Most Engineering, Procurement, and Construction (EPC) projects still cling to the ritual of 30%, 60%, 90% model reviews as design stage-gates. In theory, these checkpoints ensure the model is “mature enough” to proceed (e.g. start foundations at 30%, release steel and piping at 60%). In practice, they’ve become bloated, ill-timed meetings that invite late feedback and massive rework. By the 90% review, teams are often drowning in last-minute design changes and finger-pointing, trying to fix issues that should have been caught months earlier. A 60% review, meant to confirm design readiness, instead turns into a brainstorming free-for-all when designs should already be settled; the 90% review becomes a desperate quality catch-up with a flurry of comments that send engineers back to the drawing board. We’ve all seen it: dozens of people on a call, poring over an overloaded 3D model, discovering clashes or missing details that trigger a cycle of rework – delaying the project and inflating costs. This stage-gate process doesn’t ensure quality; it practically ensures delays, scope creep, and misalignment.
Worse, the 30/60/90 model is inflexible and out of sync with real project progress. Tying reviews to arbitrary “% complete” dates means teams often review incomplete or outdated design information. Some parts of the model at “60%” are still half-baked, leading to decisions based on guesswork; other parts are over-developed by 90%, so late changes there force expensive rework of finished work. For instance, reviewing a heat exchanger’s layout at 30% makes no sense if critical calculations (line sizing, bundle removal clearances) aren’t done yet – it gives a false sense of progress and almost guarantees future design changes. Meanwhile, entire disciplines waste time sitting through these broad reviews even if only a few systems are being discussed. Huge multi-discipline meetings require weeks of preparation (printing drawings, assembling comments) and grind other work to a halt, only to review a model that isn’t truly ready. It’s a habit the industry holds onto (“it’s just how we do things”) despite the obvious waste. In short, the 30/60/90 model review process is broken – a comfortable old ritual that no longer serves us in today’s fast-paced, data-rich projects.
A lean, LOD-based model review process aligned with execution
It’s time to ditch the percent-complete review rulebook. Leading EPC teams are adopting a more agile, execution-driven method – essentially a leaner approach that uses frequent Level of Development (LOD)–based “sprint” reviews instead of big 30/60/90 milestones. LOD (Level of Development) is a BIM standard that defines how detailed and reliable a model element is at a given stage. In simple terms, it’s a measure of model maturity: for example, LOD 300 means a component’s geometry and attributes are defined well enough for coordination and decision-making. Using LOD as our guide, we review each part of the model just-in-time when it reaches the appropriate maturity, rather than waiting for the next arbitrary calendar milestone. As soon as a set of components (say, all equipment in Area A) hits the target LOD – for instance, detailed design complete at LOD 300 – it’s flagged as ready for review. The project team then conducts a short, focused review session (often weekly or bi-weekly) to validate those components across all disciplines immediately, instead of deferring to a large batch review weeks or months later. In that session, stakeholders confirm all critical details are correct (location, dimensions, connections, clearances, etc.) for the components in question. Once everyone agrees a component or system is good, its status is advanced to a higher LOD (for example, Buro Matei uses LOD 320 to indicate an element has been reviewed and approved) and the geometry is frozen, with no further changes allowed except minor tweaks like vendor data updates. In other words, each element is reviewed once at the right moment and then locked down, instead of being revisited and reworked at 30%, 60%, 90% checkpoints. This “one-and-done” approach eliminates the churn of repeated reviews and late changes.
Crucially, these LOD-based reviews are planned to follow the logical sequence of execution rather than a percentage checklist. The model reviews are sequenced in the same order the facility will be constructed, aligning with the Work Breakdown Structure and Path of Construction. For example, civil and structural elements (foundations, steel frameworks) are finalized first, then equipment is placed and approved, then piping is routed to those fixed equipment locations, then electrical and instrumentation details are added, and so on. You never review a dependent system before its prerequisite is ready – no reviewing pipe runs until the pumps and vessels they connect to are fixed in place, no finalizing equipment until the supporting steel and foundations are set. This careful sequencing ensures each discipline’s design is reviewed at the optimal time with the right context, which dramatically reduces downstream changes. In effect, the model review process mirrors the construction plan: we’re verifying each chunk of the design in the same order it will be built, so that by the time construction crews get to work, there are no surprises left.
