Mark's thoughts...

It’s time to end “manels” at tech events

markw@markwilson.co.uk (Mark Wilson) — Fri, 13 Mar 2026 00:00:00 +0000

With International Women’s Day taking place last weekend, I’ve been reminded of a couple of industry events I attended late last year. I thought I’d posted on LinkedIn at the time about the poor female representation, but I can’t find those posts now. So while we’re celebrating women, it’s worth making the point again – because it’s still an issue.

The content at the events was… okay. One was a major IT channel/distributor event – and it was probably a bit too panel-heavy for me. The other event was run by a storage vendor and I was soon bored of tech product-and-feature talk (I only stayed for the excellent afternoon keynote). Both brought together people from a multitude of IT companies. Unfortunately, almost everyone on stage was male.

At the first event, it was well into the afternoon before the first female presenter took the stage. And the day was almost over before we saw a panel that wasn’t an all-male line-up.

The number of “manels” (all-male panels) was so noticeable that I started to keep a tally. At the end of the day the final speaker count was 30 men and 4 women. Thirty to four. In 2025. In a sector that claims to value progress and innovation.

At the second event, the only woman on stage had opened the event and quickly handed over to male colleagues. We didn’t hear a female voice again until after the lunchtime break.

To be fair, I’ve attended some really good events recently that have been far more balanced. That just proves it can be done. But those two stood out – for all the wrong reasons.

Yes, IT is still a very male-dominated industry. But that’s not an excuse for putting on events that simply mirror the imbalance. If anything, it’s a reason to work harder to widen the pool and to champion the diverse voices that are out there.

And to be clear: I’m quite happy to have my share of the limelight. But I’m a middle-aged, white, straight man. If someone else can bring a different perspective to the table, I’ll happily stand aside. We don’t move our industry forward by hearing the same voices over and over again.

There are organisations trying to change things. For example, techUK committed to ending “manels” back in 2017 and reaffirmed that commitment in 2020. For March 2026, they have a TechTogether campaign that is:

“Celebrating what organisations are doing to support diversity, equity, and inclusion, while acknowledging how much more needs to be done.”

And Thoughtworks has written about the importance of bringing diverse voices to tech conferences. But that was back in 2020 - and the events I described earlier make it clear that progress is still too slow.

So a simple ask for event organisers: try to make your events diverse. I know it’s not always easy. But if your speakers are mostly white men, and your panels are all “manels”, it might be worth reconsidering your line-up.

Events shape narratives, and narratives shape culture. If we want a more inclusive industry, we need to start by changing what – and who – we put on stage.

What exactly is a Frontier Firm?

markw@markwilson.co.uk (Mark Wilson) — Thu, 12 Mar 2026 00:00:00 +0000

Back in October 2025 I spent three days at Microsoft’s SME&C EMEA Sales and Partner Summit in Dublin. More than 3000 people were there — a mix of Microsoft staff and partners working across the Small, Medium Enterprise and Channel (SME&C) sector.

At the time I wasn’t entirely sure how much I was supposed to be sharing publicly, and I certainly didn’t get the chance to download any slides. Since then, Microsoft’s Ignite conference, Microsoft AI Tour, and a steady stream of other announcements have brought much of that thinking into the public domain.

One phrase I heard repeatedly during the event was clearly designed to stick in the mind:

“Frontier firm.”

It’s one of those terms that feels destined to appear in slide decks and conference presentations for the next few years. But behind the branding there is an interesting idea about how organisations might evolve as AI becomes embedded in everyday work.

Where the term comes from

The phrase didn’t originate at the summit. Microsoft introduced it earlier in its 2025 Work Trend Index report, which was titled “The Year the Frontier Firm Is Born”.

The idea is that organisations are moving towards a new operating model built around hybrid human-AI teams, where people increasingly direct AI assistants and agents to complete work.

In a blog post titled “Becoming a Frontier Firm: Unlocking the business value of AI”, Microsoft described the concept like this:

“A Frontier Firm is defined not by its size or industry, but by its mindset and execution. These organizations lead with AI-first differentiation, embedding intelligence across every layer of the business — from employee experiences to customer engagements to core processes. ‘Becoming Frontier’ means moving beyond experimentation to enterprise-scale transformation…”

In other words, it’s less about a particular technology and more about how organisations operate when AI becomes part of the normal flow of work.

AI as the defining technology of our time

The summit’s opening keynote set the tone clearly. AI, we were told, is the defining technology of our time.

That may sound like the usual technology conference rhetoric, but I think most people will agree that the scale of investment and innovation is staggering. Something significant is happening.

One message that comes through repeatedly is that AI is only as good as the data and the platform behind it. That’s not unique to Microsoft. It’s a point we hear across the industry, and it’s probably one of the biggest barriers to meaningful AI adoption. Many organisations are discovering that their data is fragmented, poorly governed, or simply not ready to support large-scale AI use.

