Revision Management is now live in LIFT. It automates the manual work of comparing drawing sets, updating takeoffs, and documenting what changed, so fabricators and erectors can respond to rebids and change orders in minutes instead of hours.
Why Revisions Are So Painful
Revisions are part of every steel project. An addendum lands mid-bid. A change order comes through after award. An RFI triggers a drawing update mid-project. Each one forces the same interruption: stop what you're doing, find the revised drawing set, compare it to the one you already worked from, and figure out what actually changed.
On average, it takes around four to five bid cycles to win a job, and that doesn't include post-award change orders. Every one of those cycles ends with the same manual comparison work, under deadline pressure, while the rest of the pipeline keeps moving.
And it's not only estimators feeling it. Project managers and contract administrators live in the same revision cycle after award, when change orders and RFIs keep coming and the quantities need to stay accurate for billing and procurement.
For most teams, this means hours of manual comparison, an incomplete audit trail of what changed, and the constant risk of pricing at the wrong number because something slipped through.
What Revision Management Does
Revision Management automates the full cycle: identifying changes between drawing versions, updating the takeoff, and generating the documentation you need to bid or respond to change orders at the right price.
Upload a revised drawing set and LIFT compares it to your previous version. Every added, removed, or changed member gets flagged. Your previous work carries forward. You review only what's different. See the full Revision Management overview.
The core capabilities
Version comparison. Add a new version to an existing job, upload the revised drawings, and arrange the pages to produce a complete revised set. LIFT handles the version history automatically.
AI-powered change detection. Rather than transferring markups manually or re-analyzing from scratch, LIFT compares the two drawing sets and makes the changes for you. Processing takes roughly one to two minutes per page.
Quantity panel with change status. Every member in the takeoff is marked as added, changed, or deducted. You see the full picture of what moved between versions without hunting for it.
Overlay validation. Turn on the overlay to visually confirm the changes LIFT identified. Hover over a member to see a record of what changed, for example, from W250x33 to W360x33.
Page alignment tools. When pages don't align perfectly between drawing versions, a combination of click-and-drag and keyboard controls lets you fix alignment quickly to ensure accurate comparison.
Delta Report with tonnage summaries. A built-in summary shows net tonnage added, deducted, and changed by member type. On one beta job, this surfaced 150 tons added, 90 tons deducted, and just over 200 tons removed as the net change of moving to lighter members, all directly from the export.
BOM export with change status. Every line item in the exported BOM includes its status (added, changed, or deducted), giving downstream teams a clean record of what moved. The export is built to align with Tekla PowerFab and common internal estimation workflows.
What This Means for Your Team
For estimators, revisions stop eating hours. A revised drawing set comes in, LIFT processes it, and the updated takeoff is ready to review in minutes. No more rebuilding quantities from scratch to catch what changed.
For project managers and contract administrators, post-award change orders and RFIs become tractable. The quantities stay current, the audit trail is automatic, and billing and procurement have the documentation they need without a separate scramble.
For the business, rebids and change orders can be turned around at the right price, with confidence, instead of with whatever could be pulled together under deadline pressure.
Built with Customers, For Customers
Revisions have been the single most-requested feature in LIFT for over a year. Almost every estimator we sit with brings it up unprompted. We heard it enough times to know it was costing customers real money across the entire life of the project.
Several of our customers have been running Revision Management in beta for the past several months on real projects with real deadlines. Their feedback pushed us in directions we wouldn't have gotten to on our own, and shaped what we're shipping today.
The real credit for this release goes to them. To the customers who kept raising their hand, kept telling us where the pain was, and kept giving us feedback on the early builds. That's how good products get made.
Next Steps
Revision Management is available for every LIFT customer, new and existing. Reach out to your CSM or contact us to get access.
We're showing it live this week at NASCC 2026 in Atlanta. Stop by Booth #1127 to learn more.
Book a demo to see Revision Management on your own project drawings.
How AI Integration Transforms Existing Steel Estimating Workflows Without Disrupting Your Team
AI can slot into your existing steel estimating workflow as a fast, accurate takeoff “engine” without forcing you to abandon Tekla, Excel, Strumis, or the review habits your team already trusts. The key is treating AI as another tool in your estimating bench, not a replacement for your processes or your people.
How AI Integration Transforms Existing Steel Estimating Workflows Without Disrupting Your Team
Steel estimators face a tough decision. Stick with familiar manual processes that work but take hours, or risk disrupting everything with new technology that might not integrate with existing tools. The fear is real: what if AI does not work with Tekla, Strumis, or your custom Excel templates; what if it disrupts years of refined workflows; what if the team resists the change.
Across construction, digital takeoff and automation have already cut quantity takeoff time by roughly 80% compared to manual methods, as guides from Buildxact and other construction tools document. In structural steel, fabricators using AI‑assisted takeoff and integrated estimating workflows report 50–80% reductions in takeoff time and substantial gains in bid capacity. They are not replacing their systems; they are enhancing them.
This article explains how AI integrates with your current steel estimating workflows, what stays the same, what improves, and how to implement it without chaos.
The Integration Reality: What Actually Happens When You Add AI
Many estimators picture “AI integration” as throwing out tools like Tekla, Strumis, and Excel and starting from scratch, a fear echoed in AI‑adoption articles in construction and metal industries. That is not how successful fabricators do it.
AI for steel estimating acts as an intelligent layer on top of your existing workflow, similar to how power tools augment a shop without changing the fundamental fabrication steps. In most integrated shops, the following parts of the workflow stay the same:
Tekla Structures, Strumis, PowerFab, and Excel remain core systems.
Bid review, risk assessment, and approval flows stay in place.
Estimators still make pricing, risk, and scope decisions.
Client relationships and communication patterns do not change.
Pricing strategies, markups, and margins remain under your control.
What changes is the amount of manual work required to get to a clean bill of materials, because manual counting, measuring, and data entry can consume roughly 60–80% of an estimator’s working time. AI takeoff tools remove most of that burden so estimators can spend their time on scope, pricing, and strategy, which aligns with BIM and automation studies showing productivity gains when low‑value tasks are automated.
One structural steel fabricator quoted in SketchDeck.ai case material summarized it this way: “We still estimate the same way; we just do not count beams by hand anymore.”
Understanding Your Current Workflow Pain Points
Before you integrate AI, you need a clear view of where manual processes are actually slowing you down. Research on construction estimating shows that small quantity errors and process inefficiencies compound into major cost overruns. One summary of overrun statistics notes that about 85% of projects experience cost overruns and that 32% of overruns can be tied to estimating and scope issues.
Industry data and your own pillar content highlight a few common bottlenecks in steel estimating.
Manual takeoff time
Counting beams, columns, braces, and connections by hand across dozens or hundreds of sheets.
Recording specifications from tiny labels on structural plans, which is difficult when fonts are small or drawings are scanned.
Cross‑referencing piece marks and details across multiple sheets to avoid missed members.
Building material lists and BOMs from scratch, which BOM automation research shows can take days without structured data.
Data entry redundancy
Re‑entering takeoff data into Tekla, Strumis, PowerFab, or custom Excel templates.
