Why we built Safa Residences: the gap in UK short-let management we couldn't ignore

What we saw before we started

Most short-let companies are started by people who bought a flat, put it on Airbnb, and decided to do it for other people too. That's a fine way to learn the basics. It is a poor way to build something that holds up at scale, because it starts from a single property and works outward, rather than starting from a system and working in.

My own route in was different, and it matters for understanding why Safa is built the way it is.

The first piece is engineering. I studied robotics and mechanical engineering at the University of Sussex. That sounds unrelated to property until you sit in the two worlds side by side. Engineering trains you to distrust intuition and to build systems that produce the same output every time, regardless of who is on shift. A process that only works when the founder is personally watching is not a process — it's a bottleneck wearing a process costume. That instinct never left me.

The second piece is proptech. Time inside a proptech incubator showed me how technology actually changes property operations — and, just as usefully, where it's oversold. A lot of "tech-enabled" property businesses are a spreadsheet and a WhatsApp group with a logo on top. The ones that work treat software as the thing that enforces the standard, not the thing you mention in the pitch.

The third piece is property media. Working in property media gave me access I would never have got as an individual landlord — conversations and proximity to how the institutional end of the market operates, including the worlds occupied by the likes of Berkeley Group, JLL, and CBRE. What struck me wasn't the glamour. It was the discipline. Institutional property operations are run to defined standards: compliance is assumed, reporting is structured, nothing depends on one person remembering to do it.

Then you look at the short-let market, and almost none of that exists. Same asset class — residential property — operated with a fraction of the rigour. The gap wasn't hidden. It was visible from both sides at once: I'd seen how property operations are done properly, and I'd seen how short-let was actually being run. The distance between the two was the opportunity.

The problem with most short-let management

When you look closely, the same three things are broken almost everywhere.

1. Inconsistent standards. No two properties in a typical portfolio are managed to the same level. One flat gets a great clean because that cleaner happens to be good; the one down the road gets a rushed one because someone was running late. There's no defined standard the operation is held to, and no system enforcing it — so quality is a function of who happened to turn up that day. Guests feel this immediately, and it's the single most common reason a property's reviews drift downward over time.

2. Landlords underserved. Most management companies are organised around guest acquisition, because that's where their revenue comes from. The landlord — the person who actually owns the asset and pays the management fee — comes second. Communication is thin. Income reporting is opaque: a number lands in your account each month with little explanation of the occupancy, the rate achieved, the costs deducted, or why this month differs from the last. The landlord is treated as a supplier of inventory rather than a client of the service. That's backwards.

3. Guests paying hotel prices for a product that wasn't thought through. Short-let has spent a decade marketing itself as a better alternative to hotels, and at the same time charging hotel-level nightly rates. But a hotel earns those rates with discipline behind it — consistent housekeeping, a real check-in process, someone accountable when something breaks at 11pm. A lot of short-lets borrow the pricing and skip the discipline. The guest pays for "luxury" and gets a flat with a lockbox, a half-stocked kitchen, and no one answering the phone.

Those three problems are the same problem seen from different angles. There's no system, so standards slip, landlords are left in the dark, and guests get an inconsistent product. Fix the system and all three improve together.

What we decided to build

The easy version of this business is to take on every property that's offered, manage them all loosely, and grow the door count as fast as possible. That model is everywhere, and it's exactly the model that produces the three problems above. We deliberately didn't build it.

Safa is a selective, systems-driven operation. That means a few specific commitments.

• Selective intake. We don't take every property. A flat has to clear the basics — the right location for genuine year-round demand, a lease that permits short-let, and a realistic path to performing — before we'll manage it. Saying no to the wrong properties is what protects the standard on the right ones.
• Compliance from day one, not when forced. The insurance, the redress-scheme membership, the data registration, the client-money safeguarding — all of it in place before the first property, not bolted on after a complaint.
• A professional management agreement. A real contract that sets out what we do, what we charge, and what the landlord is entitled to expect — not a handshake and a hope.
• Dynamic pricing. Rates set against actual demand, day by day, rather than a flat nightly number that's too high midweek and too low when an event hits.
• Quality-controlled cleaning. A defined turnover standard with checks behind it, so the clean doesn't depend on who's on shift.
• Transparent income reporting. The landlord sees occupancy, rate, costs, and net — the full picture, not just the figure that lands in their account.

None of that is exotic. It's the institutional discipline I saw on the property-media side, applied to an asset class that mostly operates without it. The engineering background is the throughline: build the system first, define the standard, and make the operation produce it every time — systems thinking, not intuition.

What Safa looks like now

Here's the honest current state, because the rest of this article isn't worth much if this part isn't straight.

We are early stage and pre-revenue. We don't have a portfolio of live properties or years of trading history.

What we do have is the infrastructure most operators only build once they're forced to. The compliance stack is fully in place: £1m Professional Indemnity and £5m Public Liability cover via HDI Global Specialty SE, membership of the Property Redress Scheme (PRS058526), and ICO registration (ZC138131). Client money is held in a dedicated safeguarded account with an FCA-regulated provider, ring-fenced under the Electronic Money Regulations 2011. (Client Money Protection isn't a legal requirement for short-let operators — safeguarding at source is the equivalent protection for this model.) The website is live, and the management, pricing, and reporting systems are built and ready.

That sequencing is deliberate. Most operators take on properties first and assemble the compliance and systems later, under pressure, after something has already gone wrong. We did it the other way round: the infrastructure that takes most operators years to build is done before the first property. It's less impressive to talk about than a portfolio of doors — but it's the part that actually protects a landlord's money and a guest's stay.

What we're building toward

The goal isn't 200 properties. Scale for its own sake is exactly what produces the loose, standard-free management we set out to avoid.

The plan is a managed portfolio in specific cities where the demand thesis is proven before we commit — not a pin dropped on a map because it sounded promising. Doncaster is the anchor: a market we've researched in depth, with necessity-led, year-round demand and a clear reason to be there. From there the model extends, deliberately, to markets like Leeds, Manchester, and Edinburgh as it scales — each one entered on the strength of its own demand picture, not a growth target.

What we want at the end of it is simpler than a number. A brand landlords recognise because of the standard it holds to, and one guests choose by name because the last stay was good and they expect the next one to be too. You only earn that by being consistent, which brings it back to the system.

Why we're being transparent about being early stage

The alternative is to pretend we've been operating for five years and hope nobody checks. Plenty of new operators do exactly that — borrowed stock photos, vague claims about "our portfolio," testimonials that don't quite name anyone.

We'd rather just tell you where we are. A landlord handing over a property and the income it produces deserves to know exactly who they're working with, including the fact that we're new. The market analysis we publish — the city guides, the yield thinking, the demand breakdowns — is built on real research and third-party data, not on a track record we don't have yet. Where something is modelled, we say it's modelled. Where we don't know, we say we don't know.

That's a slower way to build trust. It's also the only version that survives contact with a landlord who does their due diligence — and the ones worth working with always do. We'd rather earn it slowly than claim it early.

Where we are right now

Safa Residences is open and operating. The compliance stack is fully in place — £1m Professional Indemnity and £5m Public Liability cover via HDI Global Specialty SE, membership of the Property Redress Scheme (PRS058526), and ICO registration (ZC138131), with client money safeguarded under the Electronic Money Regulations 2011. We're a newly launched operator building our first portfolio, and we'll always be straight with you about that.

If you own a property and you're weighing up whether short-let beats your current arrangement, we're glad to talk it through. Not to sell you anything — to explain the model honestly, look at your specific situation, and give you a straight read on whether it's worth doing at all. Sometimes the honest answer is that it isn't, and we'll tell you that too.