There is a block in Melbourne's CBD — Lonsdale Street, between Swanston and Exhibition — where four restaurants have opened and closed in the last three years. Same building, same kitchen infrastructure, different operators each time. Ramen joint. Vietnamese fusion. A café that got good reviews but couldn't make rent. Now a Korean BBQ place that will probably become the fifth closure on that block. The foot traffic is decent. The street looks fine. The nearby towers push enough lunch suits through the door to look promising on a site visit. But every single operator has walked in with capital, signed a lease, and eventually walked out with less than they came with. Meanwhile, three streets over, a cramped 40-seat Italian trattoria has been fully booked every Friday and Saturday night for six years. Same city. Same general neighbourhood. Completely different financial story. When you look at enough data — and we have — you stop being surprised by this pattern. You start to understand exactly why it happens, and the answer almost never has anything to do with the food, the branding, or the chef's technique. It comes down to location economics. And most founders are getting this completely wrong.
The Myth That's Destroying Australian Restaurants
"Just find somewhere with high foot traffic." It's the single most repeated piece of advice in restaurant location selection, and it's killing businesses across Australia. Not because foot traffic doesn't matter — it does. But because foot traffic without context is as useless a metric as a car's speed without knowing whether you're on a highway or in a school zone.
A tourist strip in Cairns might generate 8,000 pedestrians a day. Average spend per head: $12. Table turn rate: 1.2. A business district in Parramatta might generate 3,500 pedestrians a day. Average spend per head: $24. Table turn rate: 2.8 per lunch service. The Cairns location has more than double the foot traffic. The Parramatta operator makes more money. Significantly more. And yet, if you'd asked both founders during their site selection process what they were looking for in a location, both would have said "high foot traffic."
The conversation that Australian restaurant operators need to be having isn't about foot traffic volume. It's about foot traffic quality, demographic match, conversion probability, and — most importantly — whether the economic fundamentals of the specific address can ever produce a profitable business at their format and price point. These are harder questions. They require data rather than a site visit on a busy Friday afternoon. They require intellectual honesty rather than enthusiasm. And they're the questions that the top quartile of restaurant operators in our data consistently asked before committing to a location.
The number that matters more than foot traffic
Rent-to-revenue ratio. Industry benchmark for viability: 6–10% of gross weekly revenue. If your all-in weekly rent (including outgoings) exceeds 10% of your realistic weekly revenue at 70% capacity, the location is structurally unviable regardless of foot traffic. Calculate this before you visit. Calculate it before you fall in love with a space. Calculate it before the landlord starts applying pressure.
Variable 1: The Rent-to-Revenue Ratio Nobody Calculates
The most preventable failure in Australian hospitality isn't poor service or bad food. It's operators committing to rents that their format can never physically support, based on absolute dollar figures that sound manageable rather than percentages that reveal structural impossibility.
Here's how the trap works. Operator finds a space in a visible location. Rent is $5,800 per week including outgoings. That number sounds large but manageable — $25,000 a month, about $300,000 a year. The operator does some back-of-napkin math, feels confident about the concept, and signs. Then trading reality arrives and the math that was never done properly starts to reveal itself.
The restaurant seats 55 people. Average lunch spend: $26. Average dinner spend: $45. Table turn rate: 1.6 at lunch, 1.1 at dinner. The operator runs 14 services a week — lunch and dinner Tuesday through Saturday, lunch only Sunday. At full capacity across all 14 services, weekly revenue ceiling: approximately $47,000. Realistic trading at 70% of ceiling: $32,900. Rent as a percentage of realistic revenue: 17.6%. That's almost double the industry's safe maximum. The business model was broken before the first cover was served.
These ratios aren't arbitrary. They're derived from what's left after food cost (typically 28–33% of revenue), labour (32–38%), and other overheads (12–15%). At a 10% rent ratio with best-case margins across every other cost category, you're looking at 2–8% net. At a 17% rent ratio, you're operating at a guaranteed loss on every standard-week of trading. Every. Single. Week.