This LOD-driven, execution-first approach yields huge benefits over the old method. Reviews happen in small, frequent batches rather than giant infrequent meetings, which means issues are identified and resolved continuously throughout design, not stockpiled until late. Because each object or area is only reviewed when it’s truly ready and then locked, we avoid the wasted effort of revisiting the same element multiple times – leading to far less rework and iteration overall. By tying every review to actual design maturity (LOD) instead of an arbitrary date, we only review complete, up-to-date information, avoiding the trap of making decisions on incomplete or outdated data. The review sequence is aligned with execution logic, often matching the project’s Work Packages and construction work areas, which means design prerequisites are confirmed in the right order and late design changes are dramatically reduced. Equally important, this sprint-style process fosters continuous collaboration and alignment among all parties. Stakeholders from engineering, construction, operations, and the client stay engaged from the get-go, rather than tuning in only at 90% with a laundry list of complaints. Issues get raised and fixed in real time, and constructability, operability, and safety considerations are addressed iteratively during design – not as afterthoughts. Essentially, we restore model reviews to their original purpose: a quality assurance step to confirm a design that the entire team has already agreed on, instead of a late-stage design by committee. The outcome is more predictable, reliable project delivery: fewer surprises, fewer RFIs during construction, and an EPC project that progresses smoothly from design into execution.
To summarize the advantages of moving away from rigid 30/60/90 checkpoints toward a continuous LOD-based review model:
- Reviews triggered by real readiness: Using LOD thresholds as gates means the team reviews each scope only when the design is actually complete and verified, not at some arbitrary date. This ensures feedback is timely and applicable to the final design, rather than too early or too late to be useful.
- Minimal rework through “first-time-right”: Frequent, focused sprint reviews catch issues when they’re small and easy to fix. Once a model element is reviewed and approved, it’s closed out and won’t be revisited repeatedly. This one-and-done mentality results in far fewer design revisions and rework cycles, speeding up the project by eliminating the back-and-forth churn of the old process.
- Execution-driven sequencing: The order of reviews follows the project’s construction and work package sequence, not discipline silos. Foundational elements are finalized before dependent details, so each decision is made with the correct prerequisites in place. This logical sequencing (often aligned to the WBS and AWP principles) means far fewer late design changes, because you’re not approving something only to have its context change later. It ties engineering progress directly to how the project will be built – supporting Advanced Work Packaging by ensuring design deliverables are completed in the optimal order for execution.
- Continuous team alignment (no surprises): Sprint-style LOD reviews keep everyone – engineering, construction, O&M, safety, and client – involved throughout the design. Because information is shared in real time (often via a live dashboard) and reviews are routine, all stakeholders see the same single source of truth and can contribute feedback at the right time. This eliminates the classic scenario of a client or operator only seeing the model at 90% and yelling “this doesn’t work for us.” Instead, they’ve been part of the process all along. Constructability and operability input is incorporated continuously, leading to a design that everyone has essentially approved by the final checkpoint. The model review stops being a big “ta-da!” moment and becomes a transparent, collaborative progress check, resulting in full buy-in and no last-minute fireworks.
In short, moving to an LOD-based, execution-driven model review approach applies Lean principles of eliminating waste and batch delays. We conduct model reviews just-in-time when each part is ready, we remove the waste of redundant reviews and waiting, and we engage people continuously to build quality in from the start. Now the question is: how can an organization actually make this shift? Below is a practical roadmap, in ten steps, to replace the old 30/60/90 review process with this leaner LOD-driven system. These steps cover everything from execution sequencing and work packaging to team culture and client buy-in – all the pieces you need to get this right.
