With our top 150 survey this year showing 87% of consulting firms have increased their spend on AI, how exactly is this investment being used to boost the business?
One of the clearest themes from Building’s Top 150 Consultants survey is the rising importance being placed on artificial intelligence (AI) and machine learning by firms in the industry.
As mentioned elsewhere in our Top 150 coverage, the percentage of firms saying AI and machine learning is “extremely” important to the transformation of their business has grown for the third successive year.
A total of 54% of respondents say it is extremely important, up from 50% last year. A total of 91% of firms say AI is extremely or very important, up from 75% three years ago.
>>See also: AI is already paying off for consultants, but human oversight is becoming the differentiator
This importance is reflected in the fact 87% of firms have invested more in AI this year compared with last, with over a third (36%) investing “significantly” more.
The two broad areas of spending seeing the largest rises in investment from firms this year were “digital technology” (on which 84% of consultants are increasing their spending) and “improving productivity” (85%). Both are likely to include spend on AI and machine learning.
How important do you think AI and machine learning will be to the transformation of your business over the next 10 years?
Has the amount you invested in AI technology changed in the last year compared with the previous year?
Another clue to AI’s rising importance can be seen in the fact that two in three respondents (66%) said they have invested more in cyber security compared with last year.
But what specifically are we seeing on the ground? We asked firms to give examples of deliverables from implementing an AI tool or solution; in the responses, several themes emerged.
The dominant theme is efficiency and time savings
Many of the respondents report improved efficiency through using AI, especially in administrative tasks, bid writing and data processing.
Sidara: “AI is delivering efficiencies in document drafting, summarisation, requirements extraction, language editing, research and design visualisation.”
Cundall: “Our range of AI tools being developed not only help make our staff more efficient in day-to-day work, but we are also creating tools that can create flood risk assessments, and estimate embodied carbon for buildings.”
Kier Design: “Examples include using Copilot as an assistant and a time‑saver. It helps efficiently capture actions from online meetings, get a clear one‑page view of dense information, and carry out a fast first-pass summary with action points. This creates more time to focus on strategically important tasks.”
Bellrock: “We’ve built a dedicated AI function, a clear governance framework and a company-wide empowerment programme, all held to one rule: every initiative must show measurable value, whether that’s saved time, better service quality or reduced risk. It’s already landing. Our SI Bid Writer is live in production, speeding up tender drafting and improving consistency.”
Cadence: “The most significant benefits thus far have been realised through the streamlining of administrative and low-value tasks, including the production of meeting minutes, document formatting, information organisation, quality assurance reviews and the rapid dissemination of project data into bite-size/readable versions.”
Many are in the early stages of adoption and not ready to share measurable results
A sizeable proportion of firms told us it is simply too early to give measurable results from their implementation of AI tools.
Rider Levett Bucknall: “Investments have been made but payback is still being measured before being scaled. The outlook, however, is very good.”
Stride Treglown: “We are not yet looking at hard return on investment metrics. We are in an exploration stage, so we are trying to realise the maximum benefits from the technology and provide a sandbox for our staff to test new ways of working.”
Core Five: “[AI is an] emerging area of our business, with results not yet measurable.”
Other firms saying it is too early to give measurable results include EPR, Potter Raper, Rapleys, Civic, Quantem, Atelier Ten, DSSR, Vextrix, CHP, Academy Consulting, Beadmans and Provelio.
A handful did share figures, of varying detail
A few firms were able to share figures illustrating the impact of AI and machine learning, although they ranged in thoroughness.
WSP has deployed Microsoft 365 Copilot to all UK and Ireland staff, using more than 5,000 licences. It said: “Early adopters are saving approximately 2.5 hours per person per week on repetitive tasks.” In its work on Network Rail’s nationally deployed NEST asset management platform, the firm embedded AI that checks new maintenance recommendations against 20 years of historic data: “This removes around 10 minutes of review time per recommendation at the triage stage, unlocking approximately six person-years of engineering capacity annually at national scale, while keeping engineers fully in control of all safety-critical decisions.”