Maintaining separate versions of the same BOM for estimating, production, and purchasing without automated synchronization.
Updating quantities manually after drawing revisions, increasing the chance of version mismatch.
Error‑prone steps
Transcription mistakes during manual counting or data entry.
Missed materials in complex or cluttered drawings, especially where multiple disciplines overlap.
Inconsistent methods between estimators, leading to different results on similar jobs.
Version control issues when addenda or revised sets arrive late in the bid cycle.
Time‑intensive reviews
Double‑checking manual counts line by line.
Verifying material specs against schedules and details.
Comparing drawings for changes between revisions, often by eye.
Performing quality control on BOMs before pricing.
These issues tie directly into the cost overrun statistics mentioned above, where poor estimating and scope management contribute to overruns of 15–28% on many projects. AI does not remove risk, but it gives you a faster, more consistent starting point so your team can focus on catching high‑impact issues instead of doing rote tasks.
How AI Fits into Your Existing Tools
The best AI estimating systems do not replace tools like Tekla or Excel, they feed them with cleaner, more complete data.
Integrating with Tekla Structures
Tekla Structures remains the steel detailing standard for many fabricators, and AI takeoff tools such as LIFT can export member and connection data in formats that Tekla can interpret while preserving section sizes, lengths, and attributes. Studies on BIM and model‑based quantity takeoff show that when structured data flows cleanly into tools like Tekla, coordination improves and manual data entry drops.
A typical workflow looks like this:
Upload structural drawings (PDF or DWG) into the AI platform.
AI detects and counts beams, columns, and braces, then reads their sizes with OCR that can reach 98–99% character accuracy on clean, printed text at 300 DPI or above.
Import into Tekla to drive modeling, sequences, or reporting.
Continue detailing as usual.
Academic work on computer vision and BIM for construction documents and progress monitoring shows that combining CV extraction with BIM tools can improve data consistency and reduce manual work when transferring information into design models.
Nearly every estimating team has Excel templates refined over years, which your pillar article recommends as part of a strong estimating system. AI takeoff does not ask you to discard them; it populates them.
Most AI platforms offer:
Direct CSV or Excel export using standard columns.
Configurable column mappings so exported fields line up with your workbook inputs.
Consistent naming so formulas, lookup tables, and macros continue working.
This means your pricing logic, markups, and risk factors remain unchanged, but instead of spending hours keying numbers into spreadsheets, your estimators can focus on analyzing those numbers against historical performance and market conditions, in line with cost‑engineering best practices.
Your pillar article already recommends building and maintaining a historical estimating database that tracks estimated vs. actual performance by project type and tonnage. AI takeoff strengthens that advice by making it faster to populate that database with clean, comparable data for every job, similar to how BIM‑driven quantity takeoff centralizes data.
Connecting to Strumis and Other MIS/ERP Systems
Manufacturing Information Systems (MIS) like Strumis need accurate, structured data to drive purchasing, nesting, and production planning, and BOM automation research shows that when ERP systems receive high‑quality data, material waste and rework decline. AI integration focuses on delivering exactly this.
A typical integrated flow:
AI performs the takeoff and builds a BOM with sections, lengths, weights, and location tags.
The BOM exports in a format that Strumis or an ERP can read, often CSV or a dedicated import format.
Your MIS uses this data for stock checks, purchasing, nesting, and production scheduling.
08:15 – Upload the combined PDF to the AI platform.
08:45 – AI completes first‑pass takeoff and BOM for clean structural sets, leveraging computer vision and OCR.
09:00–10:00 – Estimator verifies flagged items, spot‑checks key grids, and compares total tonnage to benchmarks from similar projects.
Same day – Pricing starts, and the bid can often be submitted before the end of the day.
The steps are the same receive, understand, take off, price, review, submit but where your time goes is different. Instead of spending 6–8 hours on counting and only a couple of hours on strategy, estimators can flip that ratio, consistent with time‑savings reported for digital takeoff and AI workflows.
Handling Drawing Revisions
Drawing changes used to mean rework and risk of missing updated elements. Computer vision and document comparison tools make revisions more manageable by highlighting differences between versions, which is a documented use case in AI‑driven document analysis.
With AI integrated:
Upload revised drawings into the same project in the AI platform.
The system identifies new, changed, or removed members by comparing geometry and labels.
You review a change list instead of redoing the entire takeoff.
Only affected quantities in Tekla, Excel, or Strumis are updated; unaffected items stay as they are.
Large or fast‑track projects often require more than one estimator. AI integration can improve collaboration rather than complicating it, especially when the AI platform serves as a shared source of truth.
Common patterns include:
A central AI project where all extracted members and quantities live.
Multiple estimators reviewing different areas or floors but all working from the same AI‑generated baseline.
Shared filters and views so everyone sees the same version of the BOM and flags.
One consolidated export to Excel, Tekla, or Strumis instead of several inconsistent spreadsheets.
Construction collaboration research shows that shared, structured data reduces rework and miscommunication between teams, which directly supports better estimating outcomes.
The Human Element: Getting Your Team On Board
Common Concerns You Should Expect
“Will AI replace my job?” Evidence from AI adoption in construction and manufacturing shows that AI tends to shift task mix rather than eliminate roles, moving people from repetitive work into oversight, analysis, and client interaction. Estimators still own scope interpretation, risk decisions, connection complexity, and pricing; AI handles counting and data entry.
“Can I trust automated counts?” Benchmark studies show that even strong computer vision models still make errors, especially on complex symbols, which is why human verification remains essential. AI quality‑control guides suggest treating AI output as a first draft, using confidence scores and discrepancy flags to drive verification rather than redoing everything manually.
“Our workflow is unique; can AI adapt?” While every shop has its own quirks, most steel estimating workflows share core steps, and flexible AI platforms support configurable exports, naming conventions, and templates so outputs line up with your standards. Training‑data research also shows that domain‑specific models trained on structural drawings perform better than generic OCR or image tools.
“Will the learning curve slow us down?” Studies of digital takeoff adoption show that estimators reach higher productivity within weeks, not months, when tools are designed around their tasks rather than programming workflows. A short pilot with parallel runs is usually enough to build confidence.
Run AI takeoff and manual takeoff side by side on a few test projects.
Compare results to build confidence and identify any recurring differences.
Week 2: Gradual handoff
Use AI for straightforward projects and keep manual workflows for complex or high‑risk jobs.
Start using AI output as the primary source for Excel templates and Tekla imports, with manual checks.
Weeks 3–4: Full integration for standard work
AI becomes the default takeoff method for common project types; manual work remains for special cases.
Team develops new “check‑first” habits and refines verification patterns.
Month 2 and beyond: Optimization
Measure time saved, error rates, and bid volume changes.
Adjust workflows and templates to remove remaining friction points and take advantage of advanced features.
Case studies from SketchDeck.ai customers like Ennis Steel and King Steel show that this phased approach leads to faster adoption, higher estimator satisfaction, and measurable gains in bid capacity without team burnout.