Another challenge is choosing the right starting point. Successful AI projects tend to begin with a clearly defined business problem — something that can be measured and improved — rather than with the technology itself.

Unsurprisingly, Microsoft’s framing focuses heavily on the platform and the tools. After all, that’s what Microsoft sells.

But the underlying message still holds: without strong foundations in data, platforms, and clear use cases, AI initiatives rarely move beyond experimentation.

The path to becoming a Frontier Firm

At the summit, Microsoft described the journey toward a Frontier Firm as a progression — from humans with assistants, to human-agent teams, and ultimately to human-led, agent-operated organisations.

Stage	Description
Human with assistant	Employees use AI assistants embedded in everyday tools. Copilot (other AI assistants are available) helps draft documents, summarise meetings, analyse information and automate routine tasks. Humans remain firmly in control.
Human-agent teams	Specialised AI agents begin handling parts of workflows. Humans still provide direction and oversight, but agents gather information, analyse data and trigger actions across systems.
Human-led, agent-operated	Humans set strategy and intent while AI agents execute much of the operational work needed to deliver outcomes.

Whether organisations will actually operate this way remains to be seen. But it’s a useful way of describing the direction Microsoft believes AI-powered organisations are heading.

Microsoft also outlined a success framework for organisations becoming Frontier Firms. Rather than focusing purely on technology, the framework highlights four areas where AI is expected to reshape how organisations operate:

Enrich employee experiences
Reinvent customer engagement
Reshape business processes
Bend the curve on innovation

These themes reflect where Microsoft sees AI having the greatest impact across organisations — not just in productivity tools, but in how businesses interact with customers, design processes, and innovate.

Data foundations

We’ve already established that good data is essential for successful AI adoption, so it’s no surprise that this is one of the areas Microsoft spent a lot of time talking about.

The problem is easy to understand. Most organisations didn’t design their data environments for AI. Systems have grown organically over many years, often across multiple platforms, acquisitions, and generations of technology. Data ends up scattered across applications, databases, file stores, and analytics tools — often with inconsistent governance and visibility.

Microsoft’s answer to that challenge centres on Microsoft Fabric. Fabric is the platform designed to bring together analytics, engineering, integration and governance capabilities in a single environment.

Underneath Fabric sits OneLake — a SaaS-based data lake intended to act as a single, unified data store for the organisation.

The goal is to reduce fragmentation and make data accessible across analytics, applications and AI workloads. In Microsoft’s view, that kind of unified data estate is what allows organisations to move from isolated AI experiments toward something that can operate at scale.

The platform underneath the AI story

Behind all the talk of AI sits something more familiar: the cloud platform.

Microsoft currently describes its portfolio in three broad areas — AI business solutions, cloud and AI platforms, and security — built on what it calls a trusted and secure foundation.

Microsoft Azure has matured significantly since the early days of Windows Azure more than fifteen years ago. Today it provides the infrastructure, development platforms, data services and AI capabilities that underpin much of Microsoft’s technology stack.

Developers also play a central role in this story. Tools like Visual Studio and GitHub sit at the heart of how organisations build applications, integrate systems and increasingly create AI-powered solutions.

In other words, while the conversation may focus on Copilot and agents, the underlying platform — cloud, data and developer tooling — is what makes those capabilities possible.

Trust becomes critical

If AI systems are going to operate across business processes, trust becomes even more important.

AI can surface information that previously sat quietly in documents, emails and collaboration platforms. That creates new challenges around governance, security and compliance.

Identity, data protection and threat detection therefore move from the background to the centre of the conversation.

Security is no longer something that sits alongside AI. It becomes part of the foundation that makes AI possible.

Microsoft’s answer to that challenge sits in its broader security and identity platform — services like Entra for identity, Purview for data governance and compliance, and Defender and Sentinel for threat protection and monitoring — alongside endpoint management through Intune.

From experimentation to transformation

The shift Microsoft is describing with the idea of a Frontier Firm isn’t about a single product or capability. It’s about how organisations operate when AI becomes part of everyday work.

That means embedding AI into real business processes, supported by strong foundations in data, cloud platforms and security.

So what exactly is a Frontier Firm?

In simple terms, a Frontier Firm is Microsoft’s way of describing organisations that embed AI deeply into how they operate.

Employees work alongside AI assistants. Agents begin to handle parts of workflows. Human judgement focuses more on direction and decision-making, while systems take on more of the operational work.

Whether the phrase itself will stand the test of time is another question. Technology marketing has a habit of introducing new labels every few years.

But the underlying idea is straightforward enough: organisations using AI not just as a tool, but as part of how work gets done.

