The call tends to arrive in the same shape. An owner has a building, a rough sense that something inside it is old, and one question they want settled before any other: how much of the work will the state pay for? It is a reasonable place for a non specialist to begin, because it is the only part of the process that already has a number attached to it, and because the industry has spent years training clients to ask exactly that. In an office in Gräfelfing, on the western edge of Munich, the reply is usually another question, and it has nothing to do with money. What is this building? When was it put up, what has been done to it since, how does it actually behave in February? Dmitri Berdnikow, who runs the engineering practice BBP Engineering there, will not price anything before he has an answer. Clients occasionally read this as evasion. It is the opposite: it is the one step of his method he refuses to shorten.
The question he puts last
Subsidies are not a side issue in German building renovation. They are large, they are technical, and knowing your way around them is a genuine professional skill. Berdnikow’s argument is narrower and stranger than a rejection of them. It is about position in a sequence. A subsidy programme is written by an authority, for a class of buildings, to encourage a class of measures. It is not written for the building in front of you, and it does not know how that building behaves. Start the conversation there and you have quietly let a document you did not write set the technical scope of a project you will live with for two decades.
Run the other way round and the same programme becomes something else entirely: a discount on a decision you would defend anyway. First establish what the building is, then assemble the options that are technically sound for it, then work out what each of those options costs to own over the years it will be owned, and only then look at which of them the state supports. The subsidy stops being the goal of the exercise and becomes a consequence of getting the exercise right. Sometimes a well engineered solution turns out to sit outside the programme, and the honest version of that conversation is that the owner is choosing between a supported measure and a correct one, which is a decision an owner is entitled to make with the trade off visible.
| Step | The usual order | The order he works in |
|---|---|---|
| 1 | Which programme will pay, and how much? | What is this building, and how does it behave? |
| 2 | Which measure fits the programme’s conditions? | Which measures are technically sound here? |
| 3 | Design the measure so that it qualifies. | What does each option cost to own over the horizon? |
| 4 | Discover afterwards what it costs to run. | Which of the sound options does the state support? |
The two columns look like a matter of emphasis on paper. They diverge in the third year of ownership, when the running costs of a measure chosen for its eligibility arrive on a bill nobody modelled, and by then the subsidy has been disbursed once and the building has another seventeen years to go.
Twenty years to get his name on the work
The sequence is not a young man’s idea. Berdnikow has been in construction since the 1990s, and he climbed the German technical ladder in its slowest and most literal form: technical draughtsman first, then technician, then engineer on large projects, then energy consultant. The detail that matters is not the ladder but the letterhead. For more than twenty years, every drawing he produced went out under somebody else’s name.
That changed in 2017, the year the first commission arrived that stood on his name rather than an employer’s. By his account the working relationship behind it lasted nine years, running until 2026. The company followed later, as it usually does with practices that begin before they incorporate: BBP Engineering GmbH sits in the commercial register of the Amtsgericht München under HRB 256629, entered in May 2020 under the earlier name BBP GmbH, active, with Berdnikow as managing director.
It is tempting to treat 2017 as an administrative date. It is closer to a change in what a calculation is. A twenty year projection produced by an employee is a document that leaves the building with the client and stays with the firm as a file. The same projection produced by a man whose own name is on the cover is a liability he will still be holding in year eight, when someone calls to ask why the figure looked different. Nothing about the arithmetic changes. Everything about who absorbs being wrong does, and that is the pressure that turns a preference for the right order into an actual procedure.
What the city confirms, and what it does not
Munich has been building schools at an unusual pace, and the projects are public. The city’s own page for the Bildungscampus Lerchenauer Feld describes a Gymnasium, a primary school, a double and a triple sports hall and a swimming hall, with construction running from the first quarter of 2024 to the second quarter of 2027. A separate school construction project on Triebstraße appears in district council papers and in the public record of the architects working on it. Those are the city’s facts, and they are solid.
His part in them is a different order of claim. Berdnikow says that from 2017 until this year he worked on school and childcare buildings for the city of Munich, and we could find no independent record naming him or his practice on either scheme. German municipal construction rarely publishes the engineering firms sitting under the architects, so the silence is weak evidence of anything, but it is silence, and this article will not convert an unverified role into a credential. The practice also supplied construction volumes for those projects. Those figures appear nowhere outside its own website, so they are not printed here.