Measuring Integration Success: Metrics That Matter
Time Metrics
Takeoff completion time per project
Baseline: Hours from receiving drawings to complete BOM before AI, often 4–8 hours for a mid‑size project.
Target: 50–80% reduction for standard projects, matching digital takeoff and AI case studies.
Bid turnaround time
Baseline: Days from RFQ to bid submission, often 3–5 days for complex packages.
Target: 30–50% faster on average so you can respond in 1–3 days on many jobs.
Bids per estimator per week
Baseline: Existing throughput.
Target: 40–60% increase, similar to the ~40% bid‑capacity improvements reported by some AI‑adopting fabricators.
Quality and Risk Metrics
Takeoff error rate
Baseline: Historical discrepancies between estimated and actual quantities on completed projects.
Target: 95–99% piece and size accuracy on clean structural drawings, consistent with digital takeoff benchmarks and computer vision performance on structured documents.
Change orders linked to missing or miscounted steel
Baseline: Frequency and value of such change orders.
Target: 50% reduction in takeoff‑related changes, in line with BOM automation studies that show sharp drops in downstream corrections when data is structured and validated.
Bid win rate
Baseline: Percentage of bids won today.
Target: Maintain or improve win rate while handling more bids; faster and more accurate estimates often help here.
Business and People Metrics
Revenue from bids submitted
Track whether increased bid capacity and faster turnaround lead to more awarded work, as AI‑adoption reports in construction and metals suggest.
Overtime hours and estimator satisfaction
Reduced manual drudgery and weekend work are strong indicators that integration is helping; retention research in construction notes that better tools and fewer late nights improve job satisfaction.
Your existing guide explains how even a 5–10% miss on quantities or unit rates can erase a 10–15% target margin. AI should help push you toward tighter, more consistent estimating, not just faster output.
Common Integration Challenges and Practical Fixes
Inconsistent drawing quality
Ask clients for native PDFs or higher‑DPI scans (300–600 DPI). Guides on scanning architectural drawings and document resolution show how DPI affects clarity and OCR accuracy.
Use AI platforms with pre‑processing (deskew, denoise, contrast), as shown in construction OCR and document analysis implementations.
Keep manual workflows for truly unreadable sets.
Resistance from senior estimators
Involve them in vendor selection and pilot design.
Let them run the side‑by‑side comparisons and sign off on accuracy.
Emphasize that AI amplifies their judgment rather than replacing it, consistent with patterns seen in other industries adopting AI.
Integrating with legacy systems
Use CSV or Excel as a universal bridge format, a common approach in BOM automation.
Build simple import macros or scripts once, then reuse them.
Choose vendors that support flexible exports instead of rigid, proprietary formats.
Consistency across teams
Define standard operating procedures (SOPs) for AI use, including verification steps.
Standardize naming conventions, templates, and checklists.
Appoint an AI “champion” who supports the team and coordinates feedback, as recommended in AI continuous improvement literature.
Is Your Workflow Ready for AI Integration?
You are likely ready if:
Manual takeoffs consume 20+ hours a week per estimator.
You are turning down bids due to limited capacity.
Accuracy issues and surprises are causing margin erosion.
Competitors are delivering bids faster or winning work on responsiveness, something AI‑trend reports emphasize for early adopters.
Many shops are turning down bid opportunities because the estimating team is at capacity. At the same time, competitors with optimized workflows and carefully chosen automation are bidding three to five times more projects with the same headcount. The difference is not effort. It is the way work flows, where it stalls, and how much time gets lost to rework, waiting, and context switching.
This article provides a complete framework for building a high‑performance estimating workflow. It draws on lean construction principles, Theory of Constraints research, and real results from steel fabricators that have systematically eliminated bottlenecks and scaled their bid volume.
What Makes a Workflow High‑Performance?
A high‑performance estimating workflow delivers three outcomes at the same time:
Speed: reduced cycle time from RFP to bid
Accuracy: consistent quality across estimators and projects
Scalability: capacity grows faster than headcount through process improvement
Construction productivity research emphasizes that workflow optimization should focus on removing constraints rather than pushing people to work harder. The Theory of Constraints (TOC) says that every process has a single binding bottleneck at any given time, and improving non‑bottleneck activities has no effect on overall throughput.
Stage 2: Takeoff and Quantity Extraction (traditionally 40–50% of Time)
Purpose: produce an accurate, well‑documented set of quantities with minimal rework.
Time‑study work on construction processes and models for estimating duration both show that measurement and takeoff phases dominate preconstruction time on many projects (https://etd.lib.metu.edu.tr/upload/12610696/index.pdf).
Key practices in a manual or semi‑manual environment:
Always follow a consistent drawing review order. For example, framing plans first, then elevations, then details and notes.
Apply a standard material classification and naming convention across all jobs.
Explicitly note moment frames, braced bays, special connections, and non‑typical conditions while counting.
Build simple accuracy checks into the process, such as comparing total floor tonnage against benchmarks for similar buildings.
Capture assumptions and open questions as you work, rather than relying on memory.
Technology leverage at this stage:
This is where automation delivers the most value. Research on automation in lean construction shows that automating repetitive, rule‑based tasks across the project lifecycle, including preconstruction, can significantly reduce non‑value‑added time and improve overall performance (https://www.sciencedirect.com/science/article/pii/S2666165924001005). In steel estimating, AI tools like LIFT automate detection and counting of structural members and many connection features directly from PDF drawings, then output a structured bill of materials in minutes rather than hours or days.
Customer data from SketchDeck.ai illustrates the impact:
Use a fixed self‑review checklist. For example: check that total tonnage looks reasonable, compare major member counts to grid logic, confirm that all scopes on the intake checklist are covered.
Require a second set of eyes on complex projects, whether through peer review or a senior estimator review.
Check that all RFIs and clarifications have either been resolved or clearly listed.
Reserve management sign‑off for margin and go or no‑go decisions.
If takeoff is the bottleneck, you should protect estimator focus during that stage.
Tactics:
Assign clear project ownership so estimators can move a job from intake to submission with minimal handoffs.
Block off two or three uninterrupted hours for takeoff on complex jobs.
Use simple rules for communication, such as responding to non‑urgent questions in defined windows, so that estimators are not pulled out of concentration repeatedly.
Time saved on the bottleneck stage (often takeoff).
Change in bids per month at the same headcount.
Impact on accuracy or rework rates.
Customer stories from LIFT users show reductions of 50 to 95 percent in time spent on takeoff for certain project types and large increases in bid capacity, while maintaining or improving quality:
Cycle time from RFP to bid, segmented by project size and complexity
Bids per estimator per month
Accuracy variance between estimated and actual cost on completed jobs
Percentage of estimates that require major rework
On‑time bid submission rate
Also track time by workflow stage so that you can see where improvements have the most effect. Over time, you should see Stage 2 consuming a much smaller share of the total as you standardize and automate it.
Use Metrics to Guide Action
Metrics should inform decisions, not just fill reports:
Focus improvement efforts on the current bottleneck stage.
Compare performance before and after workflow or technology changes.
Estimate financial return by multiplying time saved by loaded estimator cost.
