Seven days of real estate news, filtered down to what could actually change how you run your business, not just what happened in the market. Covers September 21 to 28, 2026, prioritizing Southern California and Orange County where local data is meaningful.
Agent Business: Top 5
1. Rates whipsawed toward 7.5% this week, and pending sales took the first real hit of the year
Freddie Mac PMMS, Sept. 24, 2026; MBA Weekly Survey, Sept. 23, 2026; HousingWire Housing Market Tracker, Sept. 27, 2026
What changed: Freddie Mac's weekly survey put the 30-year fixed at 7.03%, up from 6.95% the week before and roughly 70 basis points above a year ago. MBA data for the week ending Sept. 18 showed purchase applications down 1% and refinance activity at its slowest pace since February 2025. By midweek, daily rate trackers cited by HousingWire showed the 30-year spiking as high as roughly 7.5% intraday before settling near 7.4%, driven by bond-market volatility. Altos Research pending-sales data cited in the same report showed pending sales fell to 59,316 nationally versus 65,152 a year earlier, the first clear non-holiday demand pullback of the year. The share of listings with a price cut rose to 42.5%, up from 41.5% the prior week.
What it means for an agent: This isn't a one-week blip. Three independent sources are telling the same story: financing costs are rising and buyer follow-through is softening in real time, ahead of what the lagging monthly reports will show. If your qualification conversations and payment-shock talking points are more than a couple weeks old, they're already stale. Rising price-cut share also means sellers who “want to test the market” are now competing against more reduced listings than they were two weeks ago.
The question to ask yourself: When did I last update the rate-and-payment numbers I use in a buyer consultation, and would today's numbers change the conversation?
Sources: Freddie Mac PMMS · MBA Weekly Survey · HousingWire Tracker
2. MBA cuts its 2027 mortgage origination forecast, citing two more rate hikes, not a temporary slowdown
HousingWire, Sept. 22, 2026
What changed: The Mortgage Bankers Association lowered its 2027 origination forecast to $2.101 trillion, down from $2.144 trillion projected in August, citing 30-year rates averaging 7.32% and an expectation of two additional Fed rate hikes over the next year as inflation stays above target. Both purchase and refinance applications have fallen in consecutive recent weeks.
What it means for an agent: A forecaster revising its 2027 numbers downward, not just its estimate for the rest of this year, signals the current rate environment is being treated as structural, not a passing phase. Staffing, marketing spend, and pipeline strategy should assume a smaller transaction pie for at least another year, not a snapback.
The question to ask yourself: If transaction volume in my market stays roughly where it is today through 2027, does my current business model still work?
Source: HousingWire
3. New home sales ticked up in August, but builders cut average prices 8.8% to move inventory
Census Bureau/HUD, reported by Inman, Sept. 25, 2026
What changed: New single-family home sales ran at a seasonally adjusted annual rate of 684,000 in August, up 6.4% from July, though Census flagged that gain as not statistically significant, as with the 2% year-over-year decline. The one figure that was statistically significant: average sales price fell 8.8% year-over-year to $478,700, from $525,100. Median price was down 5.8% to $393,700. Months of supply held at 8.5.
What it means for an agent: Builders are the only sellers with the margin and motivation to cut price meaningfully rather than just offer concessions, and they're doing it. When your resale listing is competing with new construction, list-price comparisons alone won't tell the client the real story; you need to know what the builder down the street is actually netting, not what they're advertising.
The question to ask yourself: Do I know the current new-construction pricing and incentive package within a mile of my active listings, or am I working from what I remember a few months ago?
Source: Inman
4. Compass buys an 85-agent Texas brokerage: another independent chooses consolidation over competing solo
HousingWire, Sept. 28, 2026
What changed: Paragon Realtors, a Dallas-Fort Worth brokerage with more than 85 agents and over $260 million in trailing-12-month volume, joined Compass, continuing to operate under its own leadership as “Paragon Group” within the Compass brand. Owner Mike Miller cited industry consolidation and shifting commission structures as reasons for the move.
What it means for an agent: This is one more data point, not the first and won't be the last, in independent brokerages deciding that scale, brand, and tech infrastructure are worth trading some autonomy for, post-settlement. Whatever side of that trade you'd make, it's worth noticing that owners who built real books of business keep choosing platforms over going it alone.
The question to ask yourself: If I ran the numbers honestly, is my current brokerage giving me a competitive advantage, or am I paying for the option to be independent without getting its benefits?