None of this is a reason to drop the subject, because the subject is not the school buildings. Subjects of a Vocation Review portrait supply facts, documents and interview answers, and they get a window to correct facts about themselves before publication; the framing and the conclusions belong to the desk. What interests the desk here is a working sequence that can be examined on its own merits, and one episode in which the sequence changed shape because it failed him.
The analysis that walked out of the room
For years, the most demanding part of Berdnikow’s work was free. A prospective client would arrive with a building, and he would do what he does: walk it, read it, work out how it actually performs, assemble the options that hold up technically, put numbers against each of them, and explain the reasoning in detail, because explaining the reasoning is how a technical person sells. Then some of those clients took the reasoning to a cheaper contractor and had it carried out there.
Nothing improper happened. The hard thinking had already been done and handed over in a form clear enough to act on, and what remained was execution, which many firms can do adequately and one of them will always do for less. The usual lessons available at this point are that clients are ungrateful or that the market is a race to the bottom. Berdnikow drew a third one, and it is the reason he is worth writing about: he had mispriced his own product. He had been treating the analysis as marketing for the execution, when the analysis was the part that carried his judgement and could not be reproduced cheaply, and the execution was the part that could.
The free consultation was not generosity. It was giving away the one step that carried his judgement, and then competing on the step where judgement barely matters.
So he rebuilt the process rather than his sales pitch. The detailed preliminary analysis became a separate service with its own price and its own deliverable, ordered on its own terms and not bundled into the hope of a later contract. Two things follow from that, and both of them are structural rather than attitudinal. Anyone who orders the analysis has already decided to pay for thinking, which filters the room before the work starts. And the analysis itself is free to conclude that the owner should do nothing yet, or should do a great deal less than they imagined, without costing him a project he was quietly angling for, because the project is no longer what he is angling for.
This is a first hand account of his own practice, and the clients who walked away are not available to us for comment. What can be said is that the change is not a slogan. A philosophy costs nothing to announce; a priced service line with a deliverable attached is a decision that shows up in what the practice will and will not do for free.
Where the forecast stops
The habit that makes the reversed order workable is a smaller one, and Berdnikow applies it unprompted: he marks the line between what is measured and what is modelled. A building’s fabric can be surveyed. Its equipment can be inventoried. What it consumed last year is on a meter. Everything after that, the twenty year cost of owning it, the savings from a given measure, the point at which an investment returns, is a projection built on assumptions about energy prices, occupancy, maintenance and weather, and the uncertainty widens with every year further out.
Blurring that line produces a better looking document. It also produces a client who is angry in year six, holding a printout that promised a number the world declined to deliver. Marking it produces a duller document and a client who calls back. The habit is also what keeps the sequence honest at the front end: if you begin with the subsidy, the model turns into an argument for a conclusion that has already been chosen, and the assumptions drift, one at a time and each defensibly, towards supporting it. If you begin with the building, the model has to survive the possibility of being wrong before anybody has committed to anything.
By his own account, he also teaches. He describes himself as an Honorardozent at the Munich and Upper Bavaria chamber of crafts and says he has sat on an examination board since 2025. The chamber names no such thing publicly that we could find, so it stands here as his statement rather than ours. If it is as described, it fits: an examiner is obliged, every time, to say which part of an answer is knowledge and which part is estimate, which is the same discipline as the marked forecast, practised on someone else’s work.
What the order actually buys
Nothing in this sequence is exotic, and that is the honest difficulty with writing it down. State it as a principle, that you should understand a building before specifying work on it, and the entire industry will nod. The difference is procedural, and procedure is where agreement usually dies. The market’s intake conversation starts with the money question because clients ask it first and firms answer what clients ask, and holding a different order has a cost that lands precisely at the point of sale: you lose the owner who wanted a figure in the first ten minutes, and you spend expensive attention before anyone has committed to anything.
Which is why the two halves of this story are really one decision made twice. Refusing to start with the subsidy and refusing to give away the preliminary analysis are the same claim about where the value sits: the judgement is the product, and the drawings, the applications and the subsidy paperwork are downstream of it and largely reproducible. A practice that hands over its judgement in order to win the reproducible work has arranged its economics against itself, and it will experience that as a pricing problem, or as a problem with the character of clients, when it is a problem of sequence.
Berdnikow runs a small office on the edge of Munich. It took him more than twenty years to get his own name onto the work, and a run of prospective clients walking out with his thinking to arrange the work in the other order. Both of those are ordinary experiences in a technical trade. What is not ordinary is treating the order of the questions as the method itself, and everything after it as drawings.