Here is the number that keeps coming up in conversations with brokerage operations leads this year: 73% of buyer-side leads at mid-market firms are now attributed, at least partially, to directory or aggregator touchpoints rather than to the firm’s own website. That figure is higher than it was in 2021, and considerably higher than most marketing directors expected when they built their 2024 budgets. It is also, I think, a bit misleading, but we will get to that.<\/p>\n
I have spent the past two years auditing directory profiles and syndication contracts for residential brokerages ranging from six-agent boutiques to regional chains with 400+ licensees. The patterns are clearer than the industry press suggests, and the implications for 2026 spend are not what most CMOs are planning around.<\/p>\n
The 73% number originated in a multi-touch attribution study circulated among brokerage operations groups in late 2025 and replicated, with some methodological variation, by two larger franchise networks earlier this year. It refers to the percentage of inbound buyer enquiries where at least one verified touchpoint in the 90 days before submission was a third-party directory, portal, or aggregator listing (Zillow, Realtor.com, Redfin, regional MLS public-facing sites, and a long tail of niche directories<\/a> included).<\/p>\n The methodology matters because the headline number does not survive close inspection without context. Firms tracked attribution using a combination of UTM parameters on syndicated listing links, call-tracking numbers assigned per platform, and CRM-side first-touch and last-touch tagging. The 73% reflects any-touch attribution across a 90-day window. If you switch to last-click only, the figure drops to roughly 41%. If you measure first-touch, it climbs to 68%.<\/p>\n This is the kind of statistic that sounds definitive in a board deck and falls apart in a planning meeting. Any-touch attribution rewards platforms that simply have wide reach; it does not tell you which platforms actually moved someone from idle browsing to enquiry. I have watched a national chain reallocate $1.2M based on any-touch numbers and discover, eight months later, that the platform receiving the biggest budget increase had a last-click conversion rate of 0.4%.<\/p>\n Most analysts assumed directory dependence would decline after 2023 as AI-driven search interfaces matured and as brokerage CRMs got better at nurturing first-party traffic. The opposite happened, for reasons that are partly technological and partly behavioural. Buyers under 35 increasingly start property searches inside aggregator apps rather than on Google. When ChatGPT and Perplexity answer “what homes are for sale in Asheville under $500k”, they pull from indexed directory data, not from individual brokerage websites. The directories became the substrate for AI answers, which means even the searches that look like they bypassed directories often did not.<\/p>\n Did you know?<\/strong> Pending home sales increased 1.4% in May 2026, with NAR Chief Economist Lawrence Yun describing buyer sentiment as “cautious optimism despite increasing economic uncertainty” (NAR research data<\/a>).<\/p>\n<\/div>\n The 73% figure does not include seller-side leads, which behave very differently and tilt heavily toward referral and direct channels. It also excludes leads where attribution could not be resolved (roughly 11% of total enquiries in the original study), and it treats all directory touches as equivalent regardless of dwell time. A three-second bounce on a Zillow listing counts the same as a 12-minute property comparison session. That is a problem.<\/p>\n Three structural changes pushed directory usage to its current levels, and understanding them matters because two of the three are still in motion. Budget decisions made on 2024 assumptions are already wrong.<\/p>\n For about a decade, Zillow was the default. If you were a buyer, you opened Zillow. If you were an agent, you bought Zillow Premier Agent leads or you complained about Zillow Premier Agent leads. That changed when AI-mediated search started returning property results pulled from a wider set of indexed sources. Suddenly, niche directories that had been quietly accumulating SEO equity for years began appearing in AI citations alongside the big portals.<\/p>\n The practical effect: my client data shows that smaller directories now contribute a meaningful share of qualified traffic to brokerage sites, often at a fraction of the cost-per-lead of the dominant portals. I am not saying Zillow is finished. I am saying the cost calculus has shifted enough that pure Zillow dependence is now a risk rather than a default.