A reverse-engineered approach to franchise site selection — starting with the financial goal, not the map.
Every lease is a bet. Most operators place it backwards.
The standard process starts with a map. A broker circles a few suburbs that "feel right," pulls a demographic heatmap from a free tool, and presents three options with population counts and not much else. The lease gets signed on vibes dressed up as data.
Lokko starts from the other end. Before we look at a single suburb, we ask one question:
to make a 10-year lease at this rent defensible, and which suburb can actually deliver it?
This is Territorial Intelligence: a financial guardrail model that works backward from the number your board needs to see, through population, capture rate, and membership revenue, to a maximum defensible rent — before you fall in love with a site. Every recommendation in this paper is built on that logic, illustrated using a fictitious brand we call "Apex Gym."
A poor site decision doesn't fail quietly. It shows up as a decade of under-target revenue, a rent line the P&L can't carry, and a franchisee who blames head office. The decision deserves the same rigour as any other six-figure capital allocation — but it rarely gets it.
Why the traditional approach keeps failing operators
None of this means the traditional inputs are wrong. Population, income, and traffic data are necessary. They're just not sufficient. What's missing is the translation step: turning demographic layers into a dollar figure a landlord, franchisee, and board can all interrogate.
Apex Gym is evaluating a 1,400 sqm full-service format and needs to know, before it looks at a single lease listing, what rent it can actually afford to pay. Here is the five-step calculation Lokko runs for every candidate suburb.
That final figure — $194/sqm — is the number that matters most in the negotiation. It's not a benchmark pulled from a market report. It's the maximum rent Apex Gym's own catchment can support, before profitability is compromised. If a landlord is asking for $230/sqm, Lokko has already told Apex exactly how much negotiating room exists — and how much doesn't.
This is the reverse-engineering principle in practice: financial ceiling first, site selection second.
This is what the question on the cover actually costs. A ten-year lease is a commitment to roughly $15M of cumulative revenue (modelled) — and it gives the operator a success metric before signing: if the chosen suburb can't project the annual revenue the model requires, the ten-year term isn't ambitious. It's financially imprudent.
Every number in this walk-through is modelled from underlying demographic and revenue assumptions calibrated to the client's business — not asserted. Revenue and rent figures are always labelled (modelled) in client deliverables, because the value we sell is the analysis, not a claim to measured fact.
The Lokko model uses official ABS Statistical Area Level 2 (SA2) boundary polygons — not shortcuts to the centroid. That distinction is not cosmetic. SA2 centroids can sit kilometres from where people actually live: a suburb shaped by coastline, rail corridors, or urban growth boundaries will have its geographic centre in a paddock or a car park, not in its population core.
When a site is matched to a centroid rather than a polygon, the demographics assigned to it belong to the geometric middle of the suburb, not to the ground the site stands on. The result is a different suburb classification, a different score, and — critically — a different strategic recommendation.
When Apex evaluated a candidate site in Melbourne's western corridor, the two matching methods produced opposite advice.
Same coordinates. Same site. Same date. Two completely different strategic recommendations. The centroid method assigns the site to the wrong SA2 — importing a competitor from the adjacent boundary — and flags a contested market. The boundary polygon places the site correctly, with real, census-verified demographics behind it. But because this SA2 mixes residential and agricultural land, the higher score is a reason to look closer at the ground, not a green light on its own.
This is not an edge case. In high-growth corridors where suburb boundaries are drawn along roads, rivers, or rezoning lines — exactly the areas where franchise expansion is most active — centroid shortcuts misclassify sites routinely. Lokko's point-in-polygon matching eliminates the error at source.
All revenue figures in this example are modelled estimates derived from population-based demand modelling using ABS 2021 Census data. They are not representations of actual or likely earnings.
Type the suburb name into a consumer map app and it shows the gazetted locality — a mostly agricultural precinct that reads as empty and would make most operators dismiss it on sight. The ABS statistical boundary for the same name draws the line differently: it captures a residential strip of roughly 17,000 people concentrated in the north, hard against Werribee town centre. That strip already competes with the operators sitting in Werribee town centre — which is exactly why Lokko's land-use guard flags this aggregate for site-level verification rather than clearing it automatically. The model tells you where to look; the ground decides.
The gazetted locality most map apps display for "Werribee South" — mostly agricultural precinct, dashed here because it is a legal/postal boundary, not a population boundary.
The ABS statistical area for the same name — its northern strip is where the suburb's 16,810 residents (census-verified) actually live, adjacent to Werribee town centre.
Locality boundary: © OpenStreetMap contributors, ODbL 1.0 (via Nominatim). SA2 boundary: Australian Bureau of Statistics, ASGS 2021 (CC BY 4.0). Both simplified for display; not for surveying use.
The rent ceiling tells you what a suburb can afford. It doesn't tell you whether the suburb is a good fit, or whether someone already owns the demand. That's what the Lokko Opportunity Score is for.
Every suburb in a client's target region is scored from 0 to 100 across five factors:
The output is a ranked, defensible shortlist — not a black box. Every score can be broken back down into its five inputs, and every input can be interrogated by a franchisee, a landlord, or a board member. That transparency is deliberate.
The model informs. You decide.
The Lokko Territorial Intelligence Dashboard: mapping opportunity scores, financial guardrails, and real market constraints in real-time.
Lokko doesn't tell a client where to sign a lease. It tells them, with evidence, which suburbs can carry the revenue their format needs, which have room in the market, and which are already spoken for. The judgement — brand fit, deal terms, timing — stays with the operator. The guesswork doesn't have to.
Lokko's entry point is a $299 Territory Audit — and every dollar of it is credited against your full project if you proceed. It ranks suburbs and surfaces whitespace: a genuinely useful starting shortlist. It is also, deliberately, incomplete.
The Territory Audit does not pair your demand-side rent ceiling against what landlords are actually asking. That comparison — modelled affordability versus real market rents and lease comparables — is where Territorial Intelligence earns its keep, and it's the first deliverable in every Lokko Phase 1 engagement:
This is the difference between a shortlist and a decision. The Territory Audit gives Apex Gym three suburbs worth a closer look. Phase 1 tells them which of those three they can actually afford to sign — before they're sitting across the table from a landlord. And because the $299 is credited against the full engagement, the audit isn't a cost — it's a deposit on the decision.
Book a live demo and we'll run this exact model against your target region, live.