The calculation that saves businesses is simple: (Seat count × average table turns per service × average spend × number of services per week) × 0.70 = your realistic weekly revenue. Divide your all-in weekly rent by that number. If the answer is above 0.10, you need to either negotiate the rent down significantly or find a different location. If the answer is above 0.15, the location will not generate a profitable business for your format regardless of how well you execute.
Variable 2: The Anchor Proximity Effect
Every business location exists within an ecosystem of neighbouring businesses. The composition of that ecosystem either compounds your traffic or quietly destroys it — and most operators assess this ecosystem with a ten-minute walk around the block rather than any systematic analysis.
The single most underrated driver of consistent, reliable weekday lunch revenue in Australian CBDs and metro areas is proximity to large, stable office buildings with 300+ workers. Not shopping centres (too variable, too dependent on anchor tenant performance). Not universities (high volume, low average spend, collapses in semester breaks). Not hospitals (complex demographic, time-constrained, heavily habitual). A boring, unglamorous commercial tower full of accountants, HR departments, and mid-level management is a gift to the right food and beverage operator.
Why? Because office workers generate the most predictable, highest-frequency dining behaviour in the market. They eat lunch within a 400-metre radius of their workplace five days a week, fifty weeks a year. They have a spend threshold they're comfortable with ($18–$28 for lunch in most Australian metro markets). They develop habitual loyalty quickly once a format suits their constraints. And they generate dinner trade opportunities through after-work drinks and team dinners that residential catchments rarely produce with the same reliability.
The anchor business combinations that predict high performance:
The most dangerous proximity signal — one that appears repeatedly in the autopsies of failed restaurant locations — is being situated next to a complementary business that generates foot traffic but doesn't convert to your format. A high-end florist. A jewellery store. An interior design showroom. A premium gym. These businesses attract people in a browsing, aspirational frame of mind who are not necessarily in a dining decision moment. The foot count looks reasonable. The conversion never materialises.
Variable 3: The Competition Paradox
Every operator has an intuition about competition. "I want to be somewhere with no direct competitors" or "I want to be in the restaurant precinct where there's already dining culture." Both instincts have data behind them — but only in specific contexts, and most operators apply the wrong strategy for their format.
For fast casual and QSR formats, competitive clustering is demonstrably positive. The aggregation of multiple food options in one precinct creates a destination for decision-fatigued consumers — people who know they want to eat but haven't committed to a specific cuisine. The precinct drives collective foot traffic that no individual operator could generate alone. Being in a food court or an eat-street cluster with 12 other operators is not a disadvantage for a counter-service rice bowl concept. It's a structural advantage that the individual concept benefits from without having to pay for it.
For experience-led, sit-down, premium-dining concepts, the calculus inverts. If guests can see a directly comparable restaurant from your front window — similar cuisine, similar price point, similar positioning — you are in a discretionary choice competition at the point of the dining decision. Every Friday night when someone is deciding between your two options, you are splitting potential covers rather than capturing incremental demand. Fine dining concepts and experience-led mid-range restaurants consistently perform better with geographic distance from their direct competitors.
The competition rule most operators get backwards
Cluster if you are: high-volume, fast casual, commodity food, QSR, or in a cuisine category that benefits from aggregation (Asian food courts, taco strips, etc.). Differentiate geographically if you are: experience-led, fine dining, mid-range sit-down, wine-led, or building a destination rather than intercepting a decision. Applying the wrong strategy for your format costs 15–30% of potential revenue in competitive markets.
Variable 4: The Demographic Decay Trap
This is the most expensive variable to get wrong because it's a slow failure. You open in a suburb that's clearly trending. The demographic indicators look encouraging. Your first eighteen months are decent — not spectacular, but the trajectory feels right. Then something shifts. The demographic is changing, but not in the direction your format serves. Or the transition is happening, but two years behind the pace your lease assumed.