“AI Eats the World”

markw@markwilson.co.uk (Mark Wilson) — Wed, 11 Mar 2026 00:00:00 +0000

At the recent NashTech Connect event, Benedict Evans delivered a keynote titled “AI Eats the World”.

Benedict (Ben) regularly publishes large presentations exploring long-term trends in the technology industry. As he describes it on his website: “Twice a year, I produce a big presentation exploring macro and strategic trends in the tech industry.”

“AI Eats the World” is Ben’s current presentation.

What I liked about this talk is that it combines several familiar themes into a clear narrative, illustrated with historical reference points.

Platform shifts

The talk opened with the idea that the technology industry moves in platform shifts.

We know the pattern by now. Mainframes gave way to PCs, then the web, then smartphones. Each time, the old world didn’t disappear, but innovation, investment and company creation moved to the new platform. New gatekeepers emerged. Old gatekeepers lost their grip. New markets appeared, old ones were reshaped, and some things were simply destroyed.

That matters both inside and outside the technology industry. Inside tech, platform shifts change which companies matter. Outside tech, every business ends up asking the same questions: is this just a new tool, a new source of revenue, or an existential threat? How much attention does it deserve?

Microsoft is a good example. When the PC sat at the centre of the industry, Microsoft dominated. When the centre shifted to smartphones, Microsoft became much less relevant for a period because it was no longer central to the platform that set the agenda. After missing the smartphone revolution, Microsoft rebuilt its position through cloud, but with far less dominance than it once had in the PC era.

[I personally witnessed Microsoft’s drive to push Azure as the company attempted to prevent Amazon Web Services (AWS) from gaining market dominance.]

Ben also pointed out that the first companies into a new market are rarely the ones that capture most of the long-term value. That was true for PCs, browsers, search, social media and smartphones. It may well prove true for generative AI too.

One consequence of this is that a lot of what surrounds a platform shift will turn out not to matter. There will be noise, hype, dead ends and ideas that look important for a while before fading away. That is normal.

Bubbles, hype and uncertainty

Platform shifts often bring bubbles with them.

People draw straight lines on charts, talk about exponential growth and insist that this time everything is different. In one sense they are right. Every major technology wave is different from the last one. But that doesn’t mean it can’t still be a bubble.

After the noise clears, the world has usually changed. The technology becomes woven into everyday life. It stops feeling novel and starts feeling normal.

That is why platform shifts matter even when the hype gets ahead of reality.

There is one important difference this time, though. With previous platform shifts we broadly understood the physical limits. We knew what phones couldn’t do. We knew what networks couldn’t support. With AI, we don’t really know why these models work as well as they do, which makes it harder to judge how far they might go.

That means the range of possible outcomes is unusually wide. This may turn out to be “only” as significant as the Internet, the PC or the smartphone. Or it may be something more fundamental. We simply don’t know yet.

Capital is flooding in

From there, the keynote moved inside the technology industry to look at capital, starting with two quotes from industry leaders:

“The risk of under-investing is significantly greater than the risk of over-investing.”
– Sundar Pichai

“The very worst case would be that we have just pre-built for a couple of years.”
– Mark Zuckerberg

That belief is driving extraordinary spending, apparently based on a fear of missing out (FOMO). The biggest platform companies are pouring vast sums into infrastructure. NVIDIA can’t keep up with demand. Infrastructure suppliers often benefit first in platform shifts – Ben compared the moment to earlier cycles where companies like Sun Microsystems rode the wave of new computing platforms.

The power industry is struggling to keep pace too. In some cases, access to electricity is becoming a bigger constraint than access to chips, raising questions about power grids, infrastructure planning and the wider political implications of AI’s energy demands.

“It’s been almost impossible to build capacity fast enough since ChatGPT launched.”
– Kevin Scott, Microsoft CTO

The scale is enormous. Hundreds of billions of dollars are being committed to data centres and associated infrastructure, with more to come. Some of that is funded from the enormous cash flows of highly profitable companies. Some increasingly depends on leasing, borrowing or other financial engineering as firms try to build faster than their balance sheets might naturally allow.

Ben illustrated how different this is from earlier eras of software economics. Microsoft’s historic model was effectively a one-dollar CD in a ten-dollar box sold for hundreds of dollars. Software used to be extraordinarily capital-light, and often benefited from powerful network effects that created dominant platforms.

AI infrastructure looks very different. The frontier keeps moving forward, but each step is more expensive than the last. That raises a difficult question: how long do you keep chasing the frontier?

If models continue to converge in capability, the risk is that they become commoditised.

More models, but not much separation between them

The investment boom has produced rapid progress. Models keep improving. New ones appear constantly. There are proprietary models, open-source models, sovereign models and a growing alphabet soup of acronyms.