Source: HousingWire
5. Brokers say retention beats recruiting, but most can't spot a flight risk, and it's not about the split
Inman/Lone Wolf 2026 Broker & Owner Survey, Sept. 28, 2026
What changed: In a new survey of brokers and owners, 74% ranked retention above recruiting as a priority, but only 15% felt confident spotting a flight risk early. Lone Wolf's data put the average cost of an unaddressed agent departure at $18,036 and roughly 200 days of disruption. Notably, 94% of brokers cited culture or leadership, not compensation, as the top reason agents actually leave, and 64% of team departures traced back to coordinated competitor recruiting rather than a single grievance.
What it means for an agent: If you're an experienced agent with inconsistent production wondering whether the problem is the market, your brokerage, or you, this is a useful gut check from the other side of the desk. Brokers themselves say the reason producers leave is rarely the split; it's whether leadership engages before a departure becomes a decision.
The question to ask yourself: If I left my current brokerage tomorrow, could I name specifically what was missing, or would it just be a general sense that something felt off?
Source: Inman
Clients & Representation
Only three developments this week met the bar for this section, reported as three rather than stretched to five.
1. New data: “coming soon” listings are outselling private/off-market listings by $10,000-plus
ARELLO study, reported by Inman, Sept. 25, 2026
What changed: A new Association of Real Estate License Law Officials study using 2024 to 2026 data found that “coming soon” listings, publicly visible but marketed ahead of full MLS activation, sold for up to $7,000 more than homes that went active immediately, while private, fully off-MLS listings (“High Confidence Private Listings”) sold $3,000 to $4,000 below standard MLS listings. Coming-soon transaction share rose 4 percentage points while private-listing share fell 2 points over the study period. The study notes NAR kept its Clear Cooperation policy intact while adding a “Delayed Marketing Exempt Listing” option that lets sellers delay public syndication without going fully private.
What it means for an agent: This gives you an actual number the next time a seller asks about going off-market for “privacy” or to “test the price.” The data says the opposite of the common pitch: broad, time-delayed exposure (coming soon) currently outperforms both immediate-active and fully private listings on net price. If you've been defaulting to what a seller requests rather than what the data supports, this is worth bringing into the listing conversation directly.
The question to ask yourself: The last time a seller asked to go private, did I show them a number, or just my opinion?
Source: Inman
2. Compass redesigns its portal to hide days-on-market and price-drop history on some listings
RealEstateNews.com, Sept. 22, 2026
What changed: Compass relaunched Compass.com under a “See It First” model that surfaces listings at every marketing stage across its nine brands (coming-soon listings are publicly visible, while private exclusives are gated behind registration or direct agent contact) without showing days-on-market or price-reduction history. Reports also indicate Compass offered New York agents a $1,000 marketing-package incentive to pull listings off StreetEasy. An independent broker publicly warned that reduced cross-portal visibility and hidden listing history could confuse consumers and erode trust.
What it means for an agent: Whatever brokerage you're at, this is worth watching: it signals a real fight over how much listing history buyers get to see before making an offer, and it raises a legitimate disclosure question. If a buyer can't see that a home sat for 60 days or dropped price twice, are they negotiating with full information? Have a clear, honest answer ready for buyer clients who ask why the history looks different across sites.
The question to ask yourself: Can I explain to a buyer, in one sentence, why the same listing might show different information on different sites, and would that explanation hold up if they pushed back?
Source: RealEstateNews.com
3. Should agents disclose active license restrictions to clients? One industry voice says yes; nothing has changed yet
HousingWire (opinion), Sept. 28, 2026
What changed: In an opinion piece, industry commentator Troy Palmquist argues that agents and brokers under active disciplinary restrictions currently have no obligation to disclose that status to clients, even though the underlying orders are technically public through state license lookups (in California, the DRE). He proposes a standardized short-form disclosure (restriction type, dates, regulator, and a link to the order) delivered at signing. This is a proposal, not a rule change: no regulator or association has adopted it.
What it means for an agent: Nothing is required today, but the argument is public now, and professional-standards debates like this have a way of eventually reaching MLS or DRE rulemaking. It's a reasonable moment to think through your own comfort level with transparency around your professional standing, independent of whether a rule ever forces the issue.
The question to ask yourself: If a client looked up my license status today, is there anything there I'd rather explain myself than have them find on their own?
Source: HousingWire
Technology, AI & Lead Generation
1. NAR's 2026 tech survey: nearly half of Realtors now use AI weekly or daily
HousingWire, Sept. 22, 2026 (based on NAR's 2026 Technology Survey)
What changed: NAR's newest technology survey found 48% of Realtors now use AI tools weekly or daily (23% daily), up sharply from prior years. Only 12% say they don't use AI and don't plan to, down from 32% who said the same last year. The top uses are writing listing descriptions (75%), social posts (56%), and email or follow-up (52%). Among users, 81% cite time savings as the main driver, up from 66% a year ago; the top barriers are the learning curve (63%) and cost (59%).
