Can I Use a Spreadsheet for My REPS Hours?

Yes — but only if it is contemporaneous. The regulations do not require any particular tool, and nothing says you need software. What gets logs thrown out in Tax Court is almost never the format. It is when the entries were created and what is missing from them.

Treas. Reg. Section 1.469-5T(f)(4) is unusually generous on this point. Participation "may be established by any reasonable means," and it says explicitly that "contemporaneous daily time reports, logs, or similar documents are not required." Appointment books, calendars and narrative summaries are named as acceptable.

So a spreadsheet is fine. The problem is what people actually put in one.

What a defensible entry contains

Every row needs to answer four questions, and most spreadsheets answer two.

The fourth column is where cases are won and lost. "Property work, 3 hours" tells an examiner nothing, cannot be corroborated against anything, and looks identical whether you did the work or invented it afterwards.

What actually gets logs rejected

Reconstruction after the fact. This is the big one. In Moss v. Commissioner (135 T.C. 365), the court dismissed what it called "postevent ballpark guesstimates." Once an examiner concludes the log was written after the notice arrived, its evidentiary value collapses — not because reconstruction is forbidden, but because nothing in it can be checked.

Rounded, undifferentiated blocks. In Penley v. Commissioner (T.C. Memo. 2017-65), a calendar with rounded entries that lumped driving in with other activities, without separating them, contributed to the court rejecting the log in its entirety. That is the danger with a sloppy log: it is not scored line by line. One implausible pattern taints everything around it.

Hours that are not work. Several categories reliably get struck:

• Being on call. Moss held that time on call for the rental properties does not satisfy the 750-hour requirement. Hairston v. Commissioner (T.C. Memo. 2019-104) refused to credit hours spent watching contractors work — at best, the court said, he was on call.
• Education. Antonyshyn v. Commissioner (T.C. Memo. 2018-169) excluded logged seminar hours. Studying and monitoring your investments is investor activity under Treas. Reg. Section 1.469-5T(f)(2)(ii), which is excluded from participation.
• Investor-type work. Reviewing financial statements, paying bills and preparing analyses for your own use are named in the regulation as investor activities.

Totals that fight your own records. If your log says you spent 1,400 hours on real estate in a year you also worked a full-time job, an examiner does the arithmetic. Logs that conflict with an employment record or a calendar rarely survive.

Where a spreadsheet genuinely struggles

None of the above is an argument for software. These are the practical failure points, and you can engineer around most of them in a spreadsheet if you are disciplined:

• No timestamps. A spreadsheet records the date you typed, not the date you claim. There is nothing in the file to distinguish an entry made that evening from one made fourteen months later. Cloud version history helps here, and it is worth turning on.
• Nothing enforces the year. It is easy to file a January 2026 activity under the 2025 tab, and nothing objects.
• No prompt for the fourth column. When you are logging at 10pm, "property work" is what you type unless something asks you for more.
• Comparing against other people is manual. For the 100-hour test you have to show no other individual participated more. That means tracking your cleaner's and manager's hours too, and a spreadsheet will not remind you.

The honest summary: the law does not care which tool you use. Discipline is what matters, and the tool only matters to the extent that it makes discipline easier or harder.

If you are starting late in the year

You are allowed to write down work you actually did. The regulation's "any reasonable means" language covers it. What you should not do is invent round numbers.

Anchor every entry to something dated that already exists — an email to a tenant, a contractor invoice, a bank transaction, a hardware store receipt, a calendar appointment, a photo with metadata. Then the log is not your memory, it is an index into evidence that already existed. That is the difference between a reconstruction an examiner accepts and a guesstimate they discard.

And be conservative. A reconstructed log that clears the threshold by four hours invites exactly the scrutiny it cannot survive.

Corroboration is the part people skip

Your log is a claim. What makes it credible is everything around it.

Keep the emails, the invoices, the work orders, the receipts, the bank records, the guest messages. You do not need to attach them to every entry, but you need to be able to produce them for a sample of entries when asked. A log with no corroborating trail is one person's word; a log you can spot-check against six independent sources is a record.

That is the standard to build for — not "did I write it down," but "can I show it happened."