The classic Australian version of this plays out in inner-suburban gentrification corridors. An operator opens a mid-range concept with a $42 average spend in a suburb where the current residential population has an average lunch budget of $18. The operator's rationale: the area is gentrifying, the demographic is arriving. The problem: "arriving" and "already here in sufficient numbers to sustain your business" are profoundly different statements separated by a gap that might be 2–4 years. In a 5-year lease, being 2–4 years too early is approximately the same as being wrong.
The demographic timing trap has a specific signature in the data. The business does well on weekends when the early-adopter, higher-income residents seek out new openings. It struggles on weekdays when the mass of the current catchment — who don't match the format — doesn't materialise. The operator interprets this as an execution challenge and doubles down on marketing, service improvements, and menu refinement. None of these interventions address the structural issue, which is that the weekday catchment isn't there yet. The business eventually closes, and eighteen months later a very similar concept opens in the same location and trades successfully because by then the demographic has actually arrived.
2–4 yrs
Typical lag between suburb "trending" and demographic viability for premium formats
68%
Of "wrong timing" restaurant closures occurred within 36 months of opening
$340k
Average financial loss for operators who misjudged demographic timing in Australian inner suburbs
Variable 5: Transport Access and the Last 200 Metres
In Sydney and Melbourne CBDs, transport access to a restaurant is barely a consideration — the car has already been left at home and the decision framework is entirely pedestrian. But the moment you move outside the inner city into suburban and regional Australia, transport access becomes one of the most significant predictors of trading performance, and it's one of the most consistently underweighted variables in operator site selection.
Specifically: distance from the nearest public transport stop matters enormously in markets where public transport is a realistic travel option, and dedicated parking availability matters enormously in markets where it isn't. Operators in outer-suburban areas situated more than 200 metres from the nearest bus stop or train station and with fewer than 15 dedicated or readily accessible parking spots consistently underperform comparable operators with better transport access by 18–26% on weekly revenue. That gap compounds across 52 weeks. Over a five-year lease it can represent $500,000 or more in cumulative revenue difference.
The nuance: the right transport variable to assess depends on the market. In inner Sydney or Melbourne, the question is about pedestrian walkability from public transport hubs. In middle-ring or outer suburbs, it's about car parking and whether the format's evening trade can realistically function without a sober travel option nearby. A wine-forward dinner restaurant without public transport proximity in a suburban market has a structural ceiling on its evening trade that doesn't appear in any surface-level location assessment.
The Five Most Common Location Mistakes in the Data
What separates the operators who fail from those who don't:
- 1
Signing a below-market rent without understanding why it's below market. Below-market rent in a genuinely viable location gets bid up immediately. If it's sitting available and cheap, a development project is about to disrupt your foot traffic, a major nearby employer is relocating, or the previous tenant failed for reasons the landlord isn't volunteering.
- 2
Conducting site visits exclusively during peak periods. A Saturday morning brunch assessment of a café location tells you almost nothing about Tuesday morning trade, which is the trade that determines whether the business survives its lease. Visit at 7am on a Monday. Visit during a rainy Tuesday lunch. Visit on a school holiday weekday. Visit in the dead week between Christmas and New Year. The gap between peak and trough is the single most important commercial characteristic of many food and beverage locations.
- 3
Ignoring the development application pipeline. Council DA registers are publicly accessible, legally required to be maintained, and almost never checked by first-time restaurant operators. A construction project that starts in month three of your lease can reduce your foot traffic by 40% for a year. A competing food precinct that received planning approval six months ago and breaks ground after you sign is now your commercial problem. This information is free and public. It takes two hours to check. Most operators never do it.
- 4
Calculating rent viability at capacity rather than at 70%. Nobody runs a restaurant at capacity every service. The rent-to-revenue calculation at 100% occupancy is a fiction. The calculation at 70% — which is a reasonable estimate for a well-run, established operator — is the one that determines whether the business model works in the real world.
