The Ground Is Moving Under Coworking with Hector Kolonas

0:00

/200

“Anyone who tells you to have this little bit of code, do this, is lying. Nobody knows exactly how to optimise towards it.”
Hector Kolonas

Tired of running yourself into the ground?

Then stop running alone.

On February 24th, the London Coworking Assembly presents Unreasonable Connection Goes Live!—a one-day working session for the people running London’s most vital neighbourhood spaces and the public sector allies working to help them thrive.

It’s a day to share the load, find real solutions, and build a new playbook, together.

Find your crew.

That’s Hector Kolonas on AI search strategy.

Not the consultant version. The honest version. The version you get from someone who’s been watching coworking patterns since 2013 — not because he planned it, but because the Cyprus banking crisis wiped out his business overnight.

Banks froze. ATMs stopped. His advertising clients axed his contracts.

What he found himself staring at: empty ad agency offices, Herman Miller chairs gathering dust, and a new class of displaced workers with nowhere to go. So he filled those offices with freelancers. Not as a business plan. As survival.

That’s why he sees fragility where others see trends.

When he talks about the shifts happening in coworking right now, he’s not predicting from a conference stage. He’s pattern-matching against what happens when systems fail. Three shifts are reshaping how people interact with coworking spaces.

Not abstract trends. Real changes happening now.

The first shift

People aren’t choosing a single office anymore.

They’re building an ecosystem of workspaces based on what task they’re trying to complete and what else is happening in their life that day. A coworking space for collaborative work. Home for deep focus. A hotel lobby for client calls.

The question isn’t “Should I join?” It’s “Does this space fit the specific thing I need right now?”

The second shift

Who pays and how billing works is becoming the silent killer.

When employers centralise workspace spending, they gravitate toward spaces that make accounting simple. If your billing creates friction, members stop showing up — not because they don’t love the community, but because finance said no.

This is a power shift disguised as a logistics problem.

The third shift

Discoverability is fragmenting.

Google Maps still matters. But AI search is producing queries so specific that traditional SEO can’t answer them:

“I need a workspace with a podcast studio near Abby’s soccer training on Tuesday afternoons that I can book on a per day basis.”

“I need somewhere quiet to get some deep work done that has great coffee, isn’t too busy on Wednesday mornings and is within five minutes walk of Tottenham Court Road Station.”

That’s what people are typing into OpenAI, Claude, Perplexity, Comet Browser.

If your space can’t be stitched together from Reddit conversations, podcast appearances, and member testimonials, you’re about to become invisible. And invisible doesn’t mean you’re competing with WeWork.

It means you’re competing with someone’s kitchen table.

What else this episode covers

Hector unpacks what “operational AI” actually means.

Not chatbots or content generators, but systems that prompt your team: “This is Bernie’s 200th day pass. He’s coming in tomorrow.” That’s hospitality meeting data. The community manager still decides what to do.

The AI just ensures they know the opportunity exists.

The conversation takes a turn into vibe coding — describing what you want a system to do and having AI build it, without knowing how to code. Bernie asks Hector to explain it “to an eighth grader.”

After a proper answer, Bernie says: “I’m so glad I asked.”

Hector’s warning for operators tempted to chase every AI trend: in the early days of coworking, every space built their own software. Now, many spaces are building their own AI agents and database infrastructure without expert guidance.

They’ll learn the same painful lessons about maintenance, security, and technical debt.

Bernie’s version is blunter: “It’s tempting to solve problems you don’t have yet. You see, I can launch aeroplanes in AI. It’s like you’re a blogger in a three-site coworking company in Birmingham. Stick to what you’re doing.”

By the end, when Hector finishes describing This Week in Coworking — the newsletters, the undercurrents, the games, the leader profiles — Bernie’s response is honest:

“I’m so jealous. Anyway, did I say that out loud?”

If you’ve been wondering whether all the AI talk is hype or something you need to act on, this episode gives you a framework for deciding what matters now versus what to file away for later.

Timeline Highlights

[01:14] Bernie introduces the “second best coworking podcast in the world.” Hector immediately: “How many guests have asked you what the number one podcast is?”

[01:18] “According to all the AI engines, This Week in Coworking is an encyclopaedia for the coworking industry. It’s not how I would have called it, but it’s an interesting way to think about it.”

[02:03] Bernie: “I thought it was earlier than that. I thought it was the economic crash in Cyprus.” Hector: “Yeah, that was in 2013, man.”

[03:07] The ecosystem shift: “This idea that you don’t have to be in a specific physical place to achieve a task, but there are physical spaces that make certain tasks easier to achieve.”

