Foundation design on made ground and former industrial sites
Fill varies over metres, not hundreds of metres. The design has to cope with that variation, and differential settlement is the failure it must prevent.
Made ground foundation design starts from an uncomfortable fact: you cannot assume anything about fill until you have looked at it, and looking at it in three trial pits tells you about three trial pits. Ground that has been tipped, levelled, backfilled or demolished into place varies over metres, not over hundreds of metres, and the foundation solution has to cope with that variation rather than with an average.
For a self-builder on a former yard, a filled quarry, a demolished factory or a levelled garden, this is usually the largest single unknown in the budget.
Why made ground behaves unpredictably
Natural ground has been consolidating for thousands of years. Made ground has not.
Fill placed without compaction continues to settle under its own weight for years. Fill containing timber, paper, plaster or vegetation decays, leaving voids. Demolition arisings include lumps of concrete alongside fine material, so stiffness varies sharply over short distances. Ash, clinker and industrial residues can be aggressive to buried concrete. And old basements, tanks, wells and foundations sit under fill in places nobody recorded.
The consequence for a building is not usually catastrophic failure. It is differential settlement: one part of the structure moving relative to another, cracking walls, jamming doors and distorting drainage falls.
That is the failure mode the foundation design has to prevent, and it is why "the ground felt firm when we dug the trench" is not evidence of anything.
Investigation that finds the variability
The investigation has two jobs: find the depth and nature of the fill, and find the competent stratum beneath it.
A reasonable scheme for a single dwelling on suspected made ground usually combines several techniques. Trial pits give a direct view of the fill, its composition and any obstructions, to the depth a machine can safely reach. Boreholes or window sampling go deeper and recover samples for testing. Dynamic probing gives a low cost indication of relative density between the deeper holes, which is how you map variability without paying for a borehole every five metres.
Location matters more than quantity. Put investigation points at the corners of the proposed footprint and at any point where historic mapping suggests a change: the line of a demolished wall, the edge of a former pond, the position of a filled pit.
Testing should include chemical analysis for sulfate and pH, because these determine the concrete specification for buried elements, alongside the contamination suite. Where organic content is suspected, ask for it to be measured rather than described.
Work to the current British Standard on site investigation and have the results interpreted by a geotechnical engineer, not just reported. A factual report without an interpretative report leaves the designer to make the judgements the specialist should have made.
Foundation options and their trade-offs
Five approaches cover most small sites.
| Option | Works best where | Main drawback |
|---|---|---|
| Deep trench fill to competent ground | Fill is shallow, say under 2.5 metres | Cost rises steeply with depth, excavation stability |
| Dig out and replace with engineered fill | Fill is shallow and the area is small | Disposal cost, needs testing and compaction records |
| Piled foundations with ground beams | Fill is deep or variable | Cost, access for rigs, obstructions, vibration |
| Raft foundation | Fill is uniform and reasonably competent | Vulnerable to differential settlement if fill varies |
| Ground improvement such as vibro stone columns | Granular fill of moderate depth | Unsuitable in fill with voids or high organic content |
The decision is rarely purely technical. Access for a piling rig, the cost of removing contaminated arisings to landfill, and the presence of neighbouring buildings all change the answer, and on a tight urban plot the theoretically best solution is sometimes not buildable.
Where piling is chosen, the pile type matters as much as the decision to pile. Driven piles are faster and often less costly but generate vibration that becomes a problem next to existing buildings, which is the subject of ground movement and piling next to existing buildings.
Settlement and differential movement
Total settlement is rarely the problem. Differential settlement is.
A building founded partly on competent natural ground and partly on fill will move differentially, and the transition is where the cracking appears. Where the footprint crosses a boundary between fill and natural ground, either found the whole building in the same way or design a movement joint at the transition.
Trees complicate this further on clay. Shrinkable clay under made ground, with mature trees nearby, requires foundation depths set by the species, the mature height and the distance, and warranty providers publish tables for this. Removing a tree before building can cause heave as the clay rehydrates, which damages foundations designed only for shrinkage.
Suspended floors rather than ground bearing slabs are usually specified over made ground, because a slab bearing directly on fill will settle with it. A suspended beam and block or precast floor spanning between the foundations avoids that, and the void beneath it also helps with ground gas ventilation where that is a concern.
Drainage needs the same thought. Rigid pipework laid in settling fill will fail at the joints, so flexible jointed pipes and flexible connections at the building are standard practice.