Turner & Townsend said it has piloted and deployed retrieval-augmented generation (RAG) – a technique that enables large language models (LLMs) to retrieve and incorporate new information from external data “to process and map complex programme information at scale”. The firm said: “The impact on onboarding and programme delivery has been significant: saving clients approximately £35,000 per user by reducing reliance on traditional onboarding and support models, accelerating time to value from three months to under three weeks, halving the resource required to onboard and support design teams and reducing client requests for information by 75%.”
Red Engineering Design said it has a broader portfolio of AI deployments and offered as an example its development of an AI-powered report writing assistant that uses historical project reports. The firm said: “Performance assessments demonstrated a reduction in report preparation effort of approximately 25% compared with traditional delivery methods.”
Improving workflow automation
Hollis: “AI is now embedded in day-to-day activities – from drafting and summarising documents to data analysis and workflow automation – driving efficiency gains and supporting faster, higher-quality outputs.”
Data analysis and decision-making
Ridge and Partners: “AI is speeding up concept generation and visualisation. In the engineering disciplines, it’s doing the less glamorous but more valuable work: data extraction, optioneering, modelling, design assurance. Building surveying now uses drone-enabled inspections with AI supporting defect identification and reporting.”
Newmark: “AI helps us transform complex market, portfolio and ESG datasets into clear, interactive dashboards and visuals, so clients can see key risks, performance drivers and ‘what if’ scenarios far more quickly. This improves the speed and quality of decision‑making and makes our advice more transparent and easier to communicate to non‑technical stakeholders.”
Kendall Kingscott: “ AI tools such as Microsoft Copilot have improved efficiency across business support functions by automating routine tasks, supporting faster decision-making and streamlining administrative activity.”
Design, technical and project delivery support
Exigere Project Services: “Our digital platform has enabled us to collate data across our projects to provide faster and more accurate benchmark information, enabling us to analyse the impact of design decisions in real time, improving outcomes for clients and internal efficiency.”
ECE Group has “implemented AI tools across design and business support functions, delivering measurable time savings in tasks such as research, document drafting, data analysis and visualisation, while also improving consistency and enabling teams to provide faster, more responsive services to clients.”
XCO2 has completed a live tool for energy assessments. “It is currently being trialled on the company’s projects, providing added value to clients and project teams at the earliest stages of projects, and helping shape decisions. It helps reduce redesign and abortive work by various disciplines.”
Risk, limitations and human oversight
Some consultants stressed that while they are using AI they are also keen to de-risk it by being careful to retain the guardrail of human oversight.
Mesh Construction Consultancy: “We’ve definitely found measurable results in time through AI uses for our staff, whether that be utilising software to help with recording/minutes for meetings or supporting the speed of cost reports – whilst more efficient, they still do need a human review to ensure there are no errors.”
Baqus Construction and Property Consultancy: “There is a concern of an over-reliance on the tools and we encourage our staff to review everything AI generated to ensure it is correct and they take responsibility.”
Red Engineering Design: “Human oversight remains a fundamental component of all AI-enabled processes to ensure accountability and trust.”
TAKE THE NEXT STEP AT BUILDING THE FUTURE
The biggest challenges facing construction won’t wait. Building the Future Conference brings together the people shaping what comes next – from building safety and skills to delivery, technology, costs and net zero.
Sessions that should not be missed include:
- Keynote speech: Adam Daniels, chief executive, Vistry Group
- Fireside chat: Suzannah Nichol, chief executive, Build UK and Rob Perrins, executive chair, Berkeley Group
- Building safety: lessons, compliance and future directions
- Regen Connect: delivering regional regeneration through collaboration
- The Future of Steel Construction: Raising the Bar – How BS 99001 Transforms Compliance and Competence
- Building skills for the future: workforce development in construction and design
- Data centres: delivering resilient, safe and sustainable digital infrastructure
Hear directly from industry leaders, challenge the thinking and connect with the clients, consultants, contractors and suppliers driving change across the sector.
Wednesday 7 October | Convene Sancroft, St Paul’s, London
No comments yet