Learning from data: it’s all a matter of trust

An image of trust
Image: Santiaga |
Companies need to pool data to tackle the sector’s big challenges, says Gareth Parkes. He explains to Denise Chevin how the Construction Data Trust can help achieve this.

After three years in the making, the Construction Data Trust (CDT) was launched in 2020 by Grant Findlay, then director of strategy at Sir Robert McAlpine, Robert Bryan, partner at BPE Solicitors, and Martin Paver, chief executive of Projecting Success, a data analytics expert with a project management background.

Gareth Parkes, now a board director of the CDT and head of data and analytics at Sir Robert McAlpine, worked with them from the early days to get the organisation off the ground, with the aim of gathering specific data from multiple organisations across the industry. Sir Robert McAlpine provided the initial funding to get the CDT up and running.

With a sufficient volume of data, advanced analytics and machine learning can pick out trends and barriers to improvement. This is difficult to do in individual companies because no one organisation has enough data themselves.

The publication in November 2022 of Trust and Productivity – the Private Sector Construction Playbook called for more developers and construction companies to join the CDT to help improve industry productivity.

Parkes says: “We’re still very much in our start-up phase. But ultimately the big prize is a consistent data set that the sector can leverage for improvements in productivity, health and safety and wellbeing, together with sustainability, carbon emissions and climate resilience purposes. There are many themes that run through the sector where there is almost a moral obligation for us to be sharing data.”

Parkes says there is a great deal of groundwork still needed to reach a stage where the CDT will be able to provide useful benchmark data to its members. It is resourced on a voluntary basis by its board and currently operates on a project-by-project basis. Members contribute financially to the cost of projects.

Gareth Parkes

Gareth Parkes

Parkes joined Sir Robert McAlpine six years ago, and has been in his current role for the last two.

He joined as the company’s knowledge manager, but says he quickly realised the construction sector needs to get data sorted first, otherwise the knowledge and insight is not necessarily built on the strongest of foundations.

He has degrees in chemical engineering and economics and policy of energy and the environment, and worked in the energy sector for the first part of his career.

BIMplus: Tell us more about the need for a data trust.

Gareth Parkes: There’s longstanding recognition, not only within the private sector construction playbook, but also right across the sector, for the need for greater collaboration, a no-blame culture, and also the ability to share knowledge and insights.

All those things come from trust. And yet when it comes to data, there wasn’t a mechanism that allowed organisations to trust each other at that granular level. The CDT is bringing that idea to life by providing independent stewardship of project data.

A data trust is a specific type of institution where the data is independently stewarded. It’s a specific legal entity. The trust holds the data securely within the cloud environment on behalf of its members. Projecting Success acts as our data steward.

Working with the Construction Productivity Taskforce – the coalition of companies behind Trust and Productivity – helped facilitate the process of setting up the Trust’s governance framework. The taskforce shared our vision of collaborating to learn, share and improve industry performance. We’ve also been working closely with the Open Data Institute to inform our development as a data trust.

Where is the work of the CDT now?

We collected our first project data from eight members of the Construction Productivity Taskforce – a combination of clients and contractors including Landsec, British Land, GPE, Lendlease, Mace, Skanska, Morrisroe and Sir Robert McAlpine. Our main project to date has been working with them on drawing up metrics for the seven-step framework set out in a document published in May 2022, Measuring Construction Site Productivity.

Projects that have fed in data include Landsec’s the Forge.

To collect those metrics, what you’re really collecting or starting to collect is several core datasets that have multiple purposes. The obvious example is working hours’ data, which is an aspect of the productivity calculations, but is important as well for health and safety and wellbeing. So, this is where a mechanism like a data trust starts to come into its own, because by pooling data, you can use one set of information for lots of different purposes.

It brings a level of efficiency, and importantly, a level of standardisation to the data that is shared across the sector. I would say there are parallels with the introduction of standards in the world of BIM, and how those standards have helped to embed good practices across the sector.

The value of data often rests on its level of granularity and the most granular level is often within the supply chain, at the sharp end where the work happens. There is a real need to engage and collaborate with supply chain partners to improve data quality. The data trust provides a mechanism for those conversations and a backbone to those conversations that levels the playing field.

What are the achievements to date?

We don’t have enough data as yet – that will come in time – to provide a data benchmark service to members. But the fact that we’ve been able to collect and securely pool data is an achievement in itself – and confounded a number of doubters, who felt that it would not be possible.

“The value of data often rests on its level of granularity and the most granular level is often within the supply chain, at the sharp end where the work happens.”

Gareth Parkes

When we came up with the idea several years ago, we put in a bid to Innovate UK, but we were unsuccessful in explaining our proposition. The feedback that we got was that the proposal was too innovative and the construction sector was not yet ready to collaborate at this level. We are proud that we’ve been able to demonstrate with very limited investment that it is entirely possible and that we’ve got parties that want to do it.

Having demonstrated a willingness to engage, the next step is homing in on exactly what data is going to make a difference. Otherwise, we could spend an awful lot of time and energy collecting large volumes of data that ultimately don’t add positively to the sector.

And I guess that leads then on to the second project that’s in train: getting into incredibly granular detail about the problems that we are looking to solve, collaboratively, to do with health and safety and wellbeing. We need to make sure that the organisations that are sharing their data have a shared view of exactly what it is they’re sharing. So does everybody have the same definition for working hours, for instance?

We aspire to send people home feeling better at the end of the day than they were when they came to work in the morning. It’s a very ambitious aspiration considering that around 40 people are dying annually on construction sites in the UK, and the awful suicide rates among construction workers.

The process we follow is to hold workshops with all parties involved in a project to understand the high-level problems the group wants to try to solve together. From here, we undertake detailed discovery to understand specific problems and pain points and do a high-level gap analysis of the data. In other words, does the data exist to potentially solve the problem? And are organisations willing to then collaborate with their data to solve some of these problems?

Only once there’s this clarity of purpose to solve a problem and belief that the data exists to solve it, can we then start to pool the data. There are various kinds of technical mechanisms that allow us to do that. And once that’s done, the data can be analysed and deliver insights, which, ultimately, must lead to action. It’s those actions that are really going to influence productivity, or health, safety and wellbeing, or climate.

“We’ve come a long way already, being able to demonstrate the art of the possible and demonstrate the willingness of organisations to collaborate.”

Gareth Parkes
What’s next?

The overriding message is that we’ve come a long way already, in terms of being able to demonstrate the art of the possible and demonstrate the willingness and ability of organisations to collaborate at this granular level. The next challenge that we’ve got to face collectively is one of confidence. How do we build confidence in our relationships with one another? How do we build confidence in the data that we’re all looking at together? And ultimately, how do we build confidence that such an innovative way of working with each other can really benefit the sector?

In the work we’ve done to date, we have created several different dashboards for aspects like waste generation, pre-manufactured value, and productivity. We’re able to demonstrate that collecting more of this data is possible, and that we can deduce trends on the back of it. We’ve demonstrated that for it to really work, there needs to be an increase in the quality of the data. And there also needs to be an increase in the volume of data to arrive at increasingly reliable conclusions.

For more information on the CDT, head to its website.

Don’t miss out on BIM and digital construction news: sign up to receive the BIMplus newsletter.

Story for BIM+? Get in touch via email: [email protected]

Latest articles in Analysis