Most project managers start their jobs with the odds already set by decisions made elsewhere in the business. Andy Pritchard explains why technical strategies should put project performance first – and let the margin follow

When a job goes late, we ask whether the project manager was good enough. We rarely ask whether the business gave them a fair chance.
Long before a project manager is appointed, the business has decided how the job will be bought, when the design will be procured, who writes the programme and how performance will be measured. Those technical strategies set most of a project’s odds before anyone reaches site. For many contractors, commercial considerations come first – buy each package late, at the lowest price, and protect the margin on paper.
I think that gets cause and effect the wrong way round. Commercial results are the consequence of operational performance, not the cause of it.
The saving that doesn’t survive
I have analysed around £5bn of procurement spend across several tier 1 contractors. Three patterns turn up repeatedly, and each feeds the next. Packages bought late, on price alone, cause delay. Design, especially the contractor’s designed portion, arrives incomplete and late, often because the package was bought late. And the resulting delay is found too late to recover.
When the Italian city of Turin switched to awarding public works on the lowest price, winning discounts rose sharply. But at least half of that saving went on renegotiation and extra costs, and jobs finished later. On Californian road paving contracts, plans changed after award. The hidden cost of handling those changes, from disruption to haggling, came to around 10% of the winning bid.
To be fair to lowest-price award, a separate study of private building contracts in California found it works poorly when the job is complex and the design unfinished. On simple, fully designed packages, it remains a reasonable choice. Complex, unfinished work is exactly where project managers get hurt.

Performance first, commercial second
The alternative is to write every technical strategy to answer one question first: what does this do for the performance of the project? Three changes do most of the work.
First, choose and measure partners on outcomes, not just price. Supply chain charters set out what each side can expect of the other. Each key trade gets a framework built on the outcome that matters most to it: quality for fire protection, say, or social value for labour.
Second, buy the design on time. Let packages with a contractor’s designed portion while the design is still at RIBA Stage 3 or early Stage 4, so design isn’t late because the package was bought late.
Third, measure from the day the job is won. Procurement from tender to order, design delivery, earned value, the master programme and weekly planning on site. Each is an early warning of one of the three patterns.
Some of the strongest evidence comes from utility works. When an Italian utility began awarding future work to its grid maintenance contractors on audited quality and safety scores, compliance rose from about 25% to about 90% and held for at least a decade. Overall prices did not change significantly.

Make it weekly, and make it shared
Measurement only works if someone acts on it quickly.
Every site reports by Friday. An independent team reviews it on Monday. Concerns go to the project leads on Tuesday, anything off track goes to operations the same day, and anything still open reaches the board on Wednesday. In my experience, the time from a problem appearing to the leadership team knowing about it fell from about 60 days to three.
The same data should go to every key partner, every week. But feedback makes performance worse in over a third of cases, often when it turns attention to the people rather than the task. Share trends and talk them through together. Don’t publish a league table.
The weekly heartbeat is PPC, or percent plan complete – the share of tasks committed to for the week that are fully finished. On projects using the Last Planner System, PPC typically runs at 60% to 80%. Buildots’ global benchmarks report, drawn from more than 250 projects and reported by Construction Management in October, found the typical project delivers only 63% of its scheduled weekly output, and 70.7% of activities start more than a week late. That measures work installed, not PPC, but the story is the same.
PPC alone won’t tell you a job will finish on time, though. One Brazilian study found PPC on its own explained less than 10% of cost and time overruns. The reasons logged behind missed work explained far more. The value is in fixing those reasons before the work is due.
The real payoff is before site
When you know which partners perform, you can afford to involve them early. With design procured earlier, the specialists who will install the work help resolve it while it is still moving. And each partner writes the programme for its own package rather than inheriting one. A programme written by the people delivering it, on resolved design, is one they can keep.
None of this depends on a heroic project manager, but on the business. Researchers looking at US federal works and services contracts found that a clearly more competent buying office cut cost overruns by about 29% and delays by about 23%.
Do it this month, whichever seat you are in
If you commission the job. Ask your contractor how it measures its partners, how often they see that data, and when the design packages will be bought. The answers tell you more about your programme than the programme does.
If you run the job – find your site’s PPC and the reason behind every missed commitment last week. In the event that you can’t, that is the first strategy that needs writing.
If you lead the business – read your supply chain, design and planning strategies and ask of each: is this written for project performance, or for the buying price? Then ask how a partner’s performance affects who you buy from next.
Project managers can’t fix on site what the business got wrong before they arrived. Give them strategies built for performance and the margin will follow.
Andy Pritchard MCIOB is a business improvement consultant, Construction Manager of the Year Gold Medal winner and author of Going for Gold: Constructing Project Managers.









