By Elton Boocock, founder of Thinkivity, providing AI consultancy and training exclusively for the UK glazing industry.

In last month’s column, I talked about the quiet advantage, the small, almost invisible ways AI is already saving time in everyday tasks. Writing Excel formulas faster. Prioritising emails properly. Turning raw data into clearer meeting slides. None of it dramatic, but all of it useful.

This month, I want to focus on something far more significant; Documents. And specifically, how AI could be helping to save you thousands of pounds. Now that’s the AI we all like!

Across the glazing supply chain, our businesses run on documentation. Quotes, surveys, order forms, confirmations, purchase orders, invoices, stock lists and technical specifications move constantly between teams, suppliers and customers. Every one of them represents instruction. Every one of them carries financial risk.

Most of the costly problems in our sector do not start in production. They start on a document.

A measurement slightly wrong. A finish misread. A hardware option overlooked. A specification that does not quite match what was agreed at survey stage. None of these errors are dramatic in isolation, but in glazing they can quickly turn into remakes, wasted transport, strained relationships and margin erosion.

For years, this has simply been accepted as part of the industry. Mistakes happen. Remakes are unfortunate but inevitable. A certain level of loss is written off as the cost of doing business.

Not because anyone wanted it that way, but because there was no realistic alternative. The only defence was careful people, double checking, and more pressure on already stretched teams. That is what has changed.

AI, when applied properly and within glazing specific systems, is exceptionally strong at comparison and validation. It does not get tired late in the afternoon. It does not rush because the phone is ringing. It does not assume something is correct because it looks familiar.

AI can compare one document against another.

It can flag inconsistencies between a quote and an order confirmation. It can highlight missing fields, mismatched specifications, or unusual deviations from standard configurations. In effect, it becomes a constant second pair of eyes.

The human still signs things off. The human still applies judgement. But the heavy lifting, the repetitive cross checking and pattern spotting, can now be supported in a way that was never previously possible.

This is not just relevant to installers and retailers. Fabricators process complex inputs from multiple customers, often in different formats. Component suppliers and system houses rely on accurate technical data flowing through accurately. As documentation moves down the supply chain, the impact of a small mistake can multiply rapidly.

For the first time, the industry has access to tools designed specifically to tackle this problem at scale. At Thinkivity, we have developed DocChecker to do exactly that, automatically reviewing glazing documents to identify inconsistencies and potential errors before they move further down the chain. It is not about replacing people. It is about supporting them and reducing losses that were previously considered unavoidable. That shift is important.

Human centric AI means using technology where it is strongest and allowing people to focus on what they do best.
In document workflows, AI is strong at checking and validating. Humans are strong at interpreting, questioning and approving.

When those roles are combined properly, order processors gain headspace. Managers gain confidence. Margins are better protected. Stress levels reduce because obvious issues are flagged early rather than discovered late.

In glazing, the cost of a small mistake has always been real. What is new is that we no longer have to accept it as inevitable. And that is not AI hype. It is real progress.