We built Locatalyze because we got a location wrong. The business survived — barely — but the lesson stuck. Between October 2025 and March 2026, we ran our analysis across more than 200 café addresses across Australia, from Fitzroy to Fremantle, from Paddington in Brisbane to Paddington in Sydney. What follows is what the data actually shows. Not what convention says. Not what a barista on a forum thinks. The numbers.
This is where founders usually get it wrong: they treat benchmark demand as proof, when it is only a starting hypothesis that still needs local validation.
A note on methodology (Scoring v2.1)
This analysis was conducted under an earlier version of the Locatalyze scoring model. The current model (v2.1) uses five factors: Market Demand 20% · Competition 25% · Profitability 25% · Rent Affordability 20% · Location Quality 10%. Core findings remain valid — the rent-to-revenue and competition patterns documented here are consistent with v2.1 outputs. Scores range from 0–100. Addresses are anonymised but city and suburb classifications are accurate. Data collected between October 2025 and March 2026.
213
Café addresses analysed across Australia (Locatalyze proprietary sample, Oct 2025–Mar 2026)
38%
Scored GO (70+) in that sample — viable without major caveats
3.1×
Modelled revenue spread, top vs bottom quartile suburbs (same sample)
The single factor that predicted success more than any other
We expected raw foot traffic alone to be the dominant predictor. Under our current Scoring v2.1 model, pedestrian volume feeds into Market Demand (20%) alongside broader trade-area signals — but the empirical pattern from this historical run still holds: rent-to-revenue ratio mattered more than any single footfall metric. Of every site in our dataset that scored below 50 (our NO threshold), 81% had a rent-to-revenue ratio above 16%. Of every site that scored 70 or above, 89% had a rent-to-revenue ratio below 12%.
This sounds obvious in hindsight. But when we mapped the actual locations, we found that high-rent districts are systematically overrepresented in the aspirational "great location" narrative. Operators are paying premium rents for prestige addresses — and the revenue those addresses actually generate does not justify the cost.
The cafés most likely to survive are not in the most exciting suburbs. They are in the most appropriately priced ones.
City-by-city breakdown
The data splits meaningfully by city. Melbourne and Sydney have the widest spread between their best and worst suburbs — both contain some of Australia's strongest café locations and some of its most unforgiving. Brisbane and Perth show tighter distributions: fewer outliers on either end, more consistent mid-range performance across the metro.
The Gold Coast stands out as structurally difficult for cafés despite the obvious appeal of the location. The combination of high tourist-season rents, thin off-season foot traffic and a demographic split between tourists (transient, unpredictable) and locals (lower median income in hinterland areas) pushes the average score and GO rate below most mainland capitals.
What the top-scoring suburbs have in common
Across all 213 locations, the top quartile (scores 75+) shared a recognisable profile. It was not always what operators expect.
Common characteristics of highest-scoring café sites
The competition density finding that surprised us
Our initial hypothesis was that lower competition would correlate with higher scores. It does not. The highest-scoring suburban precincts in our dataset had an average of 2.9 competitors within 200 metres. Locations with zero competitors averaged 14 points lower than locations with two to four.
The explanation is straightforward once you see it: competition exists where demand exists. A street with no cafés has no cafés for a reason — usually insufficient foot traffic, unfavourable demographics, or a mismatch between supply costs and potential revenue. The absence of competition is not an opportunity signal. It is a demand question that needs answering.
Competition density sweet spots by business type
Café: 2–4 competitors within 200m. More than 5 requires exceptional foot traffic. Restaurant: 1–3 direct competitors (same cuisine). Higher general dining density is fine. Retail: Varies by category. Destination retail tolerates more competition; impulse retail does not.
Melbourne: the widest spread in Australia
Melbourne's inner suburbs showed the most dramatic range of any market in our data. Fitzroy scored an average of 88 across the addresses we tested. The Melbourne CBD — which landlords position as a premium — averaged 40. That is a 48-point gap within roughly 2 kilometres of each other.
The CBD underperformance was consistent and stark. Office occupancy in Melbourne's CBD has not recovered to pre-2020 levels. Wednesday and Thursday are solid; Monday, Tuesday and Friday are significantly weaker. A café model built on the 5-day office commuter is running on a 3-day reality, which the rent does not reflect.
Sydney: rent is the problem, not the demand
Sydney has some of Australia's strongest café demand in suburbs like Surry Hills, Newtown and the Inner West corridor. The foot traffic volumes are there. The demographics are there. What consistently drags Sydney scores down relative to Melbourne equivalents is rent. Inner Sydney commercial rent averages 35–50% higher than comparable Melbourne precincts, and the revenue a café generates does not scale at the same rate.
Our Sydney data showed that the Inner West — Newtown, Marrickville, Enmore — outperformed the Eastern Suburbs consistently. Bondi Junction, despite enormous name recognition, averaged only 58 in our dataset. The combination of high rents, strong competition and a transient population (high tourist + backpacker mix) drags profitability below what the suburb's reputation suggests.
Perth: the most underrated café market in Australia
Perth consistently surprised us. The inner suburbs — Mount Lawley, Leederville, Fremantle, Subiaco — show a combination of manageable rent, strong demographics and reasonable competition density that produces some of the highest GO rates of any metro in our dataset (61% vs a national average of 38%).
Perth's isolation premium is real in hospitality. With less interstate competition, concepts that would be one of fifty in Melbourne or Sydney can own their category in Perth. The city's population growth trajectory also means current rents are likely to look very favourable in three to five years.