The fabricators winning more work today tend to share the same pattern. They have:
Clear, standardized workflows from intake through submission
Simple, enforced checklists for critical steps
Minimal redundant data entry, with clean integration between tools
AI and automation focused on the true constraint, not on low‑leverage tasks
A culture where estimators help design and refine process changes
Small, consistent improvements compound over time. The most important next step is simply to begin: map your current process, identify your slowest stage, and improve that one area. Then measure the result, and move to the next.
To see how AI can relieve the Stage 2 bottleneck in your own workflow, consider running one of your recent projects through LIFT in parallel with your current process and comparing time, coverage, and accuracy: https://sketchdeck.ai/demo/
How AI Reads Structural Steel Drawings: The Complete Guide for Modern Estimators
How Human Estimators Read Drawings
When an experienced estimator opens a structural set, they are not just reading lines; they are reconstructing a 3D structure, load paths, and risk profile in their head.
Estimators quickly lock onto high‑value regions such as connections, irregular framing, and load transfer points, rather than scanning every square inch at the same level of detail.
Years of experience turn drawing conventions into instant signals: dashed lines for hidden members, hatch patterns for materials, weld symbols for labor, and dimension strings that define geometry and piece sizes.
This mental model lets human estimators handle vague notes, contradictory dimensions, or incomplete details and still understand the intent of the design. The trade‑off is fatigue: after hours of counting repetitive beams on a big‑box roof plan, attention drops and mistakes slip in.
AI takeoff tools like LIFT do not see beams and columns first; they see pixel grids and patterns that are converted into objects through computer vision models.
A typical 24×36 inch sheet scanned at 300 dpi is about 7,200 × 10,800 pixels, or roughly 78 million data points per page.
Each pixel is represented numerically (for example, 0–255 per channel in 8-bit images), and convolutional neural networks (CNNs) learn to detect edges, shapes, annotations, and symbols across that grid. Stanford's CS231n notes on CNNs offer a good introductory explainer.
CNNs process the image in stages: early layers pick up edges and corners, deeper layers combine these into shapes such as rectangles, angle profiles, or bolt clusters, and higher layers learn to associate these shapes with labeled objects like beams, columns, braces, or callouts.
Object‑detection architectures divide the drawing into regions and assign a confidence score to each detected object, such as “beam, 0.97” or “handwritten note, 0.40.” A high‑level overview of this pattern is in Ujjwal Karn’s “An Intuitive Explanation of Convolutional Neural Networks” (https://ujjwalkarn.me/2016/08/11/intuitive-explanation-convnets/).
LIFT is trained on large volumes of structural steel drawings, which allows it to detect main structural members on most clean digital plans with roughly 95–99% accuracy.
Importantly, the model recognizes patterns but does not understand intent or constructability the way a human does. It can correctly tag a member as W18×40 with high confidence without “knowing” whether that member is part of a moment frame or whether the connection is buildable.
For a broader perspective on computer vision applied to construction drawings and quantity takeoff, see:
When you map AI’s strengths onto the estimating workflow, they line up almost exactly with the tasks human estimators find tedious and error‑prone.
Speed on repetitive counting: On typical clean PDFs, AI takeoff tools can cut the initial counting phase from many hours to minutes, especially for framing plans with regular grids and repeated conditions.
Consistency across pages: Models apply the same detection logic to every page, so the 49th sheet is treated with the same attention as the first, without “Friday afternoon” drift.
Coverage in congested areas: In dense framing zones or overlapped linework, AI systems systematically scan all pixels and often flag members that humans miss when they are visually overloaded.
In SketchDeck.ai’s customer base, fabricators report that what used to take an estimator days to count now takes minutes, which frees that time for connection strategy, pricing, and value engineering. LIFT quantifies main members in minutes, not hours, and is already helping teams reduce estimating time by up to 80%, with more than $25 billion in bids processed through the platform.
Even as AI gets better at reading drawings, human estimators remain critical for interpreting intent, resolving ambiguity, and making commercial decisions.
Low‑quality inputs: On clean CAD exports or high‑resolution PDFs, AI models routinely reach 95–99% identification accuracy for standard members, but performance drops on noisy scans, marked‑up copies, or old drawings with smudges and artifacts.
Notes, exceptions, and implications: AI can read “TYP UNO” or “FIELD VERIFY,” but it does not automatically know which areas are exempt from the typical note or what extra risk “FIELD VERIFY” introduces for your schedule and margin.
Constructability and shop constraints: Evaluating whether an ironworker can get a wrench into a connection, how a detail interacts with your shop’s equipment limits, or how a GC handles change orders is still human work.
The most reliable approach is a hybrid workflow: AI performs the exhaustive takeoff and labeling, and human estimators audit the results, focus on low-confidence detections and complex conditions, and make the final calls on scope, risk, and pricing. For a deeper discussion of realistic expectations, see What AI Can and Cannot Do in Steel Estimating.
A Practical Hybrid Workflow with LIFT
Leading shops are organizing their estimating process so the AI handles the heavy lifting and estimators spend their time on decisions, not counting.
Automated detection and BOM creation. Upload your PDFs to LIFT and let the model scan each sheet, identify beams, columns, bracing, joists, and other structural members, and build an initial bill of materials. For most mid-size structural packages, this automated pass completes in seconds to a few minutes per project, instead of hours of manual counting.
Targeted estimator review. Use LIFT's confidence scores and filtering to focus on low-confidence detections, unusual framing, and key connection details, rather than re-checking every member. Cross-reference AI output with your standard scope review, RFI, and risk-check routines from The Ultimate Guide to Steel Estimating so you maintain consistent quality and protect margins.
Refinement, pricing, and export. Apply your shop's waste factors, labor rates, regional pricing, and margin strategy to the verified quantities, then export from LIFT into tools like Tekla, Strumis, or Excel to complete your detailed estimate and production workflows.
Fabricators using this model often see their estimator time on a typical structural set drop from a full day to roughly 1–1.5 hours, while also increasing the number of bids they can comfortably respond to each week.
To see how this works on your own projects, book a live demo of LIFT. The team will run one of your recent structural sets through the platform so you can compare AI takeoff output against your current manual process.
The Ultimate Guide to Steel Estimating: Best Practices for Fabrication Success
This guide walks you through the complete steel estimating process, from reading blueprints to submitting your final bid. Whether you're a seasoned estimator looking to improve efficiency or a shop owner evaluating your current process, you'll find practical frameworks and industry best practices to strengthen your estimating operation.
Understanding Steel Estimating Fundamentals
What Steel Estimating Really Means
Steel estimating in fabrication is different from general construction estimating. While construction estimators work at a broader project level, steel fabricators need extreme detail at the member and connection level.
You're not just pricing square footage. You're calculating every beam, column, brace, plate, bolt, and weld. You're estimating shop labor for each fabrication step and field labor for erection. You need precision.
Key Components of a Steel Estimate
Every complete steel estimate includes these cost components:
Material costs:
Structural shapes by weight (beams, columns, angles, channels)
Even small estimating errors create big problems. Research from the Association for the Advancement of Cost Engineering shows that a 5-10% miss on quantities or unit rates can push project costs outside acceptable ranges.