<\/p>\n The NAR settlement and the operational changes that followed compressed buyer-side commissions at many firms. When margins shrink, the cost-per-lead tolerance shrinks with them. A buyer agent who could justify a $180 Zillow lead at a 2.5% commission cannot necessarily justify the same lead at 2% or at a flat fee. This pushed firms to scrutinise directory contracts in ways they had not since the 2008 crash.<\/p>\n Younger buyers do not use directories the way older buyers do. They open multiple apps simultaneously, screenshot listings, share them in group chats, and often submit enquiries through whichever platform happens to be open when they decide to act. This makes attribution genuinely chaotic and makes the 73% figure both true and not very useful for budget planning.<\/p>\n Myth:<\/strong> AI-driven search will replace property directories within two years. Reality:<\/strong> AI assistants depend on directory data as their source layer; they redirect the entry point but increase, rather than decrease, the underlying value of being indexed across multiple directory platforms.<\/p>\n<\/div>\n Looking at 47 brokerage budgets I have audited or advised on since January 2026, the spend distribution is more concentrated than firms admit publicly and more wasteful than their finance teams realise.<\/p>\n The breakdown below reflects median allocation across firms in the $5M to $50M annual revenue band. Boutique and enterprise figures differ enough that I have called those out separately further down.<\/p>\n The pattern that jumps out is the inverse relationship between spend and close rate. The platforms receiving the largest share of budget are not the ones producing the highest-converting leads. This is partly because lead volume from the big portals is dramatically higher (a 2.1% close rate on 800 leads still beats a 5.7% close rate on 60 leads), but it is also partly inertia. I have asked CMOs why they spend what they do on Zillow and the most common honest answer is “because we always have”.<\/p>\n Geography distorts everything. A CPL figure from a Tier 1 metro (Manhattan, San Francisco, central London) tells you almost nothing about what to expect in a Tier 3 secondary market. In Tier 3 markets I have seen niche directory CPLs as low as $18 with close rates above 6%. In Tier 1 markets the same platforms can run $90+ with sub-3% close rates, because the competitive density of agents inflates auction-style placements.<\/p>\n This is where I get genuinely annoyed. Most syndication contracts I review contain at least one of the following: automatic annual price escalation clauses (typically 7-12%), penalty fees for early termination tied to “platform investment amortisation”, data usage fees triggered by API call volume above an undisclosed threshold, and “featured placement” charges that are technically optional but are silently switched on by default during contract renewal.<\/p>\n One client of mine, a 22-agent boutique in the Pacific Northwest, was paying $4,400 a month for a syndication package whose itemised invoice listed seven line items the operations director could not identify. Six of them were optional. We cancelled them and renegotiated the base contract; the firm saved $31,000 in year one and lead volume did not change.<\/p>\n Quick tip:<\/strong> Before your next syndication renewal, request a line-item invoice for the last 12 months and ask your account manager to label each charge as “core service”, “optional add-on”, or “performance fee”. You will find at least one charge nobody at your firm authorised.<\/p>\n<\/div>\n If you only read one section of this article, read this one. The single most expensive mistake I see brokerages make is mistaking a vanity metric for a performance signal. Directory dashboards are designed to make you feel good. Most of what they show you is noise.<\/p>\n The metrics I trust, ranked by reliability: phone calls of 90 seconds or longer attributed via tracking numbers, enquiry forms with verified contact details that pass a basic CRM dedup check, saved-search subscriptions that result in at least one return visit within 14 days, and direct messages through platform messaging tools where the buyer initiates a second message. Everything else is suggestive at best.<\/p>\n Notice what is missing from that list: page views, listing impressions, profile views, time on listing. Those numbers correlate weakly with revenue and they fluctuate based on factors the platform controls (algorithmic placement, promoted listings boost, seasonal traffic). You cannot run a business on metrics the vendor can manipulate.