Roadmap: 10 steps to shift from 30/60/90 to an LOD-driven review process
- Throw out the 30/60/90 rulebook – commit to LOD gates: First, acknowledge that the old percent-based model isn’t working and secure leadership buy-in to change it. Stop scheduling design reviews by “% complete” on the calendar. Instead, redefine your model review gates in terms of LOD milestones – in other words, review when the design is ready, not when the calendar hits 60%. For example, decide that a model review will occur when a set of systems (say all pipes in Unit 1) reach LOD 300 (detailed design complete), rather than when the overall model is “60% done.” This may require some tough conversations with stakeholders who are creatures of habit (and with contracts that mention 30/60/90), but be unapologetic. Show them how those arbitrary percentage checkpoints lead to reviewing immature designs and outdated data, causing the very delays and overruns everyone hates. Get your project team and client on board that “we will review when it’s ready, not when a date hits.” This top-level commitment is crucial – it empowers the team to break from the old ritual and focus on quality over tradition.
- Define clear LOD targets and criteria: Next, get specific about what “ready for review” actually means for each discipline and stage. Establish a clear LOD definitions matrix or checklist so everyone knows exactly what constitutes, for example, LOD 300 for structural steel, for piping, for instrumentation, etc.. If you don’t already have a BIM Execution Plan that defines LOD levels, create one. Make it concrete: spell out that LOD 300 means a component is fully modeled with accurate geometry and dimensions, all interfaces resolved, metadata (specs, tags) populated, and any required calculations completed to validate the design. Whatever criteria you choose, document them and ensure the whole team shares this understanding. This avoids the classic disconnect where engineering thinks something is “done” but construction later discovers missing information. If needed, introduce intermediate LOD statuses to fit your workflow. (For instance, Buro Matei uses LOD 320 to denote an element that has been reviewed and approved – essentially frozen.) The goal is a shared language of model maturity: everyone from designers to project controls to the client should know exactly what it means when you say “this item is LOD 300, ready for approval”. With clear LOD targets, you can confidently declare something review-ready – and just as importantly, push back on reviewing anything that isn’t. Nothing gets a formal review unless it meets the agreed definition of done.
- Align the model with work packages (WBS/AWP): A major flaw of percentage-based reviews is how they slice the project by arbitrary progress rather than by how the work will actually be executed. Fix that by structuring your model development and review scope around the Work Breakdown Structure (WBS) and Advanced Work Packaging (AWP). Ensure the 3D model is organized in a way that corresponds to how the project will be built – by systems, areas, modules, or other logical work packages – instead of by discipline silos. For example, break the model (and the review schedule) into geographic areas or modules aligned with the Path of Construction. If Area A will be built first, make “Area A” a distinct model package to be completed and reviewed before Areas B or C. A well-structured WBS essentially maps out the sequence of construction, allowing you to group model reviews by that execution sequence. This way, each review covers a coherent scope that will be constructed together, rather than a random 30% slice of everything. Apply AWP principles: define engineering, procurement, and construction deliverables for each work package early and tag all model components, drawings, and data with their respective work package or area codes. Then when it’s time to review “Area A – Package 1,” you can easily isolate that part of the model and ensure everything needed for it (models, drawings, BOMs) is ready. This prevents the fiasco of construction starting an area only to find a few hold-ups because some drawings or materials weren’t ready. With WBS/AWP alignment, each model review becomes a self-contained package with all disciplines’ content in place for that portion of the project. In short, by aligning model development to the WBS from day one, you’re making model reviews drive execution readiness (construction-driven packages) instead of just checking a design percentage. The model evolves in tandem with how the project will be built, which hugely improves sequencing and handoffs.
