Real estate brokerages and teams

Three documents set your buyer-side pay. Nothing reconciles them until escrow.

Bay Area brokerages and producing teams, bought by the broker of record or the director of operations. Individual agents are not the buyer here: every build needs a book of closed transactions to run against, and one agent does not have one.

The loop that bleeds

Before August 2024 the listing broker published buyer-side compensation on the multiple listing service and it flowed at close. Now it is negotiated in three separate places: seller to listing broker in the listing agreement, buyer to buyer’s broker in the Buyer Representation and Broker Compensation agreement (BRBC), and whether the seller offers a concession at all, argued inside the offer. No system compares the three numbers to each other.

California added a clock on top of that. AB 2992 caps a buyer representation agreement at three months with no auto-renewal, and every renewal has to be written, dated, and signed. Agreements expire silently mid-search, so a long-cycle buyer can go to contract on a lapsed document. The settlement separately requires a signed agreement before touring, which is earlier than the statute’s backstop, and agents are still showing first and papering later.

Both failures surface in escrow, when the leverage to fix them is gone. On a $2.5M side at 2.5 percent the gross commission is $62,500, so half a point of gap is $12,500 that the buyer funds or the agent absorbs. The regime is two years old. There is no institutional muscle memory behind it and no incumbent has built the reconciliation, which is the window this page is about.

$12,500: Compensation gap on a $2.5M side when the BRBC says 2.5 percent and the seller concedes 2.0. The buyer funds it or the agent eats it.
90 days: Statutory cap on a California buyer representation agreement, counted from the day after the last signature, with no auto-renewal and every renewal written, dated, and signed.
$15,625: One outbound referral fee at 25 percent of a $62,500 receiving side, never invoiced because the deal closed inside a brokerage you cannot see.

Where to start

Three places this pays for itself first.

Each one is fixed scope, ships in weeks, and either recovers money already earned or removes the thing blocking the work after it. Start wherever it hurts most. None of them requires you to sell more or change what you do.

Outbound referral fee recovery

What breaks

An agent introduces a client to an agent in another market, tags it in the customer relationship management system, sets a task, and the trail goes cold. The deal closes six to eighteen months later inside a brokerage you cannot see, and nobody issues an invoice. The customary fee is 25 percent of the receiving side’s gross commission, taken off the top before the receiving broker’s split. It is pure margin, and it needs only an active license and a relationship.

What gets built

Three things a customer relationship management tag does not do: generate and route the signed referral agreement at the moment of introduction, monitor for the referred deal closing, and produce the invoice and the reconciliation when it does. The monitoring is the hard part and the whole point, because the closing event lives in another brokerage’s back office in another market. The build is small: a referral register, a document template, a recurring status check against the receiving agent and public sale records, and an invoice trigger.

What already solves this

Follow Up Boss, BoldTrail, and Lofty all support tagging a contact as referred out and setting a task. That is the incumbent, and it is exactly what fails, because a task set today does not know a deal closed fourteen months from now. The back office platforms, meaning Lone Wolf, BoldTrail BackOffice, and Loft47, compute referral fees correctly on inbound deals where the brokerage is disbursing and can see the closing, and have no visibility into an outbound close at a brokerage they do not operate. Relocation networks track their own flow inside their own portals, which is a small slice of a team’s sphere referrals. The referral agreement form already sits in zipForm, so the paperwork is a solved commodity and this build should not pretend otherwise. What is absent is the register, the close detection, and the invoice trigger.

Opens onto: Inbound referral service-level tracking, commission disbursement ordering checks, and then the database health work behind everything that comes out of your sphere.

The close happens where you cannot see it. That is the whole build.

61%

Share of buyers who came from a referral or a past relationship, against about 66 percent of sellers. The pipe feeding outbound introductions is the same database that produces your best business.

NAR Profile of Home Buyers and Sellers, 2025

BRBC lifecycle and compensation guard

What breaks

A buyer signs a BRBC at 2.5 percent in March. In August the seller concedes 2.0. Nobody compares the two numbers until the commission disbursement authorization is cut, and by then the conversation with the buyer is a request for a check at closing. Separately, that agreement quietly passed 90 days in June, so on a hard read the brokerage may hold no enforceable claim on the side at all.

What gets built

One register per buyer client holding three numbers and three dates: the BRBC compensation rate and its expiry, the seller’s offered concession on each offer written, and the amount disbursed at close. It gates showing scheduling on a signed agreement, alerts before the 90-day expiry with a compliant written renewal ready to send, computes the dollar shortfall at offer drafting so the buyer conversation happens before contract, and blocks compensation terms that are not specific and objectively ascertainable, which is what “as offered by seller” fails.