But the leading models are increasingly clustered together in capability. On many benchmarks they sit within a relatively narrow range of one another. There may be a new leader every few weeks, but there is not much long-term separation.

Where there is separation is in distribution. OpenAI has broken through into public awareness in a way that others have not. Some competitors may perform just as well on benchmarks, but they don’t have the same reach.

The ambition for companies like OpenAI is to become the universal platform that others build on – the equivalent of Windows for the AI era. But the market may end up looking more like cloud computing, where organisations mix and match components from several providers.

That raises the possibility that AI models evolve more like the semiconductor industry than traditional software – capital intensive, expensive to build, and dominated by a small number of players rather than a single winner.

Models are improving quickly. What remains much less clear is where the long-term value will sit.

Most organisations are still in the “absorb” phase

Outside the technology industry, the pattern tends to be different. The difficulty is that no-one really knows what will work yet.

New technologies usually follow a sequence: first organisations absorb them. They turn them into features and automate the obvious tasks they already understand. Only later do new products, new revenue streams and real disruption appear.

The absorption phase is where most generative AI activity still sits today.

The obvious early use cases are software development, marketing, customer support and point solutions inside large organisations. In many of these cases the change is incremental but still useful: software becomes faster to build, new capabilities appear, and there are tasks where AI simply performs better than previous tools.

Part of the challenge is that generative AI behaves very differently from traditional software. It is good at things computers have historically been bad at – language, summarisation and pattern recognition – and bad at things computers have traditionally done well, such as precise calculation.

Even usage numbers can be misleading. Many people who have access to these tools don’t use them constantly. They use them occasionally, perhaps weekly or monthly, and often only for a narrow range of tasks. Even among those using them, relatively few are paying customers, and usage is often light.

Ben compared this to early web analytics. Page “hits” once looked impressive until people realised that every image on a page counted as another hit. The numbers sounded large, but they didn’t necessarily mean very much.

Many organisations are still running pilots, experimenting with use cases and trying to work out what this technology should actually do for them. The hard part is rarely writing the code. It is working out what the actual problem is, and who would use the solution.

“People don’t know what they want until you show it to them. You’ve got to start with the experience and work backwards to the technology” – Steve Jobs

Automation comes first. Change follows later.

Ben used spreadsheets as an analogy. Before VisiCalc, spreadsheets were literally sheets of paper. When VisiCalc appeared on the Apple II in 1978, it cost the equivalent of around $10,000 in today’s money.

But it could turn a project that took a week on paper into something that could be recalculated in minutes. For accountants, that was transformative. For lawyers, who rarely build financial models, it was far less interesting.

The first step was automation. The deeper changes came later.

Barcodes offer another example. Initially they were introduced to reduce the effort involved in stock-taking. But, once deployed across supermarkets, they enabled far more stock-keeping units (SKUs) to be managed and eventually reshaped the entire supply chain, with automated inventory systems and electronic reordering.

That leads to the next question: once we have automated the obvious things, what comes next?

What changes when you have “infinite interns”?

Ben framed the next phase with a thought experiment: what happens if AI gives you the equivalent of “infinite interns”?

[At this point I was reminded of Chris Weston’s 100 Homer Simpsons analogy.]

Do you do the same work with fewer people? Or do you do far more work with the same number of people?

This is similar to what happened in earlier industrial revolutions. Steam engines effectively gave nineteenth-century Britain the equivalent of millions of additional labour units. The question now is what generative AI might do as a productivity multiplier for knowledge work.

The deeper question is whether the tasks we are automating were ever the real job in the first place. Spreadsheets automated calculations, but the goal was never the spreadsheet itself – it was better financial understanding and decision-making.

That leads to a broader question: were we really trying to automate tasks, or were we trying to achieve something else entirely?

Recommendations, discovery and agentic commerce

This question becomes especially interesting when we look at recommendation and discovery online.

Before the Internet, human editors decided much of what people saw – in shops, newspapers and broadcasters. Internet platforms scaled that process through algorithms.

Platforms like Amazon often operate through correlation rather than deep understanding. They know that people who bought one SKU often buy another, even if the system doesn’t really understand the products themselves.

LLMs may create a different kind of recommendation layer. They may operate at a higher level of abstraction, interpreting intent rather than simply correlating behaviour.

There’s also a long-standing usability principle, often attributed to Bruce Tognazzini: a computer should never ask a user a question if it can work out the answer itself. Systems that understand intent more deeply may reduce the amount of searching, filtering and form-filling users currently have to do.

That could lead to very different buying journeys – asking an assistant what to buy rather than browsing a catalogue or search engine.

The closing perspective

There’s an old line that history doesn’t repeat, but it often rhymes. Technology waves tend to follow familiar patterns.

All the technologies people were excited about before generative AI are still there. E-commerce continues to grow. Autonomous vehicles are becoming more real. Smart glasses, robotics and other long-running ideas haven’t gone away.

