THE LANGUAGE OF DATA SERIES | PART 5 OF 7
The Working Feeds: BBO, R&D, and Custom
The feeds nobody sees: the one behind your CMA, the one behind tomorrow’s products, and the purpose built pipes for everything else
By Saul Klein and John Reilly September 2026
The famous MLS feeds are the ones that show listings to the public. This article, which stands alone, covers the other family: the feeds that work behind the scenes, powering analysis, products, and special purposes. If the display feeds are the showroom, these are the shop.
BBO (Brokerage Back Office): the feed behind your back office. The feed nobody outside management ever sees, and the one your brokerage may depend on most. BBO delivers the market’s listing data into your firm’s internal systems: CMA and pricing tools, transaction management, recruiting dashboards, accounting, and the market share analyses your broker presents at the sales meeting. BBO is internal use only; nothing from it may be shown to the public. Under current national policy, the MLS must make this feed available to participating brokers, complete, with no broker able to pull their listings out, because comparables and statistics only work when the dataset is whole. When your CMA tool finds every relevant sale in the neighborhood in four seconds, the BBO feed is why.
R&D feeds (research and development): the feed behind tomorrow’s products. A newer category, and the one growing fastest. An R&D feed delivers listing data, sometimes current, often historical, so that a company can build and test something new: a valuation model, a market analytics product, a search experience, an AI application. Notice what makes R&D different from every other feed. The purpose is not to display listings and not to run a brokerage. The purpose is to create something else out of the data. That is why R&D access is almost never bundled with membership. It is granted, when it is granted at all, under a specific negotiated license that defines the project, the dataset, the retention rules, and increasingly whether the data may touch an AI training environment at all. The newest MLS rules treat unlicensed research and development on other brokers’ listings as a violation, full stop. If your firm, or your vendor, is building a product on MLS data, the first question is not technical. It is: what does the license say?
Custom feeds: purpose built pipes for specific jobs. Beyond the named programs, MLSs increasingly build feeds tailored to one defined purpose under one defined license: a valuation feed for appraisers and lenders who need sold data to support opinions of value; an assessment or litigation feed for a government agency or a court matter; a match and append feed that lets a firm enrich its own records with MLS fields; an AI access license that lets a technology company query the data under negotiated no training terms. The pattern is always the same and always worth noticing: the narrower and more specific the purpose, the more the arrangement looks like a commercial contract and the less it looks like a membership benefit. That pattern is the future of this subject.
One habit will keep you oriented across all of these. For any feed, ask two questions. What may the recipient do with what arrives, that is the license. And is the output a listing being shown, or something new being made? The working feeds in this article live almost entirely on the something new side, which the industry calls derivative works, and derivative works are where both the commercial value and the legal exposure concentrate. A companion article in this series covers that distinction on its own; this one closes with the practical rule: on the working feeds, no license means no build.
This article stands on its own and is also part of the practitioner series “The Language of Data”. Copyright 2026 Saul Klein and John Reilly. All rights reserved. Commentary and general information, not legal advice.