- Plan frequent “sprint” reviews and set a cadence: With the model now structured for flow, establish the rhythm of your sprint-style reviews. Decide on a regular interval (e.g. weekly or bi-weekly on a set day) and stick to that drumbeat. Instead of two or three giant review meetings per project, you’ll have many small ones – say, a 2-hour review every Thursday morning – to continuously chip away at the design. Frequent, time-boxed reviews prevent backlog from piling up and keep momentum high. Publish a rolling model review calendar for the team. Notably, this calendar won’t list fixed scope like “60% model review” on a certain date. Instead, entries might read like “Sprint Review 5 – focus on Equipment in Area A and Pipe Rack X (now at LOD 300)”. In other words, the agenda flows with the design progress. At the end of each review, decide what looks likely to reach review-ready status by the next one, and pencil that in as the tentative focus. This is a very different mindset from waiting three months for a big review event. Now every single week (or interval), some progress must be validated and roadblocks will reveal themselves early. To make it work, discipline leads need to be continuously preparing their part of the model for the next sprint – no more procrastinating until a big deadline looms. The project manager or BIM coordinator in effect becomes a Scrum Master, coordinating between disciplines and checking that ready items are lining up for the upcoming session. While more frequent meetings may sound like more work, in reality they replace the huge, inefficient meetings with shorter, efficient ones. They create a steady drumbeat and urgency: for example, if the piping team knows the next sprint review is next Wednesday, they’ll push to get those ten lines to LOD 300 by then, rather than letting work drift until a far-off 90% review. Keep the cadence tight – it curbs the accumulation of “design debt” and prevents big surprises. The model will march toward completion in a smooth, continuous flow instead of lurching in last-minute leaps.
- Review in logical construction sequence: Ensure each sprint review’s content is organized by how the plant will actually be built, not by arbitrary discipline order. Within a given review session, tackle topics in the execution order that makes sense for construction. For example, start with civil and structural foundations, then structural steel, then equipment, then piping, then electrical/instrumentation – layering in design elements in the same sequence you’d build them. Always review prerequisites first: you do not review pipe routing until the equipment locations are approved and frozen; you don’t finalize equipment arrangement until the steel platforms and elevations it sits on are set; and so on. This sounds obvious, but traditional 60% reviews often violate this logic by trying to cover “everything” at once, leading to disjointed discussions that jump around and leave dependencies unresolved. In the new approach, if something isn’t ready in its proper order, you discipline yourself to defer it. For instance, if piping for Area A isn’t ready this week because the equipment is still being adjusted, then review the equipment now and push the piping review to the next sprint after the equipment is frozen. It takes restraint to say “we’ll wait on that,” but the payoff is zero wasted effort reviewing a model section that will just change anyway. As part of this sequence-driven review, involve construction management to perform quick constructability sanity checks on the fly. Within each review, when a design is about to be frozen, have a construction manager or experienced superintendent give a thumbs-up: “If we freeze this, can we build it efficiently? All crane access, lifting, and maintenance clearances okay?”. By baking these constructability checks into each sequence review, you effectively conduct a rolling constructability review throughout design, instead of a giant separate workshop at 90%. Following an execution-driven sequence for model reviews means each element is reviewed at the right time and in the right context, drastically minimizing late rework and ensuring the design is truly construction-ready.
- Engage all stakeholders early (no surprises): This step is about mindset and roles – turning the team from siloed departments into one integrated group. Require participation from all key stakeholders in these frequent reviews, not just the engineering designers. That means construction, operations & maintenance (O&M), safety (HSE), the client’s project representatives – everyone with a stake in the design should be involved continuously, not only at the tail end. In practice, invite the construction lead to any review touching constructability, have the O&M rep join sessions on equipment layout and access, bring in HSE when safety-related layout issues are on deck, etc.. Their job in each review is to validate that their requirements are being met while changes are still easy to make. For example, in a sprint review focusing on equipment arrangement, the operations team can immediately confirm that the model provides adequate access for maintenance (valve clearances, manways, etc.), the HSE team can verify that safety features like escape routes or lifting lugs are accounted for, and construction can ensure the equipment orientations allow efficient installation with no conflicts for cranes or scaffolding. By getting this feedback at LOD 300 for each system, you catch issues at the optimal moment – not during a late field walkdown when steel has already been fabricated. Early stakeholder engagement also fosters a sense of ownership and prevents the “big surprise” at final handoff. The client, for instance, won’t vanish after a token 30% review and then show up at 90% with a laundry list of changes. Instead, the client’s engineering manager can attend the periodic reviews relevant to them, staying informed and providing input when it’s still relatively simple to adjust the model. This eliminates the big “ta-da!” unveiling – no more shocking the operations team with a design they can’t accept at the 11th hour. All stakeholders should feel they’ve been part of the journey and signed off along the way. It also creates accountability: if O&M and others signed off on the pump arrangement at LOD 300, they can’t later claim at 90% that they “never liked that pump there”. They were part of that decision. In short, engaging stakeholders continuously ensures full alignment and no late-stage surprises, restoring model reviews to a true check that everyone has already agreed the design is good.
