Float is the slack in a construction schedule, the days an activity can slip before it pushes out completion. It's also one of the most argued-over things in construction. When an owner's late decision eats up float that the contractor was quietly counting on, who absorbs the loss? The answer depends on the contract, and most owners don't find out what their contract says until a delay claim lands.
This guide explains the two kinds of float, the common ways contracts allocate them, and how owners can protect their position before the first update arrives.
Total float and free float
A critical path schedule calculates two kinds of float for each activity:
- Total float is how long an activity can be delayed without delaying project completion. Critical activities have zero total float.
- Free float is how long an activity can be delayed without delaying the next activity's earliest start. It's always equal to or less than total float.
The difference matters because float is usually shared along a chain of activities. Picture two linked activities on a building project: underground services (8 days) followed by site works (12 days), both feeding final commissioning. The path has 50 days of slack before it would delay the project. Underground services has 50 days of total float but zero free float: any slip pushes site works later. Site works also has 50 days of total float, and all 50 are free float, because nothing else waits on it until commissioning.
Those 50 days aren't 50 days per activity. They're 50 days for the whole chain. If underground services slips 30 days, site works is left with 20. Whoever uses the float first takes it from everyone downstream.
View the data
| Activity | Days | Total float | Free float |
|---|---|---|---|
| C. Underground services | 10–18 | 50 | 0 |
| E. Site works | 18–30 | 50 | 50 |
Why float ownership becomes a fight
Float is valuable to both sides, for different reasons.
For the contractor, float is flexibility. It lets them resequence crews, absorb a late delivery or recover from a bad week of weather without missing the completion date. Many contractors see the float in their schedule as their own time contingency, built in through good planning.
For the owner, float is room to make decisions. A tenant fit-out change, a late furniture order or a slow design review can be absorbed by float without triggering an extension of time or delay costs.
The conflict shows up when the owner uses float first. Suppose an owner's late approval of a curtain wall shop drawing delays the envelope by 15 days, and the envelope had 20 days of float. The completion date doesn't move, so the owner owes no extension. But the contractor now has five days of cushion instead of 20. When their own steel supplier is ten days late a month later, the envelope becomes critical and the project finishes five days late. The contractor argues the owner's delay caused the problem. The owner argues the contractor's supplier did.
What the industry guidance says
Two widely cited references set out the default position when a contract is silent.
The Society of Construction Law's Delay and Disruption Protocol (2nd edition, 2017), used internationally, says that absent an express term, float "is not time for the exclusive use or benefit of either the Employer or the Contractor." Its Core Principle 8 says an extension of time should only be granted to the extent the owner's delay is predicted to reduce total float below zero, in other words, when it actually affects the completion date.
AACE International's Recommended Practice 29R-03 on forensic schedule analysis takes a similar view. Absent contrary contract language, network float is a shared commodity to be used in the interest of the project, not for the sole benefit of either party.
In practice, that means "first come, first served." Whoever uses the float first gets it. Owners often like this approach because owner decisions tend to come early. Contractors often resist it for the same reason.
Neither document is law in Ontario, and neither overrides your contract. Courts and arbitrators will look first at what the contract says. These documents matter because consultants and tribunals often refer to them when the contract is silent or unclear.
Four ways contracts deal with float
Owners usually have a choice of approach when they draft or negotiate supplementary conditions:
- Silent contract (shared float by default). Nothing is written down, so the parties fall back on the general position: float belongs to the project, first come, first served. Simple, but it invites argument.
- Express shared float. The contract states that float is a shared resource available to both parties and that extensions are granted only when completion is affected. This is the same principle as the default, written down so nobody can argue otherwise.
- Owner-favourable float. The contract gives the owner first use of float, or bars the contractor from "sequestering" float through artificial logic, preferential sequencing or inflated durations. This needs careful drafting and tends to raise bid prices.
- Contractor time contingency. The contractor shows a separate, visible buffer between its planned finish and the contract completion date. That buffer belongs to the contractor. Activity float within the network is shared.
