October 15, 2026 · 5 mins read

Bringing Order to Chaos: Reflections from HE Connect 

There is an idea in software development called Conway's Law, which says that digital systems come to reflect the organisations that design them. So over time, a website or digital estate becomes a kind of mirror, revealing the structures and organisational choices of the teams who build it. 

If that's true, universities face an almost uniquely difficult task when it comes to creating a joined-up digital experience. Colleges, faculties, departments, professional services, research groups, spin-outs, partnerships, libraries, museums and hospitals: few organisations are built from so many separate parts. 

So, if there was an unofficial theme from this year’s HE Connect conference, it was the urgency with which teams must try to tackle this challenge if they’re to win over potential students – along with the LLMs that shape their decisions.  

Taming Complexity 

The speakers came at the problem from different directions, but they shared an overarching perspective. A coherent digital experience isn't something you add at the surface or fix with technology alone - it’s a result of the hard, organising work that humans do behind the scenes.  

At the largest scale, Stratos Filalithis described how the University of Edinburgh is consolidating more than 8,000 separate websites, using a structured audit to decide what to keep, fix and retire. Johannes Nygaard showed the University of Copenhagen working from underneath, establishing content types and structures through a multi-year programme, of work bringing institution-wide consistency to their digital estate, department by department. 

Others focused on people and priorities. June Shannon explained how the University of Limerick has broken down silos between teams, through cross-departmental workshops and user research with site editors as well as external audiences, including users with accessibility needs. Lucia Tornero Gonzalez showed how Universidad Pontificia Comillas in Madrid has organised its communications and digital experience around memorable moments, putting its content effort where it matters most.  

On day 2, Samantha Fanning from UCL, Paul Leigh from the University of Limerick and Sarah Tan from LSE all gave incredibly practical insights on how they were tackling the mammoth and complex content challenge that faces institutions. 

Whether the end point is a new platform or a better-run existing one, the hard work starts with structure, people and priorities. 

What the Student Sees 

Our talk picked up the same idea from the other side. Inside a university, people see departments. A prospective student sees one institution. They never see the org chart; they experience the result of it, in every email, web page, open day and phone call. 

Increasingly, they also experience it through AI. Research from UCAS and WonkHE in May 2026 found that 48% of prospective students now use AI to explore their options. When those tools describe a UK university, they mostly draw on sources the university doesn't control. Our own GEO research data shows YouTube and Wikipedia are referenced most often, followed by league tables, UCAS, Reddit and GOV.UK. 

Advertising veteran Jeremy Bullmore once said that “people build brands the way birds build nests, from the straws and scraps they chance upon”. And we see AI search working in much the same way: gathering fragments of evidence from everywhere and assembling them into a picture. 

That matters because the scraps aren't only marketing. If brand is the central organising thought of an organisation, shown through everything it does, then some of the most important brand decisions don't feel like brand decisions at all: 

• When the phone lines are open 

• How and when applicants are contacted 

• Who briefs the academics and student ambassadors at open days 

• Whether to add yet another page to the website 

Each of those choices might make perfect sense to the team that owns it. Together, though, they shape the student's experience. And through reviews, forums and real stories, they shape what AI tells the next applicant. 

Shortening the Distance 

So what can universities do? Our answer was to shorten the distance between what an institution says, what students experience, what frontline staff know and what the organisation does about it. 

Often the best place to start is with the people closest to prospective students. Recruitment teams, admissions staff and student ambassadors frequently know something is wrong long before it shows up in the numbers. Bringing them together and asking what they're really hearing can be more revealing than any dashboard. 

It also helps to hold up the mirror. Track how AI tools describe your institution and compare that with how you describe yourselves. Where the two don't match, there is usually some insight to be gained and an opportunity to improve. 

Conway's Law suggests universities might never be able to fully escape their own complexity. But as the speakers at HE Connect showed, through consolidation, structured content, collaboration and focus, they can decide what the mirror reflects.

---