I stopped reading agent pro formas and it did not cost me anything

I'm still list-building, no closings, and I'd already stopped reading the summaries for a different reason than yours. Two of the three agents I've talked to used gross rent multipliers and one of them quoted a cap rate on a single family house, which told me they'd downloaded a template rather than run a building.

But I want to push on your two inputs. Seller timeline I get, that's genuinely something they can find out and I can't. Sold comps with finish level described, though, is doing a lot of work in your post. In my county the assessor publishes sale prices and a photo, so I can build a comp set myself in an evening. Half the value you're assigning to the agent might be a local data quirk.

My vote is off-market access, because everything else I can either learn or buy. What I can't do from a spreadsheet is be the person a property manager calls when an owner gets tired.

How does an agent find out a seller's timeline without just asking and getting told nothing? Everything here is new to me.

@bramble named the variable that decides this, and it's worth making explicit because it changes the answer by geography. Roughly a dozen states are non-disclosure, meaning sale prices are not published in public records, and in those states the MLS sold field is the comp set. If you're in one, an agent's data access is expensive to replace and hard to work around. If your county assessor publishes prices with photos, you can rebuild most of a comp set yourself and the premium you should pay for that specific input drops. Check what your own county publishes before you rank the four options.

On the pro formas, the failure you're describing is structural. An agent's investment summary is produced in advance of knowing the buyer, so every assumption is a default: 5% vacancy, a blended maintenance percentage, no management line. Those defaults are marketing inputs. Your capex number depends on the age of the mechanicals and the roof and on your own hold period, none of which the agent knows when the sheet is built. Treat the sheet as a description of the property and run the operating numbers yourself.

@rigging, timeline usually comes from the listing agent rather than the seller. Agents talk to each other, and a listing agent will often say what the seller needs in order to get an offer structured to fit, occupancy after closing, a specific date, a clean inspection. That conversation is between licensees and you generally can't have it directly. It's the single most common thing an experienced buyer's agent gets that a self-directed buyer doesn't.

The ranking also depends on whether your constraint is finding properties or closing them. Those are different bottlenecks and the same fee buys different value against each.

The split you predicted is real and I can feel myself crossing it. Deals one and two, the machinery was everything. I did not know what a title commitment was and my agent caught an unreleased second mortgage that would have wrecked my closing. That's worth a full fee on its own.

Eight deals in, MLS gives me nothing. Everything I want is either not listed or listed at a price that assumes a retail buyer with a 5% down loan. So my premium is now entirely on the first call, and the frustrating part is that it's the hardest thing to buy. You can't pay an agent to have relationships they don't have, and the ones who do have them are already feeding two or three investors who close faster than I do.

On the referral network, I'd rank it last and I feel strongly about it. Every contractor an agent has referred me to has been a retail-priced remodeler who quotes flip scope like it's a homeowner kitchen. The lender referrals have been better, though I still shop them. Nobody's referral list beats the crew you build yourself over three projects.