- Foster a “zero rework” culture: Changing the process won’t stick without changing the team’s mindset. Make “first-time-right” design a core value. Ingrained habits like holding feedback for formal reviews or keeping quiet to avoid conflict must be broken. Encourage and expect continuous, proactive feedback from team members at all times – not just during review meetings. This requires building psychological safety: everyone needs to know that raising a concern early is far more valued than silently letting an issue fester. Emphasize that our goal is zero significant rework – catch problems when they cost hours or days to fix, not when they cost millions during construction. One practical tactic is to introduce new KPIs for design quality, such as “No major design changes after approval” or “No scope gaps between disciplines,” and track them. Monitor how many design changes or RFIs arise after something was supposedly finalized. Celebrate when a sprint review yields zero comments because the preparation was that thorough. Conversely, if a certain issue had to be revisited three times, do a quick retrospective on why it was missed initially. The team needs to internalize that rework is a failure in the system, not an inevitable part of the job. Bust the myth that catching issues late is normal – it’s not; it’s waste that we’re determined to eliminate. Leaders should set the tone with frank, direct language (much like this article) to call out old habits. Don’t be afraid to say in meetings: “We will not wait until 90% to fix something we see at 30%. If you spot a problem or have an idea, bring it up now, not later.” Make it clear that speaking up early is rewarded, whereas keeping polite silence and causing late surprises is the real transgression. This candor might ruffle some of the old guard who are used to nice, non-confrontational reviews where issues get papered over – so be it. We’d rather have a bit of awkward honesty in week 4 than a project disaster in week 40. When the whole team shares this “zero rework” mindset, the model review process truly becomes about preventing issues proactively instead of managing rework after the fact.
- Leverage data and live model dashboards: To coordinate this process across disciplines and locations, you’ll need real-time visibility into design maturity. Say goodbye to clunky spreadsheet trackers and email chains. Implement a centralized digital dashboard or model management system to track each component’s LOD status, review progress, and upcoming readiness. Modern EPC execution platforms (for example, Buro Matei’s Brumata) provide structured ways to do this – using a central database and LOD-driven workflows with live dashboards that show exactly which model components are ready for review, which are in progress, and which are frozen. The idea is to establish a single source of truth for model maturity that every team member (engineering, construction, client, etc.) can see in real time. Imagine opening a dashboard where the 3D model is color-coded by LOD status: green for “approved & frozen” (say LOD 320), yellow for “in progress” (LOD 200–300), red for “issue or behind” (needs attention). You could instantly see that all pumps in Area A are green (done), the pipe racks are yellow (still being designed), and perhaps the cable trays are red (awaiting vendor data). This level of visibility focuses the team on what needs attention and prevents anything from slipping through the cracks. Data-driven tracking also enables automation – no more manually compiling model status reports for the project manager each week. The data is live and accessible to all. For example, when an engineer marks a pump as “LOD 300 complete,” the system updates immediately and can even notify procurement and construction that this item is ready for their actions (like starting fabrication or planning installation). Integration is key: tie the model’s LOD status to procurement and scheduling systems. If a piece of equipment is frozen at LOD 320, procurement can proceed to confirm the vendor drawings and construction can schedule its installation, confident that no further design changes are coming. Off-the-shelf solutions exist, or you can build custom dashboards with business intelligence tools – the point is to use technology to track and broadcast model readiness. This enforces accountability (everyone can see if a discipline is lagging on getting their area to LOD 300) and enhances collaboration (no more “I didn’t know you were waiting on me” excuses). In the modern EPC world, real-time dashboards replace static progress spreadsheets, allowing the whole team to rally around facts and adapt quickly. Data should drive decisions – including what to review next – rather than gut feel or outdated information. (On a side note, platforms like Brumata were developed specifically to support this kind of LOD-based model maturity tracking and dashboard visualization. Such tools provide an integrated, data-driven environment to manage model reviews and ensure everyone sees the same status in one place.)