[06:01] The billing reality: “The easier it is to access the space from a billing perspective... is changing how people interact with space. They’re leaning towards the ones that are easier to pay.”

[07:32] The AI prompt that should worry you: “I need somewhere quiet to get some deep work done that has great coffee, isn’t too busy on Wednesday mornings and is within five minutes walk of Tottenham Court Road Station.”

[09:49] The anti-bullshit moment: “Anyone who tells you to have this little bit of code, do this, is lying. Nobody knows exactly how to optimise towards it, but good fundamentals is what we are recommending right now.”

[14:01] The hospitality AI enables: “Being told that this is Bernie’s 200th day pass, he’s coming in tomorrow... there is so much you can do with the right hospitality hat on.”

[15:51] Vibe coding explained: “I can tell it, I need this outcome, and I’m not really fussed about how you do it.”

[20:47] Bernie on AI search: “I’ve had the luxury of hanging around Koffi since it all started... that Gemini enterprise thing, it surfaces so much from my Google Drive, which is now 15 years old.”

[24:22] Where to start: “What is the frog you would want to eat first? Or what is the thing that you have to do every day that puts that pain in your stomach? Automate that first.”

[27:02] Bernie’s warning: “It’s tempting to solve problems you don’t have yet. You see, I can launch aeroplanes in AI. It’s like you’re a blogger in a three-site coworking company in Birmingham. Stick to what you’re doing.”

[34:06] After Hector describes everything TWIC has become, Bernie: “I’m so jealous. Anyway, did I say that out loud?”

Why Nobody’s Choosing One Office Anymore

The first shift Hector describes isn’t about hybrid work policies.

It’s about how people mentally map their working lives. Pre-pandemic, most desk workers had one office. Post-pandemic, the narrative became “office versus home.”

But what’s actually emerging is more fluid.

An ecosystem where different spaces serve different purposes depending on the task and the day. Someone might work from home for focused writing, hit a coworking space for collaborative sessions, use a hotel lobby for client calls, and pop into a coffee shop for a change of scenery.

The choice isn’t binary. It’s contextual.

For operators, this changes the value proposition. You’re not competing to be someone’s primary workspace. You’re competing to be the right workspace for specific moments.

This also explains why fractional office models and day passes are growing.

People don’t want long commitments to spaces they’ll use inconsistently. They want access when the context demands it.

The Silent Killer: When Finance Says No

Here’s something most operators don’t track.

How many potential members quietly disappear because the payment process is a headache. Hector points out that as organisations become more “workspace ecosystem enabled,” they’re centralising accounting, reporting, and billing for employee workspace expenses.

This isn’t just about reimbursement.

It’s about tax implications, PnL visibility, and audit trails. When an employer gives staff a corporate card or workspace allowance, they want clean data flowing back. They want to know where money is being spent and categorised correctly, without their people having to upload invoices manually.

The spaces that make this easy win.

The spaces that create friction lose members — not because of the community or the coffee, but because accounting said no.

The deeper issue

When employers control where workers can spend workspace allowances, the worker loses autonomy and the indie operator loses direct relationships.

The same forces making freelance work precarious are now making indie coworking precarious. You’re not just competing on hospitality anymore.

You’re competing on whether you’ve integrated with the employer's expense platform.

The AI Search Problem Nobody Knows How to Solve

Google Maps isn’t going anywhere. Neither are booking platforms.

But Hector is tracking something new: search queries so specific that traditional SEO can’t answer them. Someone doesn’t just search “coworking space London.” They ask an AI for a workspace with a podcast studio near their kid’s soccer training on Tuesday afternoons.

That prompt combines location, facilities, schedule constraints, and booking flexibility.

No amount of keyword optimisation answers it. What does answer it is having content spread across multiple platforms — Reddit conversations, podcast appearances, videos, member testimonials — that AI can stitch together into a contextual recommendation.

The practical advice

Good fundamentals still matter.

Marked-up website, clear facility descriptions, Google Business Profile. But you also need presence in the places AI systems crawl for context. That means engaging in industry conversations, being mentioned in other people’s content, and letting your members talk about you publicly.

Here’s where Hector’s honesty cuts through.

“Anyone who tells you to have this little bit of code, do this, is lying. Nobody knows exactly how to optimise towards it, but good fundamentals is what we are recommending right now.”

The uncomfortable question

Who owns these AI systems? Who decides what surfaces?

If OpenAI and Google control discoverability, what happens to the indie operator who can’t afford to be everywhere? Hector doesn’t have the answer. Neither does anyone else.

