How to Keep a Construction Project on Budget From Day One

A construction project goes over budget for reasons that are settled long before anyone breaks ground. The three that matter most are the cost plan, the risk allowance you set against it, and the contract you sign. Fix those at design stage and the majority of overruns never get started.

Here is the backdrop for anyone costing a scheme this year. The RICS Building Cost Information Service (BCIS) published its latest five-year forecast on 4 June 2026: tender prices are expected to rise 15.5% over five years, with 5.6% of that in the next twelve months alone. A budget built on last year's rates is already behind before you appoint a contractor. So the figure you carry into board approval has to be priced for the market you will buy in, and old rates will not get you there.

We have spent since 1992 costing, monitoring and running projects across Greater Manchester and beyond, most of them for developers, funds and lenders who answer to someone for every pound. What follows is where we put the effort to keep a number honest, in the order it matters.

Where does a project's budget actually get decided?

At design stage, on paper, months before the first invoice. The cost plan is the document that turns a design into a number, and under the RICS New Rules of Measurement (NRM1) it is built in two parts: a base cost estimate for the measured work, and a separate risk allowance for what the base estimate cannot yet see. The base estimate is only as good as the design it measures. Bring a quantity surveyor in at RIBA Plan of Work Stage 2, while the layout is still moving, and you can still change the things that drive the cost. Bring one in at Stage 4, once the design is fixed, and you are left costing decisions somebody else already made.

This is also where optimism creeps in. HM Treasury's Green Book, the rulebook for public spending, tells appraisers to add up to 44% to the capital cost of a standard project at outline business case to correct for optimism bias, then to reduce it as the design firms up. Few private schemes carry anything close to that. You do not need 44%. You do need to notice that the instinct, on every project, is to price the version where nothing goes wrong.

How much contingency should you carry?

Enough that you never have to phone a lender for more, and no more than the design justifies. NRM1 gives sensible starting points, and they fall as the design matures:

Two allowances, doing different jobs. Design development risk covers the detail still to be drawn. Construction risk covers the ground, the existing building and the way you procure the work, and it is the one people underestimate on older stock.

The contingency line is also the first thing a client asks us to trim when the numbers do not fit the approval. We have done it under pressure, and we have watched it come back to bite the same project three months later, once the ground told a different story. A cost plan that only balances because someone shaved the risk allowance to hit a target has not solved anything. The overspend is still coming. It has just not arrived yet. If a scheme only works at 2% contingency at Stage 2, the honest answer is that it does not work yet.

If you want an independent read on a cost plan before it goes to your board, that is the day job for our cost management and quantity surveying team.

Does the contract you sign change the final cost?

More than most clients expect. A JCT or NEC contract is the machinery that decides who carries which risk, and risk that sits with the wrong party comes back later as a claim. The bigger lever is how you tender. A single-stage, fixed-price tender feels like certainty, because you get one number on day one. On a new build with a clean site and a finished design, it often is.

On a refurbishment or a conversion, we will not recommend it, and this is the point some of our competitors will argue with. On an old building, a fixed price obtained early is a false certainty. The contractor prices the risk they can see, loads it heavily for the risk they cannot, and recovers the rest through variations once the walls are open. A two-stage tender or a construction management route costs more in fees and gives you a number you can stand behind, because it is priced against a design that has been interrogated first. Certainty is worth paying for. A cheap certainty on an unknown building is the expensive option wearing a disguise.

What does the ground do to a Stockport budget?

More than the drawings ever will, and around here that is a local story. Stockport sits on the eastern edge of the Manchester Coalfield, where the Coal Measures reach the surface. Deep mining stopped in the 1970s, but shallow historic workings and unrecorded shafts are still under a lot of development land, along with the mine gas that can migrate through it. Any scheme in a Coal Authority development high-risk area needs a Coal Mining Risk Assessment, and if that desk study finds a shaft near your footprint, you are into intrusive investigation and possibly grouting before the foundation design is even settled.

Then there is the made ground. Much of the buildable land in Stockport and across Greater Manchester is former mill and terraced housing, cleared and levelled with whatever was to hand. It looks like a flat plot. It is often a metre or two of broken brick, ash, old foundations and buried slabs sitting on the real bearing stratum below. You find out which by drilling, not by hoping.

That gap is the risk.

So we push clients to commission the ground investigation early, before the cost plan is signed, not after it. A borehole survey costs a few thousand pounds and takes a couple of weeks to arrange in a busy market. A foundation redesign after the contractor has mobilised costs a great deal more, and it is the single most common reason a Greater Manchester budget moves in the wrong direction.

Who watches the budget once you are on site?

Somebody with no reason to tell you what you want to hear. A budget set well at design stage still drifts on site: variations get instructed, provisional sums firm up higher than allowed, and valuations quietly run ahead of the work in place. Independent project monitoring and auditing catches that early, while a change is still a conversation rather than a line in the final account. For a bank or a fund releasing money against milestones, that oversight is the whole point of fund monitoring: a check that the drawdown matches the progress on the ground, before the cash leaves the account.

Good change control is unglamorous, and it is what keeps the final account close to the tender. Every instruction priced before it is issued. Every valuation measured against the work actually standing, by our construction project management team rather than the contractor's own assessment. None of it is complicated. It is just done, or it is not.

Where to start

Before the next cost plan lands on your desk, get the ground investigation commissioned and get an independent set of eyes on the risk allowance. Those two moves prevent more overspend than anything that happens later. If you are weighing up a scheme in Stockport or anywhere across the North West and want that read early, talk to our team or call us on +44 (0)161 476 4456.

Phil Spring is the founder of Spring & Co and has worked in construction cost and project management since 1992.