What it means for an agent: The competitive floor is rising fast. When three-quarters of your peers are using AI for listing copy and half are using it for follow-up, opting out stops being a neutral choice and starts being a response-time disadvantage. The relevant question isn't whether to adopt something, but whether what you adopt touches client follow-up, the highest-leverage use case, rather than just marketing copy.
The question to ask yourself: Of the AI tools I've tried this year, which ones are still part of my actual weekly workflow?
Source: HousingWire
2. Compass pushes its AI platform to 15,000 more agents, and its own numbers show adoption is the hard part
Inman, Sept. 22, 2026; Fortune, Sept. 23, 2026
What changed: Following its merger with Anywhere, Compass has extended its integrated CRM/listing/AI-task platform to 15,000 Coldwell Banker Realty agents since July, backed by more than 9,000 one-on-one activation sessions, with leadership stating that “sentiment matters more than adoption” as the real success metric. Separately, Compass's AI Assistant, which connects roughly 100 tools and can draft outreach or build seller-lead lists, logged more than 97,000 conversations from about 15,000 of Compass's 83,000-plus agents in its first weeks, meaning most agents with access haven't used it yet. Compass's own president acknowledged, “It takes a lot of work to get agents to actually adopt these tools.”
What it means for an agent: This is a useful, honest data point from a company spending heavily to make adoption happen: even with company-wide rollout, live activation sessions, and executive pressure, most agents with access to a new AI tool still aren't using it weeks later. Access isn't the same as habit, and habit is the only thing that produces results. Worth remembering before you or your brokerage invests in the next tool.
The question to ask yourself: The last tool my brokerage or I rolled out with excitement: is it still open on my screen today, or did it quietly stop getting used?
Sources: Inman · Fortune
3. The company behind Google's real estate listing data just filed Chapter 11
Inman, Sept. 25, 2026
What changed: HouseCanary, which supplies listing and valuation data powering Google's home-search product, filed for Chapter 11 bankruptcy on Sept. 22 after a $30 million credit-facility dispute with its lenders led to a blocked foreclosure attempt. The company secured $15 million in debtor-in-possession financing and says its Google data feed is uninterrupted so far. A separate pending $175 million jury verdict against a competitor over trade-secret misappropriation is part of the backdrop.
What it means for an agent: This is a reminder, not a crisis: a meaningful piece of how your listings show up in Google search results depends on infrastructure you don't control and can't see. Worth knowing this dependency exists, and worth not building your entire discovery strategy on any single third-party pipe you have zero visibility into.
The question to ask yourself: If one of the platforms feeding my online visibility went down tomorrow, would I know it happened, or just notice my phone got quieter?
Source: Inman
4. Compass demands MLSs cut off agent production data from recruiting software
Inman, Sept. 25, 2026
What changed: Compass sent MLSs a formal demand, through WAV Group's Victor Lund, asking them to stop feeding Compass-branded agent production, roster, and transaction data into third-party recruiting platforms (specifically naming Courted, BrokerMetrics, and Brokerkit) that competing brokerages use to identify and target agents for poaching. Compass argues the data was submitted for transaction cooperation, not competitive recruiting, and gave MLSs 30 days to comply before considering legal action; franchise broker-owners could opt in separately.
What it means for an agent: Whether or not you're at Compass, this is worth knowing: your MLS production history may already be packaged into software that other brokerages use to build a target list on you, or that your own brokerage uses on agents elsewhere. It's a live test of how far data submitted for cooperation can legally be repurposed commercially.
The question to ask yourself: Do I know what production data about me is currently visible to recruiters through my MLS, and would I be comfortable with how it's being used?
Source: Inman
5. SERHANT launches an “agentic” AI assistant that takes action, not just drafts content
HousingWire, Sept. 23, 2026
What changed: SERHANT introduced “Dot,” billed as an agentic AI chief-of-staff that takes high-level commands like “prep for a listing appointment” and coordinates across calendar, email, and CRM on its own, flagging dormant clients, surfacing expiring pre-approvals, and delivering daily briefings, rather than just generating text. Note: SERHANT's headline claim that its prior tool drove 144% year-over-year GCI growth and saved agents 10-plus hours a week is a self-reported company figure with no independent audit behind it, and should be read as marketing, not evidence.
What it means for an agent: This points to where the next wave of “AI for agents” is headed: from writing a listing description to autonomously working your database, a meaningfully bigger claim on your workflow and your data than a chatbot. Worth watching for what real usage numbers eventually show, but not worth acting on the productivity claims as stated.
The question to ask yourself: If an AI tool could take actions in my CRM on its own, which actions would I actually trust it to take unsupervised, and which would I not?
Source: HousingWire