That broader perspective matters because technology change is cumulative. New waves arrive before previous ones have fully played out.

Looking further back reinforces the point:

An IBM advert from the 1950s promised that an electronic calculator would deliver the equivalent of 150 engineers.

A US government report from 1955 discussed the coming age of automation, including lifts (elevators). At the time many lifts required attendants to operate them. Then they were automated. Eventually people forgot that they had ever worked any other way. When automation works well, we stop noticing it.

Ben noted that image recognition once felt like something close to magic. Fifteen years ago it seemed almost impossible. Today it is a routine feature inside everyday software.

That brings us to Larry Tesler’s famous line:

“AI is whatever machines can’t do yet.” – Larry Tesler, 1970

Once it works, we stop calling it AI. It simply becomes software. Infrastructure. Just another ordinary part of how things work.

When design forgets the real world

markw@markwilson.co.uk (Mark Wilson) — Tue, 10 Mar 2026 00:00:00 +0000

Design trends come and go, a bit like fashion.

At the moment, one of the prevailing styles seems to be light text on dark backgrounds. On screen it can look clean and modern. Used well, it is perfectly readable too.

But sometimes design choices are made with the screen in mind and not much thought for what happens next.

The dark background problem

A good example appeared in my inbox recently. I was getting ready for attending the ServiceNow AI Summit and, as with many events, the organisers published the agenda as a graphic on their website.

That is already not ideal from an accessibility perspective. Text embedded in an image is not searchable, scalable, or friendly for screen readers.

There is also a practical problem because the agenda uses light text on a dark background. That looks fine on screen, but becomes less practical if printed.

Suddenly you are printing what is essentially a large black rectangle with white text on top. Toner disappears quickly. And if you are like me, the printer available is a perfectly serviceable black and white laser device. Great for documents. Less good for printing a mostly black page.

A slightly odd workaround

So, I took the agenda graphic and inverted it, creating a negative version which had dark text on a light background. On screen it looks dreadful, but on paper it works perfectly well.

More importantly, it doesn’t try to empty the toner cartridge in one go.

It is a small example of what happens when something is designed purely for screens without thinking about what people might do next.

Why print it at all?

You might reasonably ask why I printed the agenda.

Because I wanted to plan my day and, without an event app, I took the analogue approach. I printed the schedule and started circling sessions, and scribbling notes.

It is simple, quick, and surprisingly effective because paper is still a very good interface for thinking.

Of course, I could probably solve this with an iPad and a stylus. But that feels like rather a lot of electronics just to draw circles around a agenda items. A pen and a sheet of paper works perfectly well.

Design for how things are used

None of this is really about dark backgrounds. They can work very well in the right context.

The issue is when design is optimised for appearance rather than use.

And the problem is not limited to conference agendas. I’ve seen similar with airline boarding passes surrounded by dark-coloured ads on an A4 PDF. I’ve also experienced event tickets made available in PDF format but using dark designed graphics.

If something might reasonably be printed, a print-friendly version helps. If something contains structured information like a conference agenda, publishing it as text instead of a graphic helps even more.

And if a design choice accidentally turns a simple print job into a toner-consuming black rectangle, that is probably a small sign that the design process has drifted a little too far from real-world use.

Sometimes the most useful design choice is simply remembering how people actually use things.

AI outcomes, not AI hype: reflections from a PyData panel

markw@markwilson.co.uk (Mark Wilson) — Thu, 5 Mar 2026 00:00:00 +0000

A few weeks ago I joined a panel at the PyData Milton Keynes meetup.

PyData is a global community of developers, data scientists and engineers working with open-source data tools. The Milton Keynes chapter only began running events in 2025, and this was its first in-person gathering.

The panel was hosted by Grace Farayola. I was joined by Samantha Roberts from SDG Group, Madara Premawardhana from the University of Buckingham, and James Graham, former CEO of BsoftB.

The evening’s theme was FutureTech 2026: innovations that will shape tomorrow.

Ahead of the event Grace shared a set of questions that she intended to ask the panel. The discussion covered a lot of ground, but I’d prepared a few thoughts in advance.

Sadly, I don’t have a record of the great insights that my fellow panellists shared (it’s pretty hard to take notes when you’re on the panel). Instead, this is a short reflection on the topics I was asked to cover, based on the answers I prepared beforehand.

The future of tech is not just about “the shiny stuff”

When asked what the future of technology means from my perspective, my answer was fairly simple.

The future of tech is outcome-led: secure platforms, trusted data and modern applications solving real problems, with AI built in so value can be delivered safely and at scale.

It’s not particularly glamorous.

There are plenty of exciting innovations and bold ideas in the industry right now. But most organisations are still trying to get the fundamentals right while staying responsive to their customers, clients or citizens.

