BIM Execution Plan (BEP): What It Contains and Misses

A BIM Execution Plan (BEP) is the document that defines, before anyone opens a model, the goals and uses of the model, each discipline’s roles, the Level of Development required per phase, the software and exchange formats, the delivery schedule, and the method used to check model quality. Most teams build it around ISO 19650 or the BIM Execution Planning standard in NBIMS-US.

What the document rarely answers is what the model gets validated against. Most BEPs say “to be verified on site” without saying how, and on site that “how” ends up being a tape measure and the judgment of whoever showed up that day. At Foundtech we write the BIM execution plan for each project as part of our BIM modeling service, and the clause we check hardest is exactly that one: which data proves the plan was met.

Nine-minute read. By the end you’ll know what a BEP should contain, how it changes before and after appointment, where it breaks when the project starts from an existing building, and you’ll have a minimum template for your next job.

A BEP that wins the bid and a BEP that holds up on site are two different documents. The first lives in a PDF nobody opens again. The second gets checked, week after week, against what is actually being built.

You’ve seen it. The coordination meeting where the BEP calls for LOD 300 on structure and nobody can confirm the model on the table got there. The meeting minutes that close with “to be verified in the field” and are still open three months later. The model updated by hand after every field change, on the hope that someone wrote down the real dimension. The BEP didn’t fail. It never had anything to run on.

At Foundtech we capture as-built reality with 3D laser scanning and turn it into BIM models under ISO 19650, with more than 200 projects across Europe and the Americas and over 10 million m² (about 108 million sq ft) modeled, in alliance with BIM Facility AG in Switzerland. This article doesn’t ask you to change your template or the standard you follow. It asks you to decide, clause by clause, which data will prove what you wrote.

What a BIM Execution Plan contains

A complete BEP answers seven questions: what the model is for, who does what, at what level of detail, where the information lives, how it gets exchanged, when each piece is delivered, and how it gets checked. Templates differ in order and naming, but if one of these sections is missing, the plan is incomplete.

The third column is the one almost no template includes. It’s also the one that separates a BEP that gets met from a BEP that only gets signed.

Pre-appointment vs. post-appointment BEP: two documents, not one

ISO 19650 separates two moments in the plan. The UK BIM Framework guidance on ISO 19650-2 puts it this way: each bidding team submits a pre-appointment BEP describing how it proposes to manage project information, and the winning team confirms a post-appointment BEP with real people, schedule and resources behind it.

In practice, the first is a proposal and the second is a commitment. The common mistake is carrying the bid BEP into the contract without checking its assumptions, starting with the most expensive one: where the data on the existing building comes from. If your contract also calls for BIM deliverables, see how to align your construction contract with BIM deliverables, since the BEP is the technical document that turns that contract into detail.

The most common gap: the existing building

A BEP forces you to state what the model is for, and that purpose decides what information you need. More information doesn’t make a model more reliable. What matters is that it’s the information you need and that it can be checked. A BEP can ask for LOD 400 across the board; the useful question is whether the information about what’s already built is reliable enough for the use you’ll give it.

On ground-up construction, the design drawings are the starting condition and there’s nothing else to measure. On a renovation, an expansion or any building with history, the plan has to state four things about its starting data: what was measured, when, to what accuracy, and against which state of the building. If that data has no date and no source, the BEP starts on assumptions, however well it’s written.

That’s where 3D laser scanning comes in. The terrestrial scanners we use in our Scan to BIM service record the building with millimeter-level precision (±2 to 10 mm, roughly 0.08 to 0.4 in, depending on the equipment and site conditions) and produce a dated point cloud that can be overlaid on the model.

Two projects show what that looks like. At Claraspital in Basel, the design team for a hospital project had no architectural, engineering or MEP information on the existing building: more than 2,500 laser scans went into the BIM model. At Cine Metropol in Zurich, about 200 scan stations plus drone captures became a BIM model with MEP in three weeks, so complex auditoriums could be replanned before the renovation. In both, the BEP started from measured data, and that certainty is exactly what inherited drawings can’t give you.

Where a BEP holds up and where it stays on paper

Not every clause breaks the same way. There are three points where the gap between the document and the site gets expensive, and all three share a structure: the BEP says what should happen, but not which data confirms it happened.

