The search most people make is structural engineers near me, and by the time they make it, the drawings are usually already done. Sometimes the planning application has gone in. Occasionally the builder has started.
That sequence is backwards, and I’ve spent enough years unpicking it to know exactly what it costs. This is what actually happens on those jobs, and what I’d tell anyone standing at the start of one.
The call that comes too late
The pattern is consistent. Someone has a set of attractive drawings, a builder has priced them, and then building control asks for structural calculations.
The engineer arrives, looks at the proposal, and finds that the span is too long for the beam depth the design assumed. Now the ceiling drops by 150mm, or the beam becomes a bulkhead across a room that was supposed to feel open.
Neither outcome is a disaster. Both are avoidable, and both are cheaper to solve on paper than on site.
The worst version is when the problem sits in the foundations. Existing footings on Victorian and Edwardian houses are often shallow, and if the new structure loads them beyond what they’ll carry, the answer is underpinning. That’s a five figure conversation nobody budgeted for.
What we’re actually looking for
A structural survey isn’t a formality. On an older house it’s an investigation, and the useful information is mostly hidden.
I want to know which walls are carrying load, and that isn’t always the ones people assume. Internal spine walls in terraces frequently carry the first floor and sometimes the roof, and removing one without a properly sized replacement is the most common serious error I see.
I want to see the existing foundation depth, ideally through a trial hole rather than an assumption. I want to know the joist direction on every floor, because that determines where new support can go.
And I want to understand what’s already been done. A house that was extended in the 1990s may have a beam sitting on masonry that was never designed to take it, and adding to that structure compounds the problem.
Trees, drains and the things nobody photographs
Two site conditions cause more redesign than anything structural inside the house.
The first is trees. Clay soils across London and Surrey shrink and swell with moisture, and a mature tree within influencing distance changes the required foundation depth substantially. BS 5837:2012 governs how trees are assessed alongside construction, and root protection areas often constrain where you can build at all.
Homeowners are frequently surprised that a neighbour’s tree matters as much as their own. It does, and it can’t be removed to solve the problem.
The second is drainage. A public sewer crossing the site means a build over agreement with the water authority, and that has its own timeline running parallel to planning. Manholes inside a proposed extension footprint need diverting or accommodating, and both cost money.
Neither of these appears in the photographs that inspire projects, which is exactly why they get missed.
Why the planning route changes the structure
This is the part that surprises people most. The consent route and the structural design aren’t separate problems.
If your address sits inside a conservation area or under an Article 4 Direction, the enhanced permitted development depths disappear, and you’re into a full application. That usually means a shallower extension, which means shorter spans and lighter steels.
The reverse is also true. A scheme designed for maximum depth on the assumption of prior approval becomes structurally over specified if the depth then has to reduce.
Merton illustrates this well. Article 4 covers parts of Wimbledon while neighbouring streets keep their rights, and the practical result is that two similar houses need different schemes. Any work with architects wimbledon should establish which side of that line the property falls on before the structural approach is fixed, because the answer changes the beam schedule.
Where the sequence should actually sit
The order that works is straightforward, and it’s the opposite of what most people do.
Establish the consent route first. Survey the existing structure second, including foundations and load paths. Design the layout third, inside both sets of constraints. Produce the structural calculations alongside the design rather than after it.
Having the engineer and the architect in the same conversation is the practical version of this. It’s why Building Regulations Approved Document A compliance stops being a late stage obstacle and becomes a design input.
Extension Architecture keeps structural engineering in house for exactly this reason, which is unusual for a practice of its size. On a Malden Road project in Kingston, the structural approach was settled during the layout stage, which meant the steel schedule was priced accurately before the tender rather than revised during the build.
The party wall conversation
Two sections of the Party Wall etc. Act 1996 come up on almost every extension in a terrace or semi.
Section 3 applies when you’re cutting into or altering a shared wall, which includes seating a beam in it. Section 6 applies when you’re excavating within three metres of a neighbouring structure and below its foundation level, which most extension foundations do.
Notice periods run one to two months depending on the section, and a dissenting neighbour triggers a surveyor process. Structural design feeds directly into this, because the depth and proximity of your excavation determines whether section 6 applies at all.
I’ve seen builds stopped by neighbours who were never served notice. It’s one of the few things that can halt a project that already has planning permission and building regulations approval.
What good looks like
If the structural work is done properly and early, you shouldn’t notice it. Beams sit inside ceiling zones, foundations are right first time, and the builder prices something that doesn’t change.
If it’s done late, you’ll notice it constantly. Bulkheads appear, ceiling heights drop, the tender gets revised, and the contingency gets spent before the walls go up.
The engineer isn’t a box to tick before building control signs off. On an older house, the engineer is the person who tells you what’s actually possible, and that information is worth far more at the start than at the end. See more