Obstructions and buried structures
Obstructions turn a fixed price foundation package into a variable one, and they are the single most common source of overspend on made ground sites.
Old strip foundations, floor slabs, basement walls, fuel tanks, wells, air raid shelters and abandoned services all turn up. Historic mapping and a site walkover will suggest where they might be, but the only way to be reasonably sure is to dig.
Two practical measures. Price a contingency explicitly rather than hoping, and get the ground investigation to look where the building is going rather than where the machine could easily reach. An investigation carried out in the accessible corner of a plot, with the house then positioned elsewhere, is worth very little.
Where a tank is found, it changes the contamination position as well as the foundation position, and both have to be resolved before the slab goes down.
Cost and programme consequences
The costs to plan for are the investigation itself, the foundation solution, the disposal of arisings, and the delay if the investigation produces a surprise.
Two general points hold. Investigation is a small fraction of the foundation cost and a smaller fraction still of the cost of getting the foundation wrong, so under-investigating to save money on a made ground site is a poor trade. And the foundation decision affects the superstructure, the drainage and the floor construction, so it needs to be made before the design is fixed rather than after.
Programme effects are usually about lead times: piling rigs and specialist contractors are booked weeks ahead, and a design change from trench fill to piles in March can mean a start in May.
Evidence for the planning and building control stages
Planning and building control want different things, and self-builders routinely conflate them.
Planning is concerned with ground conditions where there is a contamination issue, a stability issue, a mining legacy, an effect on trees, or an effect on neighbouring land. The usual planning submission is a desk study and preliminary risk assessment, and where the site has a history, a Geotechnical Assessment setting out the ground model and the likely foundation approach. Where an existing building is being retained or altered, a Structural Survey may also be required. The detailed design follows under condition or under Building Regulations.
Building control is concerned with the foundation design itself under Part A of the Building Regulations, and will want the interpretative geotechnical report, the foundation design and calculations, and inspection at excavation stage before concrete is poured. Warranty providers impose their own requirements, often stricter, and on a self-build the warranty provider's technical team is frequently the most demanding reviewer of the three.
The approach to evidencing design to the geotechnical code is set out in Eurocode 7 geotechnical evidence in planning.
This is general guidance, not legal advice; check the position with your local planning authority or a qualified professional before relying on it. Geotechnical and contamination work should be carried out by suitably qualified and experienced persons. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval. For specialist technical reports, qualified professional input may be required.
Enter your postcode on PlanScope to see which reports a site with a former industrial or filled history is likely to need.
Frequently asked questions
How much ground investigation does a single house on made ground need?
Enough to find the depth and nature of the fill and the competent stratum beneath it, with points located at the corners of the proposed footprint and at any change suggested by historic mapping. Trial pits, deeper boreholes or window sampling, and dynamic probing between them is a common combination.
Why is differential settlement worse than total settlement?
Because a building can tolerate moving down uniformly far better than it can tolerate one part moving relative to another. Differential movement cracks walls, jams openings and distorts drainage falls, and it occurs where the footprint crosses a boundary between fill and natural ground or between fills of different density.
Should I pile or dig out the fill?
It depends on depth, composition, access and disposal cost. Digging out and replacing with engineered fill suits shallow fill over a small area. Piling suits deep or variable fill but needs rig access and raises vibration issues next to neighbours. Ground improvement suits granular fill without voids or high organic content.
Does made ground affect the concrete specification?
It can. Sulfate content and pH in fill determine the class of concrete required for buried elements, so chemical testing should be specified alongside the contamination suite. Ash, clinker and industrial residues are common sources of aggressive ground conditions on former industrial sites.
Do I need the foundation design for planning permission?
Usually not. Planning is concerned with contamination, stability, mining legacy, trees and effects on neighbouring land, and a desk study with a preliminary risk assessment normally suffices at application stage. The foundation design itself is assessed under Part A of the Building Regulations and by any warranty provider.
What happens if we hit an old tank or foundation?
It changes both the foundation position and the contamination position, and both have to be resolved before the slab is laid. Obstructions are the most common cause of overspend on made ground sites, so price a contingency explicitly and locate investigation points where the building will actually sit.
Sources and further reading
This article is general guidance about the planning system in England and, where stated, the rest of the UK. It is not legal or professional advice. Check your local planning authority’s validation requirements and take professional advice for your specific site.