Here's what that looks like in practice:
A shop targeting 15% gross margin submits a bid. The estimator underestimates tonnage and labor by 10%. Once the project is underway, the shop discovers the error. After accounting for change orders and rework, realized margins drop to low single digits or even negative territory.
Case studies across structural steel and industrial projects show cost model errors typically range from -1.7% to +7.3%. That variance is enough to erase your intended 10-15% margin completely.
The bottom line: accurate estimating protects your profitability. Inaccurate estimating costs you money, damages customer relationships, and can put your business at risk.
The Steel Takeoff Process
Reading and Interpreting Structural Drawings
Steel takeoff starts with understanding the structural plans. You need to identify every component that will be fabricated and installed:
Beams, columns, and braces
Base plates and cap plates
Stiffeners and gussets
Connection details and bolt patterns
Miscellaneous steel items
Coatings, fireproofing, and surface treatments
Experienced estimators read drawings systematically. They work through each sheet, mark up identified members, and cross-reference details to ensure nothing is missed.
Converting Drawings to Quantities
Once you've identified all members, you convert lengths and counts into weight. This involves:
Measuring member lengths from the drawings using scale
Noting section sizes (W12x26, HSS6x6x1/4, etc.)
Looking up foot-weights in AISC tables or standard references
Calculating total weight for each member type
Counting connections and estimating connection material
Measuring surface areas for coating calculations
You'll use standard section properties to approximate weight. For example, a W12x26 beam that's 20 feet long weighs 520 pounds (26 pounds per foot × 20 feet).
Manual vs. Digital Takeoff Methods
Manual takeoff involves:
Printing full-size or scaled drawings
Using a scale ruler to measure lengths
Highlighting members as you count them
Recording quantities in spreadsheets
Double-checking counts to catch errors
Digital takeoff uses specialized software to:
Work directly from PDF or CAD files
Digitally measure and mark up drawings
Automatically calculate lengths and areas
Generate quantity reports with one click
Reduce manual data entry and arithmetic errors
Time Benchmarks for Steel Takeoff
Manual takeoff for a typical mid-size structural package takes 4-8 hours. This includes:
Reviewing all drawing sheets
Identifying and measuring members
Calculating weights and counts
Organizing data for estimating
Quality checking for missed items
More complex projects with detailed connections, multiple building areas, or extensive miscellaneous steel can take significantly longer.
AI-powered takeoff tools have changed this timeline. Modern automation can reduce takeoff time by 50-75%, allowing estimators to complete the same work in 1-3 hours instead of a full day.
Common Takeoff Errors and How to Avoid Them
Even experienced estimators make mistakes. The most common errors include:
Missed members or details
Solution: Work systematically through every drawing sheet. Use a checklist to verify coverage of all member types.
Incorrect section sizes
Solution: Double-check sizes against the structural schedule. When in doubt, confirm with the engineer.
Misread elevations or dimensions
Solution: Pay careful attention to scale, units, and datum references. Verify questionable dimensions against other views.
Undercounted connections
Solution: Count connection bolts and plates separately from main members. Review typical details to establish connection patterns.
Forgotten coatings or surface treatments
Solution: Note coating callouts early. Calculate surface areas for galvanizing, paint, or fireproofing.
Not rounding to stock lengths
Solution: Remember that material comes in standard lengths. Round up member lengths to account for cutting from stock sizes.
Material Cost Calculation
Understanding Steel Pricing Dynamics
Steel prices fluctuate with market conditions. A good estimator tracks current pricing and maintains strong vendor relationships.
Your material cost calculation starts with total structural steel weight by section type and grade. Then you apply current unit rates from your suppliers.
The Basic Material Cost Formula
Step 1: Calculate weight for each member type
Group similar members (all W12 beams, all HSS columns, etc.)
Total the weight for each group
Separate by steel grade if needed (A36, A992, A500)
Step 2: Apply unit rates
Use current vendor quotes ($/pound or $/ton)
Adjust for shape type (wide flange vs. HSS vs. plate)
Include delivery charges and minimum fees
Step 3: Add cut and loss factors
Apply waste factors of 2-5% depending on complexity
Account for scrap from cutting and shaping
Consider end cuts and drops that can't be reused
Step 4: Include additional materials
Grating, deck, and floor plates
Connection material (bolts, nuts, washers)
Anchors, embeds, and specialty items
Each gets its own unit weight and rate
Managing Price Fluctuations
Steel pricing can change quickly. Protect your estimates by:
Requesting updated vendor quotes for major bids
Building relationships with multiple mills and service centers
Using limited-validity pricing in your quotes (30-60 days)
Including escalation clauses for long-lead projects
Adding conservative contingencies for market volatility
Shops with strong vendor relationships often negotiate better base pricing, reduced surcharges, and priority delivery. This competitive advantage directly improves your margins.
Labor and Shop Time Estimation
Breaking Down Fabrication Steps
Shop labor varies based on member size, connection complexity, and your specific equipment. Most shops break fabrication into these steps:
Cutting and preparation
Saw or torch cutting to length
End prep and beveling
Material handling and staging
Drilling and punching
Hole layout and center punching
CNC drilling or manual drill press work
Countersinking and deburring
Fitting and welding
Assembly and tack welding
Full weld-out (fillet, groove, plug welds)
Weld inspection and repair
Finishing
Grinding and cleanup
Surface preparation (blasting)
Coating application or galvanizing prep
Final inspection and marking
Using Historical Data for Production Rates
Generic productivity tables don't reflect your shop's actual performance. Build your own historical database that tracks:
Hours per ton for different member types
Hours per piece for similar assemblies
Weld time per inch by weld type and size
Setup time for different operation types
This database becomes your most valuable estimating tool. You can quickly sanity-check whether a new estimate aligns with past projects of similar scope and complexity.
Shop Capacity and Scheduling Considerations
Understanding your shop's capacity is critical for accurate labor estimating. Consider:
How many projects can run simultaneously?
Where are the bottlenecks (welding? painting? inspection)?
What's the realistic timeline for this project?
Do we need overtime or additional shifts?
A good estimator factors in current shop loading. If you're already at 90% capacity, adding another large project means overtime costs or schedule delays. Price accordingly.
Field Erection Labor Estimates
Field erection is harder to estimate than shop work. Variables include:
Crane size, type, and hourly rate
Rigging complexity and equipment needs
Site access and staging limitations
Coordination with other trades
Weather delays and seasonal factors
Travel time and per diem costs
Most fabricators estimate erection in crew-hours per ton or per piece, then adjust for site-specific conditions. Experienced erection teams can install 3-5 tons per day for typical commercial buildings, but complex industrial work might drop to 1-2 tons per day.
