Do not treat reserves as a guarantee against extra contributions. The useful questions are what the reserve is intended to cover, its current position, planned works and how unexpected costs are allocated among owners.
For the townhouse, check whether it still sits within a managed community with common charges and owner obligations. “More control” may be relative rather than complete, depending on the particular property and community rules.
That also means the model needs three buckets, not two: private costs, community or building costs, and operating or letting costs. Otherwise charges included for one property get accidentally duplicated or omitted for the other.
Vacancy risk should include time between tenants plus any period needed for repairs or furnishing. A unit can have good headline demand and still lose income through frequent turnover.
Tenant demand is very location-sensitive. Compare nearby transport, daily amenities, schools where relevant, and the actual competing stock. Property type alone will not tell you which unit rents faster.
Also separate rentability from rent level. A townhouse may command more total rent but still deliver a weaker yield at the same purchase price once maintenance and vacancy assumptions are applied.
Conversely, a small apartment’s gross yield can look attractive until management, service charges and furnishing replacement are deducted. Use net income before financing, then layer financing on separately if applicable.
I would request recent bills for both candidates rather than use generic estimates. Actual cooling, common charges, insurance and maintenance records are more useful, provided you adjust for how the previous occupant used the property.
A viewing checklist could mirror the spreadsheet: every visible component gets an owner, expected maintenance cycle and payment route. If nobody can explain who pays, flag it as uncertainty rather than entering zero.
For resale, count direct substitutes currently available in the same development or nearby community. Numerous identical units can make price competition obvious; a distinctive townhouse may be harder to value but less directly interchangeable.
Distinctive can also mean illiquid. A highly specific layout or location may appeal to fewer buyers. I would not award the townhouse a resale advantage without comparing transaction evidence for the exact areas.
Agreed. The better stress test is time: if a sale took longer than expected, which property could you afford to hold vacant? Carrying costs during the exit period matter as much as the eventual price.
There is another practical issue: access for repairs. Apartment work may involve building procedures and shared systems; townhouse work may require finding and supervising contractors directly. Decide which kind of friction bothers you less.
The matrix now needs a confidence column. An annual charge supported by records is different from a seller’s broad estimate. Apply a wider contingency wherever documents, responsibility or future pricing remain unclear.
Yes, and do not hide uncertainty inside one blended contingency. Keep separate allowances for common-cost increases, private repairs, vacancy and furnishing. You will then see which assumption changes the decision.
What is your intended holding period? Over a short period, purchase and resale friction may dominate. Over a longer one, recurring management charges, component replacement and changing tenant demand become more important.
Holding period also affects the value of control. Choosing materials and timing repairs matters more when you expect to keep the townhouse. For a shorter hold, operational simplicity may deserve extra weight.
I would challenge the idea that fewer decisions always means lower risk. With shared management, you may have less information before costs change. Ask how budgets, planned works and owner communications are handled in that building.
But more decisions can produce deferred maintenance, especially if the owner is remote. The townhouse case should include paid inspections and coordination if you will not be nearby. Otherwise the workload assumption is unrealistically low.
That returns to my first question: owner-occupied versus rented, and local versus remote ownership. Those four combinations could reverse the result even with the same properties.