What You Can't See: Why the Ground Beneath Your Development Needs Testing, Not Just a Look
- Sep 22, 2025
- 3 min read
On most projects, the ground gets less scrutiny than almost anything else, even though it is the one part of the project you cannot inspect by eye once you have built on it. The soil beneath a site affects whether foundations settle, whether retaining walls hold, whether drainage works and whether earthworks behave as expected.
Getting that wrong can be one of the most expensive mistakes a project makes. The difficulty is that visual descriptions and laboratory testing do not always tell the same story.
Figure 1. Soil mechanics testing apparatus: triaxial (left), shear box (centre), and oedometer (right).
New Zealand has an unusually wide range of soils, laid down by rivers, the sea, volcanic eruptions and wind, and how a soil behaves under load, during excavation, or once drainage occurs, is the domain of soil mechanics, the science underpinning geotechnical engineering. Geology gives a broad, qualitative picture of the ground; whereas geotechnical engineering turns that into measured, quantitative answers an engineer can design to.
It starts with getting samples, by drilling or augering, and then describing them. Soil description is the trained, visual assessment of a sample, its colour, texture and how easily it can be moulded without cracking. It's a skilled job, but it's inherently subjective: two experienced people can describe the same sample differently, and a visual description can be misleading.
That is why soil classification uses laboratory tests against a standardised system such as the Unified Soil Classification System, or USCS, to measure the soil’s water content, particle-size distribution and plasticity. Classification gives a more objective basis for understanding what the material is and how it is likely to behave.
The distinction matters because the two can disagree, and the lab result is the reliable one. A description tells you what the soil appears to be. Laboratory classification gives a more objective basis for understanding how it is likely to behave.
The difference is repeatability. A visual description depends on judgement and can vary between people. Laboratory classification follows standardised procedures, so the result is more consistent, traceable and defensible. Where strength, settlement, drainage or compaction performance matter, further laboratory testing can then measure the properties the design depends on.
That gap is not academic. On a residential project in Wellington, a sample was initially described as sand, which would have been treated as suitable for bulk fill. Laboratory testing showed it was actually a sandy clay, a material that behaves very differently and would have brought stability problems if it had been used on the strength of the visual description alone. The lab test caught it before it became a problem in the ground.
On smaller residential jobs, description alone is sometimes used instead of lab testing to save money, and sometimes that's a reasonable call. But skipping the testing is a gamble on the ground behaving as it looks, and when it doesn't, the cost of construction delays, future repairs, or a dispute with a neighbour whose property is affected, dwarfs what the testing would have cost.
On larger projects, developments, schools, social housing, rail, ports, relying on visual description alone is not sufficient, and the consequences of getting it wrong are correspondingly larger.
The case for laboratory testing is simply that it replaces a judgement with a measurement. It costs a little more upfront and reduces a large source of risk: foundations and earthworks designed for the soil that's there, fewer surprises during construction, less chance of expensive remediation later, and less material wasted designing around an unknown.
As the saying goes, what you can't see matters most, and on a building site, what you can't see is the ground.
This is the kind of decision we help clients make every day. Our geotechnical engineers and accredited testing tell you what's beneath your site, so the design rests on measured evidence rather than a visual guess.
If you are planning a project and are not sure what ground investigation it needs, talk to us. We will advise what is worth testing and what is not, so you are neither under-investigating a risk nor paying for testing you do not need.