- 5
Treating the personal guarantee as a formality. Most Australian commercial leases include personal guarantee provisions. If the business fails with two years remaining on a $5,500-per-week lease, you personally owe $572,000. That's not a business loss. That's a personal financial catastrophe. The guarantee provision isn't a formality. It's the clause that determines the worst-case scenario, and most operators sign it without properly modelling what the worst case looks like.
What the Top Quartile Did Differently
Across the location data we've processed, the operators who built sustainable, profitable restaurant businesses in Australia shared a specific set of pre-commitment behaviours that distinguished them from those who struggled or failed. These weren't personality differences or talent differences. They were analytical discipline differences.
They built revenue models before falling in love with a space. Specifically: they calculated their revenue ceiling (seats × turns × spend × services), their realistic trading figure (ceiling × 0.70), and their maximum viable rent (realistic trading × 0.10). They arrived at a rent number before entering negotiation rather than negotiating from the landlord's quoted figure downward.
They assessed competition at a 2km radius, not a visual radius. This meant identifying not just the restaurants they could see from the front door but every food and beverage operator within a 2km walk that competed for the same meal occasions. They mapped competitive density by meal occasion type rather than by cuisine, which gave them a more accurate picture of the real competitive pressure they would face.
They looked backward and forward on demographics. Backward: what is the current income and spending profile of the specific catchment? Forward: what does the development pipeline suggest about how this will change in the next 3 years, and does my lease term survive the gap if I'm early? They distinguished between aspirational demographic projections and current commercial reality.
And they checked the development pipeline — not the general area's development trajectory, but the specific planning applications for addresses within one kilometre, the construction timeline on approved projects, and the ground-floor tenancy types in buildings under construction. This is tedious work. It's also free, publicly available, and consistently reveals information that changes commercial decisions.
The analytical framework that separates profitable operators
Step 1: Calculate rent ceiling before visiting any space. Step 2: Visit only locations where the rent-to-revenue math works. Step 3: Map competition within 2km by meal occasion, not cuisine. Step 4: Verify demographic current reality vs. aspirational trend. Step 5: Check the DA pipeline within 1km. Step 6: Model a bad month. If the business can't survive 2 consecutive bad months, the rent is too high.
The Real Reason Lonsdale Street Keeps Cycling Through Operators
The block we opened with — the one that has seen four operators come and go in three years — has a specific, identifiable problem that is visible in publicly available data and has been visible for years. The all-in weekly rent for ground-floor hospitality on that specific section of the street sits between $6,800 and $7,400 per week. The pedestrian count during weekday lunch peaks at approximately 2,400 per hour, which is decent but not exceptional for inner Melbourne. The office building density within a 150-metre radius is moderate.
For a 60-seat casual dining concept at an average lunch spend of $26 and an average dinner spend of $44, running 16 services per week at 70% occupancy: realistic weekly revenue of approximately $27,000. Rent as a percentage of that: 25–27%. That's two to three times the viable maximum for this format. The location is not bad — it has decent traffic, reasonable visibility, and a fine physical footprint. It is simply mispriced for casual dining. The economics have never worked and will not work until the rent drops significantly or the format changes to one capable of generating the volume that makes a 25%+ rent ratio survivable (a high-volume QSR, perhaps, or a format with significantly higher average spend).
None of the four operators checked this math before signing. The Italian trattoria three streets over, which has been successful for six years? All-in weekly rent: $3,100. Sixty seats. Same price point. Rent-to-revenue ratio: approximately 8%. The math works. It has always worked. That's why it's still open.
Run the rent-to-revenue calculation on any Australian address before you visit. Know the math before you fall in love with the space.
Analyse your location → →About the author
Prashant Guleria
Founder, Locatalyze
Prashant has spent years at the intersection of commercial real estate data and business performance analytics across Australia. He built Locatalyze after watching smart operators make the same preventable location mistakes repeatedly.
Tools first — then a full report for your address
Free rent, viability, and break-even checks. Upgrade when you are ready for competitors, map, and numbers for a specific site.
No signup required for tools