View the data
| Approach | How float is held | Owner delay that uses float | Main risk for owners |
|---|---|---|---|
| Silent contract | Default: shared, first come, first served | No extension unless completion moves | Arguments over who used float first |
| Express shared float | Contract says float belongs to the project | No extension unless completion moves | Contractor may hide float in the logic |
| Owner-favourable | Owner gets first use; sequestering barred | Owner delay absorbed by float | Higher bid prices; drafting must be tight |
| Contractor contingency | Visible buffer before contract completion | Network float shared; buffer is contractor's | Buffer must be shown, not hidden in durations |
How contractors sequester float, and how to spot it
Under a shared-float contract, a contractor has an incentive to show as little float as possible. A schedule with zero float everywhere means every owner delay becomes a critical delay. The common techniques are well known:
- Inflated durations. Activities are given longer durations than the crew will need, absorbing float into the bars themselves.
- Preferential logic. Activities are linked in a sequence that isn't physically required, so parallel work appears to be in series.
- Constraints and lags. Fixed dates or long lags pin activities in place and hide flexibility.
- An early-finish plan with a late-finish contract. The schedule shows finishing early, with the gap labelled as the contractor's float. That's legitimate if it's disclosed and agreed, but not if it's slipped in quietly.
The defence is a proper baseline review. On programs we've supported, the time to challenge float sequestering is before the baseline is accepted. Afterwards, the logic becomes the agreed record and is much harder to unwind.
Float and delay claims
When a delay claim arrives, float is usually the first battleground. The questions an owner's team should ask:
- Did the delaying event affect an activity on the critical path at the time it happened, or one with float?
- If it affected an activity with float, did it consume all of that float, or only part of it?
- Who used float earlier in the project, and was that use documented in the updates?
- Is the contractor relying on the original baseline's float, or on the float in the update current at the time of the event?
That last question trips up a lot of claims. Float changes every month as work progresses. A contractor who measures an owner delay against the original baseline can make it look critical when the contemporaneous update showed weeks of cushion.
Negative float: when the schedule is already late
Float can go below zero. When an update forecasts completion after the contract date, the critical path shows negative float equal to the overrun. Twenty days of negative float means the project is forecasting a 20-day overrun.
Negative float is useful information, and owners should want to see it. Some contractors avoid showing it by adding constraints that force the finish onto the contract date, or by compressing future durations with no plan behind them. The schedule then looks on time while the work isn't.
Ask for updates to be run without a hard constraint on the completion milestone, so the schedule calculates its own finish date. If the result shows negative float, the next questions are which activities caused it, whose risk they were, and what recovery plan is on the table. Those answers decide whether you're looking at a contractor recovery obligation, an extension request, or both.
What owners should do
- Decide your float position before tendering. Choose one of the approaches above and write it into the supplementary conditions. Get legal advice on the wording. Silence is a choice too, just a weaker one.
- Require float to be visible. Ask for total float on every activity in every update, plus a separate line for any contractor time contingency.
- Review the baseline for sequestering. Challenge unexplained constraints, lags and preferential logic before acceptance.
- Track float consumption monthly. A simple chart of total float on the critical and near-critical paths, update by update, shows when cushion is being eaten and by whom.
- Record who used float and why. When an owner decision uses float, document it in the update narrative. When a contractor delay uses it, document that too. Records made at the time are worth far more than reconstructions later.
The bottom line
Float is the cheapest time contingency an owner will ever get, but only if the contract says how it's shared and the schedule shows it honestly. Owners who settle the float question before signing avoid one of the most common delay disputes. Owners who leave it to chance usually discover the answer during a claim.
Want a second opinion on how your contract and schedule handle float? Book a 30-minute call with Benjamin R. Consulting.
Sources: Society of Construction Law, Delay and Disruption Protocol, 2nd edition (2017), Core Principle 8 and guidance 8.5; AACE International, Recommended Practice 29R-03, Forensic Schedule Analysis (2011); Critical path method (Wikipedia). The worked example is illustrative. This article is general information, not legal advice. Get legal advice on contract wording.