- Align contracts and stakeholders to the new process: Even the best process will falter if your contract or client expectations are locked into the old way. Make sure all stakeholders understand and support this LOD-driven approach. Educate the client/owner on why you’re deviating from the familiar 30/60/90 routine – emphasize that you’re not skipping reviews, you’re making reviews continuous and more effective. It might even help to adjust contract language or project procedures that explicitly call for 30%, 60%, 90% reviews. For instance, if the contract says “Contractor shall conduct a 90% model review,” you can still comply – but that 90% meeting will simply be a high-level confirmation because all the detailed reviews have been happening along the way. (In practice, the 90% meeting becomes a formality to satisfy the checkbox, with no surprises left to discuss.) Many clients will actually welcome this change when they realize it means fewer marathon meetings and a more controlled design process. The key is to get client alignment early. Show them your proposed roadmap and maybe even pilot the approach on a small scope to prove its value before full rollout. Internally, make sure all discipline leads, project controls, and management are on the same page that LOD sprint reviews are the new normal – update your Project Execution Plan to formally describe the process. If there are external stakeholders like third-party reviewers or certifying authorities, brief them as well so they aren’t caught off guard. Essentially, make the new process official and transparent so nobody is confused about when or how model reviews will happen. It can be persuasive to share data from industry studies or past successes: for example, aligned teams (where engineering and construction were in step early) have delivered projects significantly under budget on average, whereas poorly aligned teams suffered major overruns. (One CII study showed well-aligned projects came in ~6.5% under budget, highlighting how early alignment pays off.) Stress that this isn’t some risky experiment – it’s an emerging best practice in the industry, designed to eliminate waste and improve predictability. Finally, clearly define stakeholder roles under the new regime: e.g. the client will be invited to a monthly model walkthrough, the operations team must formally sign off each system at LOD 300, the construction manager will co-lead constructability checks each sprint, etc. When everyone knows their responsibilities and expectations in the LOD review process, you remove the misalignment and confusion that often plague projects. The end goal is for all parties to act as one integrated team focused on delivering a successful facility, rather than each protecting their own silo and milestones. Make that alignment explicit (even contractual if needed), and you’ll create a united front that embraces this change.
- Monitor, measure, and refine the process: As you implement LOD-based model reviews, treat it like any other critical process – continuously monitor its performance and adapt as necessary. Define and track metrics that matter for design and execution outcomes. For example, measure how many RFIs or design change notices arise during construction due to design issues – this number should drop dramatically if your early reviews are working. Track whether you’re seeing fewer late surprises and field clashes. Is the engineering phase duration improving compared to past projects? Also keep an eye on the health of the review process itself: Are the sprint reviews happening on schedule or starting to slip? Is the team able to keep up with the cadence, or are people feeling overloaded? Use your dashboards not just to monitor design status, but to spot bottlenecks or pain points in the review workflow too. For instance, if a certain discipline is consistently not getting items to LOD readiness in time for the planned review, find out why – maybe the LOD criteria need clarification, or that team needs more resources, or perhaps the sequence needs adjustment. Conduct periodic retrospectives with the team to gather feedback: what’s working, what isn’t, and how can we improve the next cycle? The goal is continuous improvement of the process. Just as we use Lean thinking to eliminate waste in design, use it to fine-tune the review process itself. By measuring outcomes (like rework reductions, schedule adherence, fewer RFIs) and adjusting accordingly, you’ll ensure the new model review system keeps delivering superior results over the old method. In short, never “set and forget” – keep learning and improving. This adaptive mindset will cement the benefits and help overcome any remaining skepticism as people see measurable gains.
Following this 10-step roadmap will require determination and cultural change, but the payoff is enormous. You’ll transform model reviews from a necessary evil into a competitive advantage – a data-driven, collaborative process that ensures first-time-right design and a smooth handoff to construction. Projects that have made this shift have seen far less rework, far better team alignment, and no more wasted months waiting on bloated design reviews. It’s time to break with tradition: replace the 30/60/90 review grind with a leaner LOD-based approach that truly supports execution. The experienced EPC professionals who embrace this change will deliver projects faster, cheaper, and with fewer headaches – and won’t ever want to go back.
