AI promises better decisions and improved project performance, but most engineering consultancies lack the foundations required for it to work.
Large capital projects are notorious for underperforming. Independent Project Analysis (IPA) reports that 74% of completed Large and Technically Complex Projects assessed in Australia failed to meet cost or schedule objectives. While comparable data for small and medium-sized engineering projects is harder to find, most consultancies recognise the same persistent issues: unreliable estimates, weak cost control, and schedules that drift.
A major contributor is rarely addressed directly. The systems engineering consultancies use to manage their projects were never designed for how their work is actually priced, delivered, and governed. As a result, engineering firms struggle not only to execute projects well, but to learn from them. That gap is becoming more visible as AI adoption accelerates.
Tools that Don’t Fit the Work
Walk into almost any engineering consultancy and you’ll see the same pattern: Excel doing the heavy lifting but fraught with errors and unable to scale; enterprise project controls platforms built for mega-projects but too heavy and expensive for small and medium sized projects; generic project management tools that don’t align with engineering project governance; and critical data pushed manually between systems.
Each tool solves only part of the problem. Together, they create a fragmented ecosystem that inhibits commercial discipline and weakens delivery control. The impact shows up most clearly in pricing and project execution.
Pricing: Where the Wrong Tools Cost You
Get pricing wrong and you either lose money on work you’ve won or lose work to the competition. Sustainable commercial performance depends on an accurate as-sold position that carries through the project lifecycle.
In practice, most organisations still rely on Excel for pricing. Multiple versions of pricing sheets make it difficult to maintain current cost rates, apply margins consistently, or reflect client-agreed rates. Escalations and contingencies are often poorly estimated or ignored, and markups vary by author. As research shows, spreadsheets are prone to errors and lack formal controls which amplifies these problems.
The deficiencies compound at project setup. Without time-phasing, an estimate cannot become a project baseline. Without a work breakdown structure that mirrors how the project will be executed, the plan has to be rebuilt. The result is predictable. Complex projects are underpriced, simple ones are over-scoped, and without a reliable baseline, outcomes vary and performance is difficult to manage.
Spreadsheets are prone to errors and lack formal controls.
Project Delivery: Where Performance Goes Dark
Once a project is underway, tooling limitations shift from estimation to execution. When systems don’t automatically capture actuals such as time, expenses, and payments, project managers are forced to assemble data manually. Errors creep in, and trust in the data erodes.
Progress tracking suffers even more. Without earned value management support, teams can’t tell if progress aligns with costs and schedule. Without structured change control, scope changes trigger manual workarounds instead of controlled baseline adjustments. Projects drift off baseline with no reliable way to forecast final outcomes or justify recovery actions to clients.
The consequence is decision latency. Project teams spend valuable time verifying data and compiling reports instead of acting on insights. By the time issues surface in status meetings, you’re already behind. The warning signs were there: scope creep, resource bottlenecks, rework, patterns that historically lead to overruns. They stayed buried because the tools weren’t designed to surface them early.
The Industry Response: AI and Offshoring
Large global engineering consultancies have recognised these pressures and responded decisively. Work has been shifted to lower-cost delivery centres to reduce risk and improve competitiveness. This trend is accelerating with the global engineering service outsourcing market anticipated to grow from USD 1.87 billion in 2025 to approximately USD 10.40 billion by 2035.
At the same time, these firms are investing heavily in AI to automate elements of engineering delivery. Worley CEO Chris Ashton has stated that AI is fundamental to the company’s strategic direction and is intended to support a shift toward more fully integrated, digitally enabled project delivery.
While these investments can depress short-term profitability, they are widely seen as essential to remaining competitive over the medium term.
Why This Isn’t Enough
Offshoring and AI-driven engineering automation address cost and efficiency, but they leave a more fundamental problem untouched. Pricing accuracy and project performance management are still constrained by systems that were never designed for the small to medium-sized engineering projects that make up the majority of projects delivered.
Without high quality data foundations, AI tools cannot deliver meaningful returns. When pricing lives in disconnected spreadsheets and delivery data is fragmented and inconsistent, AI has nothing reliable to learn from.
McKinsey notes that many organisations struggle to integrate data into AI models and that poor data quality and governance are significant barriers to scaling AI successfully. In this environment, AI doesn’t transform performance, it simply amplifies the limitations of the systems beneath it.
The Path Forward
Small and medium-sized engineering projects cannot rely on spreadsheets or enterprise systems designed for mega-projects. Fit-for-purpose project delivery platforms are essential. They capture pricing, planning, and delivery data as work happens, provide time-phased baselines, automatically record actuals, and manage change within the same structured framework that governs performance. This integration ensures data is reliable, continuous, and ready for AI to generate meaningful insights.
Firms adopting modern delivery systems are building compounding advantages. Each proposal and project strengthens the data foundation, improving estimate accuracy and surfacing risk signals earlier. Over time, this feedback loop sharpens commercial decision-making and reduces reliance on individual effort.
The market has already decided on AI. The remaining question is whether firms will invest in the foundations that allow it to maximise value. Those who delay will eventually adopt similar tools, but with less data, less experience, and less competitive leverage, while the leaders continue to pull ahead.
FAQs about the AI gap in engineering consultancies
What is the AI gap in engineering consultancies?
AI is only as effective as the systems beneath it. Most engineering consultancies still rely on spreadsheets, disconnected tools and platforms that were not designed for how projects are priced, delivered and governed, so they lack the data foundations AI needs to work.
Why are spreadsheets a problem for pricing engineering projects?
Multiple versions of pricing sheets make it difficult to maintain current cost rates, apply margins consistently or reflect client-agreed rates. Escalations and contingencies are often poorly estimated or ignored, markups vary by author, and spreadsheets are prone to errors and lack formal controls.
Why can't AI and offshoring fix project performance on their own?
They address cost and efficiency, but pricing accuracy and project performance management are still constrained by systems that were never designed for small to medium-sized engineering projects. Without high quality data foundations, AI has nothing reliable to learn from, so it amplifies the limitations of the systems beneath it.
What should a project delivery platform for engineering consultancies do?
It should capture pricing, planning and delivery data as work happens, provide time-phased baselines, automatically record actuals, and manage change within the same structured framework that governs performance, so data is reliable, continuous and ready for AI.
What is the benefit of adopting a modern project delivery system early?
Firms build compounding advantages. Each proposal and project strengthens the data foundation, improving estimate accuracy and surfacing risk signals earlier. Over time this sharpens commercial decision-making and reduces reliance on individual effort.