1. Level of Development: what the document promises and what the model delivers

The BEP assigns an LOD per discipline and phase. What it rarely specifies is how that level gets audited once the model exists. When the problem is physical (a wall that isn’t where the model says, a slab with a different slope) no LOD checklist catches it: the checklist confirms the model holds certain information, not that the information matches what was built. The point cloud gives you a reference to audit the real LOD, not the declared one.

2. Coordination and clash detection: the BEP says how, not with what data

The coordination protocol sets meeting frequency, federation software and who resolves each clash. But clash detection is only as good as the models it compares. If the existing structure was modeled from ten-year-old drawings, the BEP can be followed to the letter (weekly meetings, signed minutes) and still carry an error nobody sees until the new piping hits a beam that, in the model, wasn’t there.

3. As-built handover: where most BEPs break

Closeout usually requires as-built drawings, but the BEP rarely defines how they’ll be produced. If they’re done by hand, adjusting the original drawing with whatever the superintendent remembers, the margin of error depends on someone’s memory. That’s what happened at Werdhölzli, a wastewater treatment plant in Switzerland: when the owner started planning its modernization, it found it had no up-to-date drawings or engineering documents. The drawings weren’t badly designed. They were outdated, and for a BEP that’s the same problem.

What an unvalidated BEP actually costs

A NIST study put the cost of inadequate interoperability in U.S. capital facilities at $15.8 billion a year, with about two-thirds of it borne by owners and operators during operations and maintenance, not by the design team. Every clause met on paper but not on site eventually turns into a change order, rework, or a clash discovered after the piping is already in.

At Foundtech we report up to 35% less rework and up to 25% time savings on projects where point cloud validation starts with the model, instead of showing up as an open item during construction.

What 3D scanning won’t fix

Laser scanning doesn’t replace the BEP, doesn’t assign roles and doesn’t repair internal coordination that isn’t working. It doesn’t measure structural performance, temperature or vibration either. If your problem is process (nobody reads the minutes, responsibilities are unclear), what’s missing is project discipline, not geometric data. Our piece on BIM adoption in construction covers why culture outweighs tools there.

And there are projects where a traditional survey is still the right call: a small space with simple geometry, a minor renovation with low risk from an out-of-tolerance dimension, or early design where nothing is built yet. Scanning there means paying for accuracy the project doesn’t need yet. The full comparison is in Scan to BIM vs. traditional survey.

A minimum BIM execution plan template

Whether the owner hands you ISO 19650 information requirements or an NBIMS-US based template, this is the baseline. Each section carries a sample line for an existing building about to be renovated.

The row most teams skip is “starting data.” If your template doesn’t have it, add it before anything else.

What this looks like the first time on your project

The process has four stages, and only the time each one takes changes with project size. First, on-site capture and registration with a laser scanner. Second, modeling from the point cloud, or updating the existing model. Third, validation: the model is overlaid on the cloud to confirm the deviation sits within the tolerance your BEP defined. Fourth, delivery of model, drawings and report.

A small project can be delivered in about a week; a complex industrial building usually takes three to five weeks, and the exact schedule goes in the proposal. You get native, editable files, RVT if you work in Revit or PLA if you work in Archicad, drawings generated from the model and, if you need them, free viewers so your team can open the model without a license. The files are yours and open in the software you already use.

Don’t start by scanning everything at maximum detail. Taking every area to LOD 500 from day one is the slowest, most expensive route, and it tends to end in a model nobody uses. Start where not knowing costs you the most: the area with the most undocumented changes, the interface between two disciplines that already caused trouble, or the space the owner will review first.

A simple test: if someone in your last coordination meeting said “we’ll verify that in the field” and that item is still open, your BEP has a clause with no data behind it. That’s where to start.

Frequently asked questions

Who writes the BIM Execution Plan?

The team delivering the project information writes it, usually led by the BIM manager or BIM coordinator, based on the owner’s information requirements.

Is a BEP mandatory?

It depends on the contract. If the owner requires ISO 19650 or an NBIMS-US based process, the BEP is part of it; on private projects without that requirement, it’s good practice, not a legal obligation.

How often should a BEP be updated?

At every phase change and whenever scope, team or starting data change. A BEP nobody has touched since the bid probably no longer describes the project.

Tell us which clause in your BEP still depends on someone “verifying it in the field,” and we’ll walk through how to back it with measured data. Request a quote.