That may not make headlines, but it’s where progress really happens.

AI is becoming a core operating layer

The more interesting question is what organisations should prioritise in 2026.

The answer isn’t “more technology”. It’s the technology that turns AI into measurable outcomes.

AI has moved beyond experimentation. It is becoming part of the operating fabric of organisations: a productivity multiplier, a margin lever and a differentiator for customer experience.

That puts CTOs firmly in the hot seat.

Technology leaders are increasingly expected to turn AI into real business impact, not innovation theatre.

The foundations that make this possible are not new:

Trusted, unified data.
Modern, composable platforms.
Secure and resilient infrastructure.
Experience-led technology that people actually adopt.
Sustainable architectural choices.
Cross-functional teams that can deliver outcomes.

Put simply: AI only works when the foundations are solid.

Turning experiments into outcomes

Many organisations are still experimenting with AI and cloud technologies.

The challenge is moving from experiments to measurable value. The simplest way to do that is to start with the outcome: what does good look like? (Remember Stephen Covey’s “Seven Habits of Highly Effective People”, where the second habit is “begin with the end in mind”.) In other words, decide which metric you want to change, then work backwards.

A practical playbook looks something like this:

Start with the outcome – which metric will move?
Prepare the data – quality, lineage and access.
Prepare the platform – APIs, automation and observability.
Put guardrails in place – security, governance and responsible AI.
Deliver a thin slice – prove the value quickly.
Scale what works.

It’s often said that a very high percentage of AI projects fail (usually quoting various analysts and academic institutions). It’s not just AI projects either. Many initiatives struggle because they start with the technology rather than the outcome.

If you can’t explain the impact in a sentence, the project probably needs more thinking.

Innovation and compliance are not opposites

One of the debates during the panel was whether organisations can realistically be both innovative and compliant.

I think they can.

Innovation and compliance are often presented as competing priorities (“while America innovates, Europe regulates”), but that’s the wrong way to think about it.

It’s more like driving a car. The brakes and steering don’t slow you down. They allow you to move faster, safely.

Technology governance works the same way. When guardrails are built into the design from the start, they enable innovation rather than restrict it.

The real shift for technology leaders

The biggest change we are seeing is not technical. It’s organisational.

AI outcomes are becoming career-critical for technology leaders.

In many organisations, AI is no longer a side experiment. It is becoming a core part of how the business operates, influencing productivity, margins and customer experience.

That means the expectations placed on technology leaders are changing.

Just as CFOs are held accountable for financial performance, CTOs will increasingly be judged on whether AI initiatives move the business forward.

Not demos. Not pilots. Not slideware. Real business outcomes.

Imagine a world where offices are designed around people, not desks

markw@markwilson.co.uk (Mark Wilson) — Thu, 26 Feb 2026 00:00:00 +0000

There is still a lot of noise about getting people “back to the office”. But that debate often skips an obvious question: back to what, exactly?

If the answer is rows of hot desks filled with people on back-to-back video calls, it’s not surprising that enthusiasm is limited. I’ve written before about the future of the office and, more recently, about hybrid work and so-called “watercooler moments”.

My view hasn’t really shifted. Offices still have a role. But that role is about collaboration, not simply attendance.

So rather than arguing about policies, it’s more interesting to ask what a genuinely good office would be like:

Imagine a world where offices adapt to people, not the other way around. Where meetings don’t start with five minutes of technical fumbling. Where space is shared fairly, energy is used sensibly, and design decisions are backed by evidence rather than habit.

Last week, I (along with colleagues from Node4’s Office of the CTO) visited Cisco’s Smart Workspace in London and that description wasn’t far off what we experienced.

The London Smart Workspace opened in February 2025 and is the fifth iteration on Cisco’s estate (after New York, Chicago, Atlanta and Paris).

It’s not a glossy showroom. It’s a working environment built around the idea that when people make the effort to come together physically, the space should support collaboration properly.

Managing space like it matters

The first thing that struck me was how deliberately space is handled.

There are no fixed desks and there are meeting rooms of all sizes — from single person booths to large collaboration spaces.

If a meeting doesn’t happen, the booking is released automatically. Larger bookings are handled through a concierge-style process rather than simply clicking a bigger room in a calendar.

The system avoids the familiar scenario of two people occupying a room built for ten while others wander the floor looking for somewhere to sit.

That might sound over-managed, but it’s really about fairness and visibility. Space is expensive. If you’re going to maintain it, you may as well use it well.

When organisational silos don’t help

Something else that came through in conversation was how closely human resources, facilities, and information technology (IT) teams need to work together to design spaces like this. Not as separate departments negotiating responsibilities, but as a joined-up function focused on the overall employee experience.

Once the building itself is part of the digital estate, those boundaries blur.

