Shell and fit-out: cost shares within building cost

This is the heart of the confusion. DIN 276 structures costs by Kostengruppen, cost groups, that is by which element or installation a cost belongs to. Rohbau and Ausbau, shell and fit-out, are instead Leistungsbereiche, work sections, a structure based on which trade carries out the work and how it is tendered.

The two systems do not coincide. Professional datasets note expressly that the work sections grouped under shell and fit-out correspond to cost group 300 only approximately. Excavation, for instance, may fall into cost group 500 where it serves the external works.

A practical rule follows: a percentage for the shell can be compared with a cost group schedule only where both refer to the same total. A shell share of building cost, a shell share of total construction cost and a shell share of the investment including the site are three different figures.

The shell covers everything that makes the building structurally sound and weathertight. The boundary is not defined by any standard and is drawn differently in different offers.

• excavation, the pit and any temporary support
• foundations, slab or cellar storey with tanking
• loadbearing walls and columns
• floor slabs and stairs as structural elements
• roof structure and covering
• flues and service shafts
• conduits and drainage runs laid during the shell phase

Not normally counted as shell are windows and external doors as finished components, facade insulation where let separately, screed, plaster, drylining, all finishes and the entirety of the building services. Anyone comparing two offers therefore checks first where the line was drawn.

The divergent percentages arise precisely at that boundary. A supplier including windows and insulation within its shell package arrives at a much higher share than one stopping at the weathertight shell in the narrow sense.

The most robust statement on how the balance has moved comes from a study of new residential construction: the share of fit-out trades including building services within cost groups 300 and 400 rose from 46 per cent in 2000 to 54 per cent in 2014. The shell share fell correspondingly from 54 to 46 per cent.

Reading that figure correctly matters. Shell costs did not fall. They simply rose more slowly than fit-out costs, and within fit-out the services trades developed disproportionately.

The causes are well established and continue to operate: tightened energy requirements, raised demands on sound and fire protection, more complex plant, mechanical ventilation with heat recovery, heat generators running on renewable energy, charging infrastructure, photovoltaics and building automation. They explain the shift in shares over recent decades.

One notable side effect concerns fees: the shift raises the chargeable costs of building services designers and tends to reduce those of structural engineers. Anyone setting ancillary costs from older experience will miss that redistribution. The framework is set out in the article on ancillary construction costs under KG 700.

The three figures do not contradict one another. They refer to three different denominators. The larger the base, the smaller the shell's percentage share, even though the absolute amount is unchanged.

The same mechanism applies to shell benchmarks per square metre, quoted between roughly 600 and 2,000 euros. The low figures describe the shell in the narrow sense; the high figures describe a completion stage already including facade, insulation, screed and early services. Without the specification, none of these figures can be placed.

Three practical conclusions follow from the distribution. All three concern where savings should be sought.

The shell is largely fixed once the design is settled. Massing, storey count, spans, whether there is a cellar and the roof form determine the shell quantities. After the design is complete only material and execution decisions remain, and their scope is limited. The effective moment for shell savings is at concept stage.

Fit-out stays open to influence longer, and stays at risk longer. Finishes, sanitaryware and specification level can still be adjusted late. Which is exactly why fit-out costs frequently drift upwards during the works: decisions fall at a point when the budget is already committed.

Services are the fastest growing block and the least often designed early. They are routinely set too low in early cost estimates because older experience is drawn on. Adding an uplift to an old benchmark corrects the price level, not the increased scope of works.

Within the shell, costs concentrate in a few large trades and a long tail of small ones. The ordering is remarkably stable across projects; the absolute shares move with construction method and foundations.

Concrete and masonry form the largest single share within shell work and one of the largest items in the whole new-build index. Excavation, by contrast, is the block carrying the greatest advance uncertainty, because it depends directly on ground conditions and is where variations most frequently arise.

For cost estimating this means a ground investigation ahead of the cost estimate is economically sensible, even though it first appears as additional expenditure within cost group 700. It reduces uncertainty within cost group 300 by an order that routinely exceeds the cost of the investigation.

Published percentages cannot replace a cost estimate, but they can check one. The procedure is straightforward.

1. Summarise your own cost calculation by work section and form the shares, each referenced to the building cost of cost groups 300 and 400.
2. Compare against the known orders of magnitude, that is a shell share around 45 per cent for conventional housing.
3. Explain divergences rather than correcting them. A conspicuously high shell share may point to demanding foundations, large spans or a complicated roof. A conspicuously low one may point to a very high plant or specification standard, or to a gap in the estimate.

The third step is the decisive one. A departure from the experience value is not an error but a signal. Adjusting the figures to match the experience value, instead of looking for the cause, discards exactly the information the comparison was made to obtain.

The Federal Statistical Office publishes price movement for shell and fit-out work separately, and the two series diverge. For November 2025 shell work was reported at 2.5 per cent and fit-out work at 3.8 per cent above the previous year.

Within the shell too, the trades scatter considerably. Over one comparison period concrete and masonry work sat in the low single digits, while roofing, excavation and carpentry rose considerably more.

For updating a benchmark this means a flat uplift on the total does not capture the internal shift. Where a cost estimate is built up by work section, the corresponding sub-indices should be used. The calculation is set out in the article on updating costs with the construction price index.

• Residential construction costs in Germany
• What a construction cost benchmark includes
• Updating costs with the construction price index
• Ancillary construction costs under KG 700
• Cost planning under DIN 276