Technology and Tools in Modern Estimating
The Evolution of Steel Estimating
Steel estimating has evolved dramatically over the past two decades:
Stage 1: Manual takeoff
Scale rulers and highlighters
Hand calculations
Paper quantity sheets
High error rates and slow turnaround
Stage 2: Spreadsheet-based estimating
Digital quantity tracking
Formula-driven calculations
Basic cost databases
Still labor-intensive for takeoff
Stage 3: Specialized estimating software
Digital takeoff from PDF files
Integrated cost databases
3D model visualization
Faster and more accurate
Stage 4: AI-powered automation
Automatic drawing interpretation
Intelligent member detection
One-click quantity generation
50-80% time savings over manual methods
AI-Powered Takeoff and Estimating
Modern AI systems like LIFT have changed what's possible for small and mid-size fabricators. These tools automatically:
Read and interpret structural drawings
Identify members, connections, and details
Extract quantities and dimensions
Generate formatted takeoff reports
Flag potential errors or missing information
The business impact is significant. A single estimator who previously completed two bids per week can now handle four or five. This capacity increase doesn't require hiring—it comes from eliminating manual takeoff bottlenecks.
Real-world metrics from fabricators using AI-powered takeoff show:
50-80% reduction in takeoff time
Ability to bid 2-3x more projects with the same team
Improved accuracy from reduced manual data entry
Faster response times on rush estimates
Integration with Fabrication Management Systems
The most efficient workflows connect estimating tools with detailing and fabrication management systems. When integrated properly:
Quantities flow directly from takeoff to estimating
Mark numbers and assemblies transfer without rekeying
Historical production data feeds back into estimates
Changes update across all systems automatically
Shops using platforms like Tekla or PowerFab can pull standard assemblies, connection details, and actual shop performance rates directly into their estimating process. This reduces both preparation time and costly errors when projects move into production.
Calculating ROI on Estimating Technology
Estimating software requires investment, but the ROI is measurable:
Time savings: If you're spending 20 hours per week on takeoff and can reduce that by 60%, you free up 12 hours weekly. That's 600+ hours per year—equivalent to adding a quarter-time estimator.
Increased capacity: More bids with the same team means higher hit rates on desirable projects and better project selection.
Improved accuracy: Reducing estimate errors by even 2-3% on a $500,000 project saves $10,000-15,000 in margin protection.
Faster turnaround: Responding to bid requests in 1-2 days instead of 3-5 days improves your competitive position with general contractors.
Best Practices and Process Optimization
Building Estimating Templates and Standards
Standardization is your competitive advantage. Create estimating templates for common project types:
Industrial frames template:
Pre-defined line items for typical materials
Standard labor factors for member types
Equipment and overhead allocations
Margin targets for this work type
Low-rise commercial template:
Typical connection details and counts
Standard coating allowances
Field erection crew sizes and rates
Buyout items (stairs, railings, grating)
Miscellaneous metals template:
Unit rates for common items
Installation factors by location (interior vs. exterior)
Typical markup ranges
Templates speed up your estimating and ensure consistency across bids. Your team doesn't need to reinvent the process for every project.
Creating a Historical Database
Track every project's estimated vs. actual performance. Record:
Total hours by project phase (cutting, welding, erection)
Over time, this database becomes predictive. You'll know with confidence that a 200-ton industrial frame with standard connections should require 3,500-4,000 shop hours. Estimates outside this range trigger review before submission.
Peer Review and Quality Control
Implement a review process for major bids:
Have another estimator review scope coverage
Check arithmetic and formula accuracy
Verify assumptions are documented
Ensure all drawing sheets are included
Review for competitive pricing vs. shop capacity
Two sets of eyes catch errors that one person will miss. This quality control step is especially important on high-value projects where margin for error is small.
When to Bid and When to Pass
Not every project deserves your estimating time. Evaluate opportunities based on:
Fit with shop capacity:
Can we execute this on time with current loading?
Does the schedule align with our other commitments?
Relationship quality:
Do we have history with this owner or general contractor?
Do they pay on time and fairly manage change orders?
Drawing quality:
Are the documents complete and clearly detailed?
How many assumptions are we making?
Project type:
Is this work we do well and profitably?
Does it match our equipment and expertise?
Margin potential:
Can we achieve our target margins on this project type?
Is the competitive landscape too tight?
Be selective. Estimating costs money. Focus your effort on projects you want to win and can execute profitably.
Common Challenges and Solutions
Incomplete or Ambiguous Drawings
Many bid packages have missing details, unclear dimensions, or contradictory information between sheets.
The problem: You can't accurately estimate what you can't clearly understand. Making assumptions increases risk.
The solution:
Document all assumptions in your bid clarifications
Submit RFIs (Requests for Information) early in the bid period
Price conservative quantities when details are vague
List specific exclusions in your proposal
Follow up verbally with the architect or engineer when possible
Don't assume you'll work out details later. Protect yourself with clear scope definitions upfront.
Rush Estimates and Time Pressure
General contractors often request quick budget numbers or value engineering options with tight deadlines.
The problem: Rush estimates are prone to errors. They can derail in-progress bids that are more important to your business.
The solution:
Use parametric or conceptual estimating for early budgets
Reserve capacity in your schedule for rush requests
Leverage automation tools to speed up takeoff
Consider charging for feasibility studies and budgets
Learn to say no to requests that don't fit your business goals
Speed matters, but not at the expense of accuracy or better opportunities.
Balancing Speed vs. Accuracy
Estimators face constant pressure to deliver bids faster while maintaining precision.
The problem: Rushing creates errors. Being too slow means missing bid deadlines or losing opportunities.
The solution:
Invest in tools that automate repetitive tasks
Build templates that reduce setup time
Create checklists to prevent missed items
Track time spent on different project types
Set realistic deadlines and communicate them clearly
The goal isn't speed alone—it's consistent, reliable turnaround time with high accuracy.
Managing Change Orders and Scope Creep
Projects evolve. Drawings get revised. Scope expands beyond the original bid.
The problem: Changes after bid award can erode your margins if not properly managed.
Price extras using the same methodology as your original bid
Don't perform extra work without written authorization
Scope changes happen. Protect your margins by documenting and pricing them properly.
Taking Your Steel Estimating to the Next Level
Steel estimating is both art and science. The best estimators combine technical knowledge, historical data, systematic processes, and the right tools to deliver accurate bids efficiently.
Key Principles to Remember
Accuracy protects profitability: Even small errors compound into major margin losses
Speed creates capacity: Faster takeoff means more opportunities to bid
Data drives improvement: Track actual vs. estimated performance religiously
Standardization reduces errors: Templates and checklists prevent missed items
Technology amplifies expertise: AI and automation eliminate bottlenecks
Audit Your Current Process
Take an honest look at your estimating operation:
How long does takeoff take for a typical project?
What's your bid hit rate on desirable work?
How often do estimates miss actual costs by more than 5%?
Are you turning down opportunities because you lack capacity?
What bottlenecks prevent faster turnaround?
Identifying gaps is the first step toward improvement.
Next Steps
Modern steel fabricators are using AI-powered tools to transform their estimating operations. Shops that once spent 6-8 hours on manual takeoff now complete the same work in under 2 hours.
This isn't about replacing estimators—it's about multiplying their capacity. With automated takeoff handling the tedious measurement and calculation work, your estimators can focus on what they do best: analyzing projects, refining pricing, and winning profitable work.