Lighting, room booking, collaboration platforms, air quality and occupancy data don’t sit neatly in a single traditional box. And when the building management system (BMS) is connected to the network, IT and operational technology (OT) stop being abstract ideas in a strategy deck and become part of day-to-day operations.

Infrastructure in plain sight

The open ceiling makes that integration visible.

Power is delivered over the network. Lights, blinds, access points, sensors and video devices all use PoE. It’s practical rather than decorative, and it reflects the reality that modern workspaces are as much digital infrastructure as physical environment.

There are thousands of sensors across each floor measuring occupancy, environmental conditions and usage patterns. Air conditioning adjusts based on actual room use rather than fixed assumptions. That combination of real-time data and automation reduces energy use without compromising comfort.

Older sensors can be a security risk. Properly managed ones reduce risk while still providing useful insight.

In some cases, ultrasound is used for room awareness and proximity detection rather than Bluetooth, because it does not travel beyond doors and windows in the same way. It’s a small technical decision, but it shows the level of thought that has gone into the design.

This isn’t presented as an all-or-nothing transformation. The phrase used was “crawl, walk, run”. Start with obvious value. Expand once you understand the benefit.

Designed for collaboration, not hierarchy

The meeting rooms themselves felt considered rather than over-engineered.

They face natural light, with blinds that adjust automatically throughout the day. Plants and natural materials soften the space. Together, it reflects what designers would call a biophilic approach — bringing natural elements into the working environment to make it feel more human — but done in such a way that it doesn’t create a maintenance burden.

Acoustic treatment is built in and even the table shape is deliberate. I looked it up and apparently it’s an isosceles trapezoid (schoolboy Mark would probably have known that - just one of many facts I’ve forgotten over the past few decades). The shape means remote participants aren’t visually sidelined and no one sits at a symbolic “head of the table”.

Different areas are designed with varying levels of stimulation. Some spaces are more open and collaborative, others quieter and lower intensity. It’s a simple acknowledgement that not everyone works best in the same sensory environment.

These may sound like small things. But small design decisions add up. In sport, they would be called marginal gains.

The technology in the room follows the same thinking. It can detect how many people are present, where voices are coming from, and who is speaking — not just that “the room” is speaking. Facial processing happens locally at the edge rather than sending images to the cloud, and any user profiles are tokenised so they are not easily extractable.

The aim isn’t surveillance. It’s to make hybrid meetings feel less awkward and less second-class for those who are not physically present.

You might expect that, in a Cisco office, everything would need to run Webex. Fortunately, that isn’t the case. Rooms can operate across multiple collaboration platforms rather than being locked to one. The equipment runs natively on Cisco’s RoomOS, allowing it to function as Microsoft Teams Room (MTR), Webex, Zoom or a SIP endpoint.

That flexibility matters. People shouldn’t have to think about infrastructure when they walk into a meeting. They should just get on with the conversation.

Getting around

One of the most impressive features is the way that the office integrates with people. That sounds cringy and AI-written (it wasn’t, BTW) but wayfinding is handled simply. Scan a QR code and you can be directed to your room. Scan another and your collaboration profile follows you into the space. It doesn’t adjust desk height or lighting at one of the individual desks (yet), but the direction of travel is clear.

There are strategically-placed displays on the office walls to guide you, integrated with mobile apps.

An office that earns its keep

Out of hours, the building scales itself back. Lighting, heating and cooling adjust automatically based on actual use. That reduces cost and provides real data on how the space is used, rather than relying on assumptions.

Although there is significantly more embedded technology than in a traditional office, operating costs have reduced. Lower energy consumption is part of that equation.

All of this reinforces a simple point: if organisations want people in offices, the space has to offer something that home working does not. Not presenteeism. Not rows of desks. But a better environment for collaboration.

The technology already exists. The harder step is choosing to design work around people rather than around furniture layouts inherited from a different era.

Resources

Three small experiences that say a lot about customer experience

markw@markwilson.co.uk (Mark Wilson) — Thu, 19 Feb 2026 00:00:00 +0000

Most organisations don’t set out to deliver a poor customer experience. In fact, many invest heavily in tools, processes, and metrics designed to do the opposite.

And yet, small moments still go wrong.

Not the dramatic kind that trigger complaints or refunds, but the quieter ones that leave people slightly dissatisfied. They are often the result of sensible decisions made in isolation, which only reveal their downsides when they meet a real person.

The three experiences below are unrelated on the surface. Taken together, they say something useful about how organisations present themselves to the people they serve.

A birthday lunch and a missing conversation

My wife booked a table at a pub she likes for a birthday lunch. Her first call went unanswered. On the second attempt, she got through and asked for a dog-friendly table, ideally close to the fire.

