Desky on building stable sit-stand desks and practical AI use
John Beaver is the founder of Desky, an Australian ergonomic workspace brand selling sit-stand desks, chairs and accessories backed by a 10-year warranty and stocked spare parts. He explains why Desky engineers frames first for stability and long-term reliability, and how the company uses AI for first-line support while keeping humans for complex negotiations and commercial rollouts.
Engineer products for long-term reliability and keep product data structured so AI can assist effectively. Use AI for repeatable tasks like first-line support and Q&A, but retain human judgment for supplier negotiations, commercial projects and final recommendations.
What’s the quick origin story of your business, and what makes what you do genuinely different from the other options your customers are looking at?
I started Desky after buying a standing desk that wobbled so badly I could not type comfortably at height, and when I returned it I found a market full of desks that looked good in photos but cut corners on frames, motors, and stability. We began as a direct-to-consumer brand because I wanted to control the components, test the lifting systems, and sell a desk that felt solid at every height rather than just another online table.
What makes us genuinely different is that we engineer the frame first and build the desktop and accessory ecosystem around it, so stability, warranty, and long-term reliability drive the design instead of marketing trends. We do not white-label the cheapest frame and wrap it in nice photos; we obsess over wobble control, motor quality, programmable controls, and commercial-grade durability because that is what actually matters when someone uses the desk every day for years
Where does most of your new business come from today, and where do you wish it came from?
Most of our new business today comes from organic search, paid search, email marketing, and repeat customers who already trust the brand, with a meaningful share from small teams and home office buyers who compare specs before they purchase. That direct online engine works because customers can research height ranges, motor types, and warranties on their own, then buy without a salesperson standing over them.
I wish more of our new business came from specified commercial projects where facilities managers, HR leaders, and workplace designers bring us in before furniture decisions are made, because that would let us shape ergonomic rollouts across entire floors instead of selling one desk at a time. That means building deeper relationships with architects, dealers, and enterprise procurement teams so Desky becomes the default specification rather than a late addition.
What’s one thing you’ve actually handed over to AI in the last year, and one thing you tried it on and went back to doing the human way?
In the last year I have handed over first-line customer support triage and product Q&A drafts to AI, because it can instantly pull details like height range, lifting speed, warranty terms, and assembly steps and give a customer a useful starting answer before a human steps in. That has reduced response times and let our team focus on complex issues like damaged shipments, commercial quotes, and fit questions that need real judgment.
I tried using AI for supplier negotiations and quarterly strategy memos, but I went back to doing those the human way because the technology missed relationship context, tone, and the unwritten history that shapes a deal. AI is excellent for structured, repeatable tasks, but it is not ready to replace the judgment and accountability that high-stakes conversations require.
Have you noticed a change in how customers find you or what they already know before they reach out? Is anyone arriving through AI assistants like ChatGPT yet?
Customers are arriving far more informed than they were a few years ago, often knowing to ask about dual motor versus single motor frames, wobble control, programmable presets, ergo-edge desktops, and whether a corner or compact desk fits their space. They have already read reviews, watched setup videos, and compared warranties, so our job is less about explaining what a standing desk is and more about confirming the right fit and value.
Yes, we are seeing a small but measurable number of people arrive through AI assistants like ChatGPT, and we have started tagging those referrals so we can understand what they ask and what they need. It is not yet a dominant channel, but it is growing, and it tells me that AI is becoming a front door for research, which means our product data has to be accurate, structured, and easy for those tools to cite correctly.
If you were advising someone in your industry on all of this for the next 12 months, what would you tell them to actually do, and what would you tell them to ignore?
I would tell them to invest in structured product data, clear warranty and service promises, and human expertise that can guide a buyer from a vague ergonomic problem to the right desk, chair, and accessory setup. The brands that win will be the ones that are easy for AI tools to understand and easy for humans to trust, because a customer who arrives with an AI-generated shortlist still needs a real person to confirm fit, stability, and long-term value.
I would tell them to ignore chasing every shiny AI gimmick, replacing salespeople with chatbots, or optimizing for one specific AI platform, because the underlying business still depends on quality, service, and relationships. Use AI to speed up research, drafting, and data hygiene, but keep humans in charge of negotiations, complex rollouts, and the final recommendation.
Thank you to John Beaver and the team at Desky for sharing what actually worked with Leaders Perception readers.
Want to share your playbook?
Leaders Perception publishes founder and operator interviews every week. A few questions, about five minutes, no calls. It is free and everyone we accept gets published.