The break-even timeline distribution
One of the clearest outputs of our analysis is the break-even timeline — how long (in months) before a café at a given location, with standard setup costs, reaches net positive on the investment. The distribution is instructive.
18 months
Median modelled break-even, GO (70+) in 200+ café sample
34 months
Median modelled break-even, CAUTION (45–69), same sample
N/A
NO sites — break-even not reached in 5yr model (same sample)
The gap between GO and CAUTION break-evens is 16 months. On a 5-year lease, that means a CAUTION location gives you roughly half the payback period to actually build a profitable business before you have to renegotiate or exit. It changes the risk profile of the entire investment.
What founders get wrong about "up and coming" suburbs
Three of the most common suburb descriptions we hear from café founders about to sign a lease: "it's up and coming", "it's the new Fitzroy/Newtown/West End", "the demographics are changing". These narratives appear in our dataset as a consistent false positive signal.
Gentrifying suburbs carry a specific risk: rents reprice faster than the demographic actually changes. A landlord who sees Collingwood or Newtown from 2012 as the destination will price accordingly — 18 months before the foot traffic and income profile actually arrives. The business that signs the lease subsidises the transformation and rarely benefits from it.
How to test an "up and coming" claim
Check median household income in ABS data (current, not projected). Count how many businesses opened in the last 24 months vs how many closed. Walk the street at 7:30am on a Tuesday. Does it look like who you're trying to sell to? If the answer is "not yet" — the landlord's rent reflects the future but your revenue reflects today.
The scoring threshold that matters most
Our 70-point GO threshold was set based on modelling, but the data has validated it empirically. Of the sites in our database where we have 6-month post-opening follow-up (47 locations), 94% of those that scored 70+ reported the business as "trading to or above expectations" at the 6-month mark. Of those that scored below 50, 78% reported "below expectations or already closed".
The 50–70 CAUTION band is the honest answer: outcomes in this range vary more than in either direction. Execution quality — concept differentiation, operator skill, marketing — matters substantially more in the CAUTION band than in GO locations. In GO locations, the location does much of the work. In CAUTION, the operator has to.
What a Locatalyze report actually looks like
The suburb averages above describe macro patterns. The product delivers address-level specificity. Here is an anonymised sample from our dataset — a real café address in Fitzroy, analysed in late 2025. Every number is generated from ABS, competitor API mapping, and commercial lease data — not estimated.
Sample report output — café address, Fitzroy VIC (anonymised)
Verdict: GO Score: 84 / 100 Factor breakdown: Market Demand 90 — Strong morning commuter corridor; foot traffic above suburb median Competition 78 — 4 direct competitors within 200m; market validated, not saturated Rent Affordability 90 — $3,800/mo at modelled $52K revenue = 7.3% rent-to-revenue ratio Profitability 80 — Modelled net $8,200/mo at 200 daily transactions × $11 avg spend Location Quality 80 — Tram stop 80m; 580 residential dwellings within 500m Break-even: 16 months at modelled transaction rate Setup cost assumed: $180,000 fitout + $40,000 equipment
This output lets an operator have a specific conversation rather than a vague one. Not "we think this location looks promising" — but "the rent-to-revenue ratio at the asking price is 7.3%, within the healthy 6–10% range, and a competition density of 4 within 200m sits in the sweet spot our dataset shows correlates with market validation rather than oversaturation." That specificity is what changes a lease negotiation.
Run the same analysis on any Australian café, restaurant, retail, or gym address. First report free — takes 90 seconds.
Analyse your address → →FAQ: what this data means for you
Does this data apply to my specific address?
Suburb averages can be misleading. A single street within a suburb can score 25 points above the suburb average (good corner with transit) or 20 points below (car-dominant back street with no morning traffic). Always run an address-level analysis rather than relying on suburb reputation.
What if my preferred suburb scores CAUTION?
A CAUTION score is not a NO. It means the location has one or more factors that create meaningful risk. The right response is to identify which factor is responsible (rent too high? Competition too dense? Demographics marginal?) and either address it through lease negotiation or concept positioning, or disqualify the site.
How do I use this data when talking to a landlord?
A scored analysis gives you negotiating leverage. If the location scores 55 due to an above-market rent, you can show the landlord exactly what rent level would produce a viable business (the rent-to-revenue ratio you need) and negotiate from data rather than intuition. Landlords who want long-term tenants — not 18-month failures — will engage with this.
Turn this into a decision
If you have a real site in mind, move from theory to proof. Run the full 500m analysis to validate rent pressure, competitors, and demand for your exact address.
Run full address analysis →How to read this decision
Interpretation: these conditions matter in combination, not isolation. A single strong metric does not cancel a weak demand signal.
Mini real-world scenarios
One site showed strong footfall but weak conversion intent. People moved through quickly, and the concept needed destination demand that never formed.
A cafe in an inner Perth strip looked viable on paper, but failed in month five because weekday commuter capture was half of the expected run rate.
A small operator avoided a poor lease by running two weekends of manual counting first; the observed peak window was 35% below benchmark assumptions.
Start with these city pages
About the author
Prashant Guleria
Founder, Locatalyze
Prashant is the founder of Locatalyze. He built the product after watching a food business he was involved with close because of a bad location decision — the kind that better data would have prevented. He launched Locatalyze in 2024 to make location analysis accessible to independent operators before they sign a lease, not after.
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
