Catchment Research
Part of Retail location and catchment research
Analysing a store catchment from available evidence
Build a working store catchment from customer origins, ABS local data, access routes and competing stores without overstating what the data shows.
To analyse an existing store's catchment, start with where recorded customers came from. Then test that pattern against access, nearby alternatives and gaps in the records. The output is a working area with its uncertainties, not a circle drawn at an arbitrary distance.
Build a customer-origin view
Choose a trading period that represents ordinary operation and define which purchases it includes. Check what the origin field means: home, delivery and loyalty-account addresses may differ. Group usable origins into areas broad enough to avoid exposing individual customers.
Record sales, transactions and identifiable customers separately; frequent purchases by one customer can produce many transactions. If origins are incomplete or mainly from loyalty members, report their coverage. Do not assume the mapped pattern represents all buyers.
Map areas contributing recorded customers, then inspect routes into the store. A compact cluster may involve a difficult crossing; a more distant area may have a direct road or transport connection. Mark a nearer core, an outer area with meaningful recorded custom, and places where evidence is sparse. These describe observed customers, not fixed limits on future demand.
Add people the sales record misses
Use Australian Bureau of Statistics (ABS) Census data to describe residents in relevant published areas. The General Community Profile is based on usual residence.
The ABS publishes annual estimated resident population by Statistical Area Level 2 (SA2) through 2025. That series can update resident numbers, but it does not update every Census characteristic or reveal shopping habits. An SA2 may extend beyond practical store reach. Where only part is accessible, do not assign its entire population to the catchment as though the boundary were exact.
| Evidence gap | Practical response |
|---|---|
| Many purchases lack a usable origin | Report coverage and avoid projecting the mapped pattern to all buyers. |
| A published area crosses the likely trading boundary | Show the area as partial or uncertain. |
| Customers may work or visit nearby | Use place-of-work or movement evidence as context, then check store use separately. |
| A nearby branch serves the same households | Examine both stores before claiming additional demand. |
Customer Origin Data vs. ABS Census Data: Key Differences in Catchment Analysis
- Source of Data
- Sales transaction records (loyalty, delivery, home address)
- Scope
- Only customers who made purchases and provided a usable origin
- Coverage Limitation
- May miss non-loyalty shoppers, occasional buyers, or those without recorded origin
- Supplementary Source
- ABS General Community Profile – residents by usual residence (SA2 level)
- Key Use Case
- Estimate total potential demand in areas with sparse sales data
Pros and Cons of Using ABS Census Data in Catchment Analysis
- ProsProvides reliable population estimates by SA2 up to 2025; reflects usual residence
- ConsDoes not capture shopping behaviour or movement patterns; SA2 boundaries may exceed practical catchment reach
Challenge the boundary
Locate competing stores and the retailer's own sites that serve the same shopping task. Compare practical journeys, hours and range. Ask where a plausible customer could shop without passing the store under review. A recorded origin outside the initial catchment may reflect an occasional or specialist trip; one transaction is not enough to redraw the boundary.
The useful output is a map and a short methods note: trading period, included channels, meaning and coverage of origins, geographic units, core and outer areas, access barriers, competing options and gaps in demand evidence. For a proposed store, comparable stores can inform a starting hypothesis, but differences in access and competition must be stated. The proposed site has no observed customer origins of its own.