What already solves this

zipForm, Glide, and SkySlope all generate and store the BRBC and the short-form touring agreement, and Glide runs a good California buyer flow, so form generation is an input here rather than the product. Transaction management tracks document presence on a checklist without reading the compensation field or comparing it to anything. ShowingTime and Aligned Showings schedule showings and have no concept of a buyer agreement, so the before-touring gate does not exist in the rail where it would bite. Back office platforms will faithfully disburse a number lower than the BRBC promised, because nobody told them what was promised. One caveat worth stating before you buy: this window closes. The regime is new and a transaction management vendor will build this eventually, so treat it as an engagement now rather than a platform bet.

Opens onto: The same document-data plumbing carries pre-review file linting and the disclosure defensibility audit, which is the compliance engagement a broker of record funds out of their own liability budget.

Three numbers, three dates, one comparison nobody is running.

2.43%

Average buyer-agent compensation in 2025, roughly flat against pre-settlement levels. The money did not compress, so there is still a full side to lose on a lapsed document.

Post-settlement commission trend data, 2025

HOA packet risk extraction

What breaks

A large share of Bay Area entry-level and urban inventory sits in a common interest development. The Civil Code §4525 packet contains everything needed to see a dead deal coming: the reserve study, the budget, the assessment disclosure, the minutes, the insurance summary, the litigation disclosure. It is long and dense and almost nobody reads it. The party who does the analysis is the lender, through the condo questionnaire, around day 25. When the project comes back non-warrantable the financing dies a week before close with no fallback and no time to find a portfolio lender.

What gets built

Parse the §4525 bundle the day it arrives and produce a one-page read: reserve funding as a share of budget, current and anticipated special assessments, critical-repair language anywhere in the minutes or the reserve study, pending litigation, owner-occupancy and single-entity concentration, master insurance exclusions and deductible against unit-owner policy requirements, and whether a current SB 326 exterior elevated element inspection exists. Pre-answer the Fannie Mae 1076 and Freddie Mac 476 fields from the same source. This does not replace the lender’s determination. It runs the same test three weeks earlier, while the buyer still holds a contingency and the seller still has options.

What already solves this

HomeWiseDocs and CondoCerts are delivery platforms. They collect the association’s documents, charge the statutory fee, and hand over the bundle. They do not read it, and neither do the management companies that supply it. Fannie Mae’s Condo Project Manager and the eligibility lists are lender-facing, consulted when the questionnaire comes back, which is the late timing that causes the problem in the first place. SkySlope and Dotloop store the packet as a PDF on a checklist item. The limit here is closer than anywhere else on this page: an agent can paste a reserve study into a general chatbot and get a decent summary. What that does not give you is the encoded eligibility ruleset, the SB 326 date check, the insurance reconciliation, and consistency across a hundred-page multi-document bundle.

Opens onto: The same extraction plumbing runs the seller disclosure package, and a brokerage that accumulates parsed packets builds a local project database that is a real asset in a condo-heavy submarket.

Same packet, same test. Only the date on it is different.

17 days

Default investigation contingency under the purchase agreement, routinely shortened to 7 to 10 days or waived in Bay Area practice. The association has 10 days to deliver the packet.

California Civil Code §4530 and Bay Area escrow practice

What I would ask you

I know the shape of this. I don't know your version of it.

These are the questions that decide whether any of the above is worth building at your firm, and they get asked before anything is scoped. Several of them can kill the project, which is the point.

In the last twelve months, how many of your buyer deals closed at a number below the BRBC rate, and who absorbed the difference?
How many active buyer clients do you have right now whose agreement is more than 90 days old?
Can you pull a list of every client your agents introduced to an out-of-area agent in the last two years?
Do you paper a signed referral agreement before the introduction every time, or only when someone remembers?
What share of your volume is condo or common interest development, and have you had a deal die on non-warrantability?
Does the brokerage have a written policy on who eats a compensation gap, and is it in the independent contractor agreement?

Where it goes

A first win buys the right to the bigger one.

Start here

Outbound referral recovery. It ships in weeks, it needs no access to compliance systems or client files beyond a contact list, and part of the value is retroactive: referrals sent in the last eighteen months may still be collectable, so the first invoice can pay for the build.

Then the buyer side

The BRBC guard, which starts with a backward run over the last twelve months of closed buyer sides and a defensible number for what was absorbed. This one puts the conversation in front of the broker of record rather than an individual agent, which is where it belongs.

Then the document layer

Once document extraction exists at the brokerage it carries HOA packets, the disclosure defensibility audit, and pre-review file linting. That is the largest engagement here, and it is the platform rather than the opener, which is why it does not go first.

Every figure on this page comes from industry research. None of it comes from my own engagements. I have not put a client's results here, because I am not going to dress up someone else's benchmark as my track record. When there is a delivered number worth showing, it will appear here with the client's name on it or not at all.