I called separately to ask whether anything could be done to mark the birthday. Nothing elaborate. Perhaps a candle on a brownie or a simple dessert. I was told that wasn’t possible, but that I was welcome to bring in my own cake.

When we arrived, we did have a dog-friendly table, but not one near the fire. That part of the message hadn’t been passed on. The staff on the floor picked up on our disappointment immediately and worked hard to put things right.

Then, when I mentioned the birthday, they suggested exactly what I’d asked for on the phone.

Later, we discovered why. The initial call hadn’t gone to the pub at all. It had gone to an outsourced “base camp”, and not everything had made it through.

No one on site had done anything wrong. The staff were excellent, and we enjoyed the lunch. But a decision taken to drive efficiency elsewhere had created friction at the point where it mattered most. From our point of view, there had been one conversation. Internally, there had been several.

The join showed.

When internal distinctions meet external reality

I recently spoke with some IT leaders about an organisation that was trying to improve its Net Promoter Score. Unfortunately, there were some inconsistencies in delivery, particularly around project work.

Internally, this prompted a familiar discussion. What counted as “service”? What sat outside it? Which activities should influence customer feedback, and which should not?

From the customer’s point of view, those distinctions barely existed.

They saw one supplier and one relationship. Projects and service were not separate experiences; they were simply different parts of the same one. Until everything worked consistently, the feedback reflected that inconsistency.

These internal distinctions are not unreasonable. They matter for governance, accountability, and commercial clarity. But customers don’t experience suppliers through operating models. They experience outcomes, and they judge the whole by the weakest part.

Once again, the organisation was presenting its internal structure to the outside world.

“Oliver”, broken data, and silence

The third example is smaller, but less forgivable.

I received a marketing email from a company offering to help me get better outcomes from my data analytics. It was addressed to “Oliver”. My name, is “Mark”.

I ignored it. They followed up. Still “Oliver”.

I replied, deliberately using the wrong name in return, and explained that I didn’t need their services. I also pointed out that there was no unsubscribe option on their email.

I never heard back.

I wasn’t a customer, and I wasn’t complaining. I was responding to an approach and giving clear feedback on why it wasn’t landing. Whether the issue was bad data, a broken CRM, or careless automation is almost beside the point.

The attempt to engage was sloppy, and the silence that followed suggested nobody was listening anyway.

What these moments have in common

These stories aren’t really about pubs, IT services, or marketing emails. They’re about how organisations choose to present themselves.

In one case, efficiency gains elsewhere fractured the experience at the frontline. In another, internal structures were allowed to leak into the customer relationship. In the third, a poorly executed attempt to win new business did more harm than good.

Different contexts, but the same underlying problem.

Customers experience organisations horizontally. Most organisations are still designed vertically. When responsibility, information, or ownership changes hands internally, customers feel it immediately.

People don’t move through departments. They move through moments.

Presentation matters more than intent

At least one of these organisations was careless. The others were well-intentioned but exposed by the way they had chosen to operate.

The outcome, however, was the same.

Each of them presented itself poorly at a moment that mattered, whether through broken handovers, visible internal boundaries, or a lack of basic care in how someone was approached.

Customers and prospects don’t see effort charts or efficiency targets. They see what lands in front of them and draw conclusions from that alone.

And those conclusions, once formed, are difficult to undo.

A first post... of sorts

markw@markwilson.co.uk (Mark Wilson) — Mon, 9 Feb 2026 00:00:00 +0000

I’ve been blogging since 2004, and I have over 2500 posts published on my blog. Unfortunately, that site has become unreliable, and I can’t work out what’s wrong with WordPress. For now, the site is still online, except for the few minutes each day that it spikes my CPU quota to 100%.

A new blog

I already moved my homepage away from the blog, with a nice, simple, static HTML and CSS website, and a Github Action to push it to my web host. Whilst one of my friends teased me for hand-writing code, it’s noticable how fast it is. So I decided to have a go with another simple site. No more SQL databases, over-complicated plugins — just back to basics, using Hugo.

Hugo has been around for a few years now, so there’s probably something else I should be using but I like the idea of writing posts in Markdown (I love working in Markdown, and have started to use it more in my day job…).

I have to admit that getting up and running with Hugo has been a bit more complicated than I hoped, but that’s the technology we’re running on for now, and I can start to publish my thoughts again.

To dos

There are a couple of things I still need to do before I can launch:

I know a few people subscribe to the old blog, so I need to get RSS up and running, and ideally maintain the URI for the feed.
I would like to bring some, if not all, of the old content across.

Credits

Before I sign off this first post on the new platform, Flavio Copes’ 2020 post on How to Create Your First Hugo Blog: a Practical Guide has been a huge help in getting started. I’ve also been Googling like crazy and finding lots of advice in the Hugo Documentation and Hugo Discourse.