How long should a pre-purchase analysis be?

Long enough to carry the evidence and short enough to be read. Most useful versions run three to six pages: recommendation, rent, expenses, comps, location, risks, and offer price. Attachments can hold the raw comp data for anyone who wants to check your work.

Should I ever recommend against a deal my client found?

Yes, and it is the most valuable thing you will send them. An analyst who has never returned a pass is not filtering. The pass also protects you, because you will manage this property and inherit whatever it turns out to be.

What if the numbers only work at a lower price?

Say so and give the number. A deal that clears at $262,000 and fails at $285,000 is a negotiating instruction, not a rejection. State the price your conclusion holds at and what happens to cash flow above it.

What should a property manager send an investor before they buy?

The pre-purchase analysis is the highest-leverage document a property manager produces. It’s the moment an investor decides whether to spend six figures, and for a remote buyer it’s most of what they’ll ever know about the property.

It’s also, at most firms, an email with three screenshots and a rent guess.

Why does one document beat five attachments?

Because the buyer is going to forward it.

Whatever you send gets read by a spouse, a partner, sometimes a lender, and in the case of an overseas buyer, an advisor in another country. Every one of those readers arrives without the phone call you had. Five attachments and a paragraph of context survive none of that.

One document, self-contained, with your name on it. It should make sense to someone who has never spoken to you.

What goes in it?

Seven sections, and the order matters more than the length of any one of them.

1. The recommendation, first. Buy, buy at a lower number, or pass, plus one sentence of why. Everything after this is the evidence.

2. The rent estimate, as a range. Covered below, and it’s the section that most often does the damage.

3. The expense stack. Every line, with the assumption visible.

4. Sale comps. Address, distance, size, sale date, price, price per square foot, and a note on what makes each one comparable or not.

5. Location facts. Flood zone, school assignment, and anything about the street a photo won’t show.

6. Risks, priced. Each one with a dollar estimate and a mitigation.

7. The offer price your conclusion holds at, and what changes above it.

What does the rent number need to look like?

A range, always, with the leases behind it.

A single rent figure is the most dangerous line in the document. If you write $2,100 and it leases at $1,875, you’ve cost the owner $2,700 a year against their model, and you did it in writing, before they bought.

Write it like this instead: three-bedroom, two-bath comparables within a mile leased at $1,875, $1,950, and $2,040 in the last ninety days. Likely rent $1,950, range $1,875 to $2,040. Then note what would move it within that band: a renovated kitchen toward the top, the busy road toward the bottom.

The width of the range is itself information. A 5% spread across eight recent leases is a market you can underwrite confidently. A 25% spread across three stale leases is a market where the owner should hold a bigger reserve, and they deserve to know which one they’re in.

Which expenses do people get wrong?

Property tax and insurance. Between them they routinely move a deal’s annual expense stack by $4,000, and both are easy to carry over from the seller unchanged.

Property tax. The seller’s tax bill is the wrong number in most of the country and catastrophically wrong in states with acquisition-based reassessment or homestead caps. A Florida seller with a long-held homestead exemption may be paying a fraction of what the buyer will pay from year one. Use the assessor’s post-sale methodology, not the current bill.

Insurance. Get a current quote rather than a percentage of value. In coastal, wind-exposed, and wildfire-exposed markets, premiums have moved enough in recent years that a stale assumption can reverse a deal’s viability by itself. A quote takes a day. A wrong assumption lasts the whole hold.

The rest of the stack is straightforward, and the discipline is showing your assumptions rather than the totals:

Line	Typical assumption	Why show it
Vacancy	5% to 8% of gross rent	Owner may hold a different view of the submarket
Maintenance	5% to 10% of collected rent	Depends heavily on age and prior work
Capital reserve	$100 to $250 a month	Separate from maintenance, and frequently omitted
Management fee	Your actual rate, on collected rent	Omitting your own fee reads badly
Leasing and renewal	Amortized over expected tenancy	Often forgotten entirely
HOA	Actual, verified	Special assessments are worth a sentence

Compute management on collected rent after vacancy, and use the same base everywhere in the document. Mixing gross and collected across sections is how two internally consistent pages disagree with each other.

What makes a comp defensible?

A defensible comp set is one that still contains the sale you wish you hadn’t found.

Drop comps that genuinely don’t belong: no square footage recorded, sales more than six to nine months old in a moving market, distressed or intra-family transfers, and the subject property’s own listing if it surfaces in the pull. Then keep what’s left, including the low one.

Note the adjustment reasoning explicitly. If a comp sold for $19,000 more and has a garage the subject lacks, write that. An investor who can follow the adjustment can accept the conclusion. One handed a number has to take it on faith, and faith is exactly what a remote buyer has least of.

A comp set where every entry supports your conclusion is a comp set someone curated. Experienced investors know this and read it accordingly.

How should the risk section read?

Every risk in the section should carry a dollar figure and a mitigation.

“The roof is older” tells the owner nothing. “Roof is original to 2004 construction, typical remaining life two to five years, replacement quoted locally at $12,000 to $16,000 for this footprint, recommend a credit at closing or a reserve from month one” tells them what to do.

Rank by expected cost, not by how alarming each item sounds. And include the risks that argue against your own recommendation, because the owner will find them eventually and it should not be after closing.

Then close the section by naming your weak data. Four comps instead of ten. An insurance quote you couldn’t get. A rental market you have thinner history in. This feels like undermining the document and it does the opposite: it’s the difference between an analysis and a sales sheet, and investors who’ve seen both can tell instantly.

What should you leave out?

Your commission math. It belongs in the disclosure, not in the analysis.

Appreciation projections presented as income. You can note historical price movement for the area from a published index. Forecasting it as return is where credibility goes.

Anything about the neighbourhood’s people. Describe housing stock, price trends, rent levels, schools by published rating, and commute. Do not characterize who lives there. Steering exposure under fair housing law is real, it applies to written materials, and the safe version is also the more useful one, because an investor needs to know what rents, not who rents it.

Certainty you don’t have. Every number in the document should be traceable to a source, an assumption, or a range. If it’s none of the three, take it out.

If this deal goes badly in eighteen months and the owner rereads this document, does it hold up? If it oversold, rewriting the sentences won’t save it.