An outturn cost forecast is an estimate of what a project will have cost when it is finished, made at a point when something can still be done about it. It is the single most useful number in construction cost reporting and the one most frequently missing.
Committed cost is easy to report and safe to be wrong about, because it is a matter of record. A forecast requires judgement and can be tested against reality later, which is exactly why some reports avoid it.
The four components
The adjusted contract sum: the original sum plus instructed variations, less omissions. A matter of record.
Anticipated change: work that is known to be coming but has not been instructed, or has been instructed and not yet valued. This is where the judgement sits and where most forecasts are weakest.
Remaining risk: what is left in the register that has not retired, valued at a realistic expectation rather than at worst case or at nil.
Costs outside the contract: fees, statutory payments, utilities, finance and client side costs. On a development these are a substantial proportion and they are frequently reported separately or not at all.
Anticipated change is the whole exercise
A forecast that includes only instructed change is not a forecast. It is the current contractual position with a different label, and it will move every month by the amount of change that was already visible.
Capturing anticipated change means asking the design team what is coming, reading the technical queries, looking at the drawing revision register, and listening to what the site is saying. It is not a spreadsheet exercise.
Where a value is uncertain, a range with a stated most likely figure is more useful than either a precise number nobody believes or an omission. A board can act on a range.
Forecasts should move less over time, not more
A well built forecast starts uncertain and converges. Early in the job the range is wide and the risk allowance large; as packages are let and risks retire, both narrow.
A forecast that stays flat for months and then jumps is not a forecast, it is a record of when somebody finally recognised something. That pattern is the clearest sign that anticipated change is not being captured.
Plotting forecast outturn month by month across the job is a useful discipline in itself, because the shape of the line says more about the quality of the reporting than any single figure.
The point of the exercise is the decision
A forecast exists so that somebody can act. That means it has to arrive early enough for the options to still exist, and it has to identify what is driving the movement rather than only reporting it.
The actions available reduce sharply over the life of a project. Early on, scope, specification and procurement route are all in play. Later, the choices narrow to accepting the cost or arguing about it.
That is why a forecast that is slightly wrong and early beats one that is accurate and late, and it is why the discussion in the monthly cost report a developer can use puts it on the front page.
Phased and mixed use schemes need it cut
A single outturn figure on a scheme with several uses or phases hides more than it reveals. A commercial element running over and a residential element running under can present as a scheme on budget.
Forecasting by phase, by block, by use and by funding source is the same discipline as cost planning that way from the start, and it depends on the same structural decision having been taken at the first issue.
That structure is set out under mixed use and regeneration.
Being wrong is acceptable, being silent is not
Forecasts are judgements and some of them will be wrong. What matters is that the basis is stated, so that when reality diverges the reason is visible and the next forecast improves.
A forecaster who is never wrong is either working on a very simple project or is not forecasting. The failure to avoid is the one where nobody put a number down at all and the position emerged at the final account.
That failure pattern is described in late design changes.
What this means for you
Ask for forecast outturn and ask what anticipated change is in it. If the answer is that only instructed change is included, the forecast is describing yesterday.
Then ask what the forecast was three months ago and six months ago. The trend tells you whether the reporting is capturing change as it emerges or recognising it after the fact. Our approach sits under cost control and variations.
Want a forecast you can rely on?
Send the cost report, the instructions and the programme. We will build the outturn position and show you what drives it.