you and ignoring the one that can.

Key takeaways

✔ The cash hole is not a fixed feature of a project — it is a timing number you can move, without changing the profit by a cent.
✔ Front-loading the SoV (Week 9) pulled $66,000 forward and roughly halved the worst-month hole — at the cost of the client relationship.
✔ Faster, cleaner certification pulls the payment lag from 35 days towards 21, freeing about $30,000.
✔ A 10% mobilisation advance is $100,000 in month one — enough on its own to bury the $60,462 hole.
✔ The cash conversion cycle is the master lever: collect in 35 days, pay suppliers in 60, and the job funds itself. It costs a negotiation.
✔ Forecast cash weekly, not just profit monthly. The profit forecast is blind to the wall; only the cash forecast shows the deepest point in time to act.
✔ Every lever is a trade — strained relationships, interest, guarantees, admin. Cash management is choosing which price to pay, not beating the system.
✔ The same $172,992 job can drain your money or fund itself. Same profit, different business.

What is coming next

That closes Phase D. You can read a project's cash, size its hole, and manage the timing so the business survives the work.

Now we lift our eyes from the single project to the company that runs a dozen of them at once. Every job has carried a share of head office — that $124,051 of overhead in the very first tender — but where does that number come from, and what happens to a company that wins its work by cutting it?

Phase E opens next week with the money that belongs to no single project: direct costs, indirect costs, and the site overheads that decide whether the company behind the projects makes money at all.