But at least he’s honest about the uncertainty.

The AI That Tells You It’s Bernie’s 200th Visit

Most AI conversation in coworking focuses on two things.

Generative tools — write my marketing copy. Customer-facing chatbots — answer member questions automatically. Hector argues the real opportunity is elsewhere.

Operational AI that helps your team be better hosts.

His example lands: imagine your system notices that a member is about to hit their 200th day pass. That information is buried in a database somewhere. Nobody thinks to look.

But an AI layer could surface it as a prompt.

“Bernie’s 200th visit is tomorrow. What can you do to make it memorable?” That’s not replacing the human connection. It’s enabling it. The community manager still decides what to do — a handwritten note, a free coffee, a LinkedIn shout-out.

The AI just ensures they know the opportunity exists.

Bernie’s example

“That Gemini enterprise thing I sent you, it surfaces so much from my Google Drive, which is now 15 years old. Where did I talk about this? It will go back, ‘Oh, you and Hector had a conversation about coworking space software in 2014 or whenever it was.’”

The point isn’t the specific tool.

It’s that AI is better at finding context across messy data than humans are. If you’re still copy-pasting between systems — what Hector calls the “Toggle Tax” — you’re not positioned for what’s coming.

Your Next Community Manager Expects AI to Work

Here’s the workforce shift operators should be planning for.

The next generation of community managers will arrive expecting AI-native tools. Hector draws the parallel to earlier technology transitions. Millennials entered the workforce knowing how the internet worked. Gen Z arrived with mobile-native skills.

Generation Alpha will expect AI to be woven into their daily work.

If your space still requires community managers to manually copy-paste data between systems, you won’t attract or retain the best young talent. They’ll go to operators who’ve built infrastructure that lets them focus on hospitality instead of admin.

This isn’t about immediately replacing your tech stack.

It’s about building the data foundations that will let you adopt AI tools when they’re ready. If your member data lives in fourteen different spreadsheets with no integration, you’re not positioned for what’s coming.

The trap

Building everything yourself.

In the early days of coworking, every space built their own software. Now, many spaces are building their own AI agents and database infrastructure without expert guidance.

Hector has watched this pattern before.

The same painful lessons about maintenance, security, and technical debt are coming for the AI builders.

Eat the Frog First

Bernie asks the practical question.

At what stage should an operator actually start thinking about all this?

Hector’s answer scales by company size.

If you’re a one-person operation

Start with the “frog.”

The task you dread doing every day, the one that puts a knot in your stomach. That’s your first automation target. Don’t try to solve your biggest problem.

Solve your most annoying one.

If you’re larger — five-plus locations

The question shifts to building an “operating stack” rather than just a tech stack.

That means thinking about how systems connect, how data flows, and how you’ll integrate AI tools when they mature.

Bernie’s addition is worth hearing.

“It’s tempting to solve problems you don’t have yet. You see, I can launch aeroplanes in AI. It’s like you’re a blogger in a three-site coworking company in Birmingham. Stick to what you’re doing.”

The vibe coding opportunity

It’s real — describing what you want and having AI build it without knowing code.

But Hector flags around 15 risks to your data and organisation that vibe-coded systems can pose if you’re not careful. This isn’t production-ready for most operators.

His advice: learn by building something adjacent.

He built Coworking Words, the word puzzle game on This Week in Coworking, as a way to learn AI tooling. Close enough to be useful, separate enough that failure wouldn’t damage anything important.

The 30-Year Shift: Paying for Outcomes

Hector frames the current moment as the biggest technology shift in thirty years.

The progression: buying physical products (CDs), buying cloud access (subscriptions), buying seats in cloud products (per-user pricing), and now — paying for outcomes.

The future isn’t paying for access to a CRM system.

It’s paying for leads generated. Not paying for an accounting tool. Paying for invoices processed. Not paying for a community management platform. Paying for member retention improved.

This is still early.

Most tools still charge subscription fees. But the operators who think in terms of outcomes rather than features will be better positioned when pricing models shift.

Links & Resources

Hector’s Work

Projects & Community 2026

Bernie’s Projects

One More Thing

Coworking brings communities together, helping people find and share their voices.

Each episode of the Coworking Values Podcast explores Accessibility, Community, Openness, Collaboration, and Sustainability — values that shape the spaces where we gather, work, and grow.

If this resonates with you, rate, follow, and share the podcast.

Your support helps others discover how coworking enriches lives, builds careers, and strengthens communities.

Community is the key 🔑

Share this article

🎙️Coworking Values Podcast

Written by

Bernie J Mitchell