Ready to see how AI can transform your takeoff process?LIFT reduces takeoff time by up to 80%, allowing your team to bid more projects without adding headcount. Request a demo to see how fabricators are using AI to compete more effectively and grow their businesses.
What AI Can and Cannot Do in Steel Estimating: Setting Realistic Expectations
This article lays out, in practical terms, what AI is genuinely good at in steel estimating, where it struggles, and how smart shops are designing hybrid workflows that combine machine speed with human judgment. It sits under The Ultimate Guide to Steel Estimating: Best Practices for Fabrication Success, which covers the full estimating process from drawings to final bid. Think of this as the expectations manual for the AI portion of that workflow.
Where AI Actually Helps Today
Modern AI in steel estimating is very capable in a few specific areas: reading structural drawings, counting elements, and standardizing repetitive tasks.
1. Reading and Counting From Structural Drawings
Well-trained computer vision models can now:
Detect beams, columns, braces, and common connection symbols on typical structural plans
Read nearby size labels (e.g., W12x26) and associate them with members
Convert geometry into lengths, counts, and preliminary weights
When AI tools correctly read sizes, grades, and camber from your drawings, they're effectively helping you build a more reliable input set for applying the design assumptions and detailing practices found in the AISC Steel Construction Manual and related AISC connection standards.
2. Handling Repetitive and High-Volume Takeoff
AI does best when the work is repetitive and pattern-based. Examples:
Large bays of similar beams in warehouses or distribution centers
Long runs of framing in parking structures or multi-bay frames
Standardized connection details repeated across a grid
Here, AI's consistency and stamina matter. It will not lose track of gridlines, miscount a row, or skip a bay out of fatigue. This supports the time savings and capacity lift discussed in the Ultimate Guide when comparing manual vs. AI-assisted takeoff.
3. Standardizing Baseline Quantities Across Estimators
Every estimator has a signature. Some are conservative on tonnage; others are aggressive. AI tools can give shops a common baseline:
The model applies the same detection logic to every job
Different estimators start from the same preliminary quantities
Variance in final pricing comes from strategy, not missed members
This helps owners compare bids and performance more reliably over time, a theme that aligns with the "estimated vs. actual" tracking recommended in the Ultimate Guide.
Where AI Still Struggles
AI is powerful, but not a replacement for an experienced steel estimator. There are clear limits that show up in real projects.
1. Poor-Quality Drawings and Edge Cases
AI performance drops when drawings move away from clean, digital exports:
Low-resolution scans
Heavy markup, coffee stains, or scan artifacts
Non-standard or hand-drawn symbols
Models trained on millions of examples still rely on recognizable patterns. When information is obscured or distorted, the AI may miss members, misread labels, or fail to understand unusual conditions. This is one reason the Machine Learning in Construction: How LIFT Gets Smarter Over Time article emphasizes continuous learning and human-in-the-loop feedback.
2. Interpreting Intent, Risk, and Business Context
AI can read text; it cannot infer implications the way a senior estimator does:
"FIELD VERIFY" suggests potential delay and risk, not just a note
"TYP UNO" requires understanding where exceptions will likely appear
Shop constraints (door sizes, crane capacity) and GC behavior patterns are invisible to the model
These factors drive decisions about contingency, schedule risk, and margin, core topics in the Ultimate Guide to Steel Estimating that still require human judgment.
A model might correctly read a weld symbol but still miss the implications of tougher inspection regimes or code requirements under the AWS D1.1 Structural Welding Code or D1.8 for seismic applications, areas where an experienced estimator and welding coordinator remain essential.
3. Pricing Strategy and Project Selection
AI can produce a detailed bill of materials and even suggest labor hours based on patterns. It cannot:
Decide which projects fit your current capacity or risk appetite
Choose when to pursue a lower-margin job for strategic reasons
Balance hit rate, relationship value, and backlog health
Those decisions depend on your shop's goals, financials, and market position. The AI can support by giving you accurate inputs faster, but it does not select your work or set your margins.
Designing a Hybrid Workflow: AI + Human Estimator
The most effective shops treat AI as a junior estimator, not as an autopilot. They design workflows that let the model do the mechanical work while people handle judgment, exceptions, and strategy.
1. Let AI Handle the First Pass Takeoff
A typical high-performing workflow looks like this:
Upload structural PDFs to an AI-based tool
Let the system detect and count members, connections, and key notes
Use confidence scores or flags to see where the model is uncertain
This aligns directly with the "Stage 4: AI-powered automation" section in the Ultimate Guide, where the goal is to move estimators off raw counting and into review and pricing.
2. Use Estimators for Review, Adjustments, and Strategy
Estimators then spend time where their expertise matters most:
Reviewing low-confidence detections and complex connections
Checking scope, sequencing, and constructability
Applying shop-specific labor factors, vendor pricing, and margin strategy
3. Connect Takeoff to the Rest of the Estimating Stack
AI is most valuable when its outputs feed directly into your broader estimating process:
Quantities flowing into material pricing modules or spreadsheets
Member and connection data feeding labor estimating templates
Consistent structure for comparing estimated vs. actual performance over time
This is the integration pattern described in the Ultimate Guide, where estimating, detailing, and production data reinforce each other.
Even with highly accurate quantities, safe and efficient erection still depends on human planning that accounts for crane limits, site constraints, and compliance with OSHA steel erection safety requirements. AI can surface the tonnage and piece counts, but it does not design the erection plan.
Common Misconceptions About AI in Steel Estimating
Several myths come up repeatedly in vendor demos and internal discussions. Clarifying them helps set the right expectations.
Misconception 1: "AI Will Replace Our Estimators"
AI can reduce takeoff time by 50–80% on suitable projects, but it does not:
Call out drawing conflicts between architectural, structural, and shop drawings on its own
Decide which RFIs to send or how to frame them
Weigh bid deadlines against current backlog and staffing
Real-world case studies (SSE Steel, MSE, King Steel) show that AI-equipped teams increase bid capacity and responsiveness; they do not shrink estimating teams. This is consistent with capacity and ROI examples in the Ultimate Guide.
Misconception 2: "If AI Is Involved, It Must Be 100% Accurate"
No system is perfect, manual or AI-based. The goal is higher, more consistent accuracy with less time spent:
Legacy automation often tops out around 60–80% detection accuracy on real drawings
Well-trained AI models can reach 95–99% detection accuracy on standard components
Human review is still required, but now focused on the 1–5% of items that are unclear
Misconception 3: "If It Says 'AI' On The Box, It All Works the Same"
Vendors use "AI" to describe very different capabilities:
Simple rules or pattern-matching branded as AI
Limited models trained on a small set of generic drawings
Full computer-vision pipelines trained on millions of structural steel examples
Your evaluation criteria should come from the work you do every day: can the system handle your drawing quality, connection complexity, and project mix, as outlined in your current estimating framework? The Ultimate Guide gives a useful checklist for aligning tools with your process.
How to Evaluate AI Tools for Your Shop
When you bring AI-based estimating tools into a demo or pilot, test them against the realities of your work, not just a polished sample project.
Use Your Own Drawings, Not Demo Sets
Upload:
Older scans with markups
Projects with mixed framing types and tight details
Jobs where you know, from experience, that certain members or notes are easy to miss
Then compare the AI's output against your existing estimates and the principles in the Ultimate Guide (e.g., completeness of scope, connection coverage, and risk items).
Measure More Than Speed
Speed matters, but it is not the only metric:
Detection and attribute accuracy by member type and connection
Time required for review and correction
Ease of exporting data into your current estimating tools and templates
These are the same dimensions that drive ROI in your broader estimating operation, time savings, capacity, accuracy, and margin protection, as outlined in the pillar article.
Aligning AI With Industry Standards and Best Practices
AI-assisted estimating is most useful when its outputs are consistent with the detailing, connection, and safety practices defined by industry bodies and recognized frameworks.
When you leverage AI to accelerate takeoff, ensure your workflow still reflects:
For fabricators who routinely bid pre-engineered metal buildings, AI-generated quantities still need to be interpreted through the lens of project specifications and metal building standards from MBMA and related system guidelines.
Many of the productivity and detailing approaches shops use alongside AI-assisted takeoff are informed by industry research on steel fabrication efficiency and long-running case studies published in the magazine.
Further Reading on AI in Steel Estimating
To see how this expectations framework fits into the rest of your estimating operation, these articles connect the dots:
Watch an AI-powered takeoff complete in minutes instead of hours.
Compare the results against your current process and the benchmarks in the Ultimate Guide to Steel Estimating.
Decide where AI fits in your estimating workflow, and where your team's judgment should stay in control.
Why 95% of AI Projects Fail (And How We're Part of the 5% That Doesn't)
Here's what's happening: too many companies are stuck in what Harvard Business Review calls the "AI experimentation trap." They're running endless pilots, building fancy demos, and chasing the latest AI trends. But they're not solving real problems for real people.
We took a different path at SketchDeck.ai, and our customers are seeing the results every single day.
The Problem With AI Experiments
Most AI projects fail because they start with the technology, not the problem. Companies ask "What can we do with AI?" instead of "What problem needs solving?"
In the steel fabrication industry, the problem was crystal clear to me. Our customers were spending thousands of hours on manual takeoffs. Not because they wanted to, but because they had no choice. Troy Ernst from King Steel told us his team was spending four days on what should be a two-day job. That's not a technology problem. That's a business problem that costs real money.
How We Built AI That Actually Works
We didn't start with AI. We started with estimators.
We spent months talking to steel fabricators, watching them work, understanding their workflows. We learned that manual counting isn't just slow. It's error-prone. It limits how many bids you can submit. It keeps talented estimators stuck doing repetitive work instead of strategic thinking.
Only then did we build LIFT. And here's the key: we built it specifically for structural steel takeoffs in the fabrication industry. Not generic construction. Not general purpose AI. We focused on one industry, one problem, one solution.
The results speak for themselves:
King Steel cut their four-day process to two days
Ennis Steel nearly doubled their bid output and has completed over $1 billion in bids through LIFT
Steelworks of the Carolinas tripled their throughput and completely transformed how they work
FabArc's estimator says what used to take two days now takes minutes
Over $25 billion has been bid through LIFT so far
These aren't experiments. These are transformations.
Why Industry-Specific AI Wins
Generic AI tools are like Swiss Army knives. They can do a little bit of everything, but nothing particularly well.
LIFT is different. It knows the difference between a W8x10 beam and a column. It understands camber and copes. It integrates with Tekla, Strumis, and other tools fabricators and erectors actually use. We didn't just apply AI to construction. We built AI for the steel fabrication and erection industry, by learning from the people who do the work every day.
Mason Carragher from MSE nailed it: "It's the first software that is actually geared towards estimators in a meaningful way." That's because we didn't build a generic tool and try to force it on the industry. We built exactly what estimators told us they needed.
Moving Beyond the Experimentation Trap
The companies succeeding with AI share three things:
1. They solve specific problems. Not "improve productivity" but "reduce steel takeoff time by 80%."
2. They measure real results. Not engagement metrics or usage stats, but actual business outcomes like bids completed and revenue generated.
3. They partner with their customers. Every feature in LIFT came from real fabricators and erectors telling us what they needed.
The Future Is Already Here
While 95% of companies are stuck experimenting, our customers are winning more work. JP Martinez from Metals Fabrication told us: "Using LIFT gave us an advantage to getting more jobs than what we were used to."
That's not an experiment. That's competitive advantage.
The steel fabrication and erection industry is at a crossroads. Labor shortages are real. Project complexity is increasing. Margins are tightening. Companies that embrace the right technology will thrive. Those that don't will struggle to keep up.
Don't Let the Noise Fool You
We're entering what some call the "post-enthusiasm wave of AI." The hype is dying down. The headlines are getting skeptical. And here's the danger: many leaders are misinterpreting the challenges of implementing AI as a signal that AI can't create value.
This is exactly what happened with digital transformation. Companies that dismissed the internet in 2001 after the dot-com crash spent the next decade playing catch up. The same thing is happening right now with AI.
The truth is simple: AI creates massive value when it solves real problems for real customers. Nathan Whitley from FabArc said it best: "What used to take an estimator two days to do, it does it within a few minutes. I've been amazed at every step of the process."
That's not hype. That's a customer whose business is transformed. And while others are pulling back on AI investments, our customers are doubling down because they see the results every single day.
What This Means for You
If you're in the steel fabrication and erection industry still doing manual takeoffs, you're not just losing time. You're losing opportunities. Every hour your estimator spends counting beams is an hour they're not working on the next bid.
But here's the good news: you don't need to experiment. The technology is proven. The results are real. Our customers have bid over $25 billion through LIFT, and they're not looking back.
Innovation in construction isn't about chasing the latest AI trend. It's about solving real problems with proven technology. That's what we do every day, and that's why we're part of the 5% that actually delivers results.
Want to see the difference? Book a demo and we'll analyze one of your own project drawings. No experiments. No pilots. Just results you can measure in minutes saved and bids won.
Because at the end of the day, AI that doesn't save you time and money isn't intelligence at all. It's just expensive experimentation.
Manage Consent
To provide the best experiences, we use technologies like cookies to store and/or access device information. Consenting to these technologies will allow us to process data such as browsing behavior or unique IDs on this site. Not consenting or withdrawing consent, may adversely affect certain features and functions.
Functional
Always active
The technical storage or access is strictly necessary for the legitimate purpose of enabling the use of a specific service explicitly requested by the subscriber or user, or for the sole purpose of carrying out the transmission of a communication over an electronic communications network.
Preferences
The technical storage or access is necessary for the legitimate purpose of storing preferences that are not requested by the subscriber or user.
Statistics
The technical storage or access that is used exclusively for statistical purposes.The technical storage or access that is used exclusively for anonymous statistical purposes. Without a subpoena, voluntary compliance on the part of your Internet Service Provider, or additional records from a third party, information stored or retrieved for this purpose alone cannot usually be used to identify you.
Marketing
The technical storage or access is required to create user profiles to send advertising, or to track the user on a website or across several websites for similar marketing purposes.