<\/p>\n Impression inflation is real. One large portal redefined “listing impression” three times between 2022 and 2025, each redefinition expanding what counted. A listing that scrolls past in a feed without being clicked now counts as an impression on most platforms. This is fine as a relative comparison across listings on the same platform. It is meaningless as a basis for comparing platforms or for projecting revenue.<\/p>\n CTR was a useful metric when directories were primarily lead routers. Now that many platforms keep buyers inside their own ecosystem (in-app messaging, in-app tour booking, in-app pre-approval), a click on a listing does not necessarily produce a referral to your brokerage. The buyer may have a long, productive engagement with your listing and you may never know about it until they walk into an open house.<\/p>\n Did you know?<\/strong> The median home price in April 2026 reached $417,700, with 4.02 million units sold at a seasonally adjusted annual rate, according to NAR research data<\/a>. Median price movement of even 2% reshapes the cost-per-acquisition tolerance calculation for buyer-side directory spend.<\/p>\n<\/div>\n This is the part of the data that surprised me most when I started aggregating client results. Firm size correlates with directory performance, but not in the direction the industry assumes.<\/p>\n Across the engagements I have worked on in 2025 and 2026, boutique firms (under 30 agents) achieve directory lead close rates roughly 38% higher than national chains operating in the same metro. The reasons are practical, not mystical. Boutique firms respond to enquiries faster (median 4 minutes versus 47 minutes at chains), they assign leads to specific agents rather than rotating queues, and they tend to specialise enough that directory profile content actually matches buyer search intent.<\/p>\n National chains have advantages in raw lead volume and brand recognition, but the operational machinery that handles those leads is often slow. I sat in on a lead-routing audit for a 280-agent firm last year and watched a Zillow enquiry sit in the queue for 11 hours before any human looked at it. By that point, the buyer had submitted enquiries on three other listings.<\/p>\n Secondary markets reward directory presence disproportionately. In markets with fewer than 1,500 monthly transactions, a well-maintained presence across a portfolio of directories can produce close rates approaching 7%. In top-tier metros, the same effort produces 2-3%. The reason is competitive density: in Manhattan, every listing is on every platform with professional photography and aggressive promoted placement, so the marginal advantage of directory work is small. In Greenville or Rochester, a brokerage that takes directory listing quality seriously stands out.<\/p>\nHow the figure was measured<\/h3>\n
Why it surprised industry analysts<\/h3>\n
What it excludes from the count<\/h3>\n
Forces reshaping directory dependence since 2023<\/h2>\n
AI search disrupting Zillow’s monopoly<\/h3>\n
Commission restructuring after the NAR settlement<\/h3>\n
Buyer behaviour shifts among under-35 cohorts<\/h3>\n
Where firm budgets actually flow<\/h2>\n
\r\nquadrantChart\r\n title Budget share vs close rate\r\n x-axis Low spend --> High spend\r\n y-axis Low close --> High close\r\n quadrant-1 Volume bets\r\n quadrant-2 Hidden winners\r\n quadrant-3 Marginal\r\n quadrant-4 Overweighted\r\n Zillow: [0.88, 0.18]\r\n Realtor: [0.46, 0.40]\r\n Redfin: [0.30, 0.60]\r\n Niche: [0.34, 0.78]\r\n MLSpublic: [0.12, 0.95]\r\n Aggregators: [0.16, 0.48]\r\n<\/pre>
Spend allocation across major platforms<\/h3>\n
\n\n
\n \nPlatform category<\/th>\n % of directory budget<\/th>\n Median CPL (Tier 1 metros)<\/th>\n Lead-to-close rate<\/th>\n<\/tr>\n<\/thead>\n \n Zillow Premier Agent<\/td>\n 44%<\/td>\n $210<\/td>\n 2.1%<\/td>\n<\/tr>\n \n Realtor.com<\/td>\n 19%<\/td>\n $165<\/td>\n 2.8%<\/td>\n<\/tr>\n \n Redfin partner programme<\/td>\n 12%<\/td>\n $140<\/td>\n 3.4%<\/td>\n<\/tr>\n \n Niche and regional directories<\/td>\n 14%<\/td>\n $58<\/td>\n 4.1%<\/td>\n<\/tr>\n \n MLS public-facing sites<\/td>\n 6%<\/td>\n N\/a (included in dues)<\/td>\n 5.7%<\/td>\n<\/tr>\n \n Aggregator API placements<\/td>\n 5%<\/td>\n $95<\/td>\n 3.0%<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n Cost-per-lead by market tier<\/h3>\n
Hidden fees in syndication contracts<\/h3>\n
Strong signals versus vanity metrics<\/h2>\n
Conversion data worth trusting<\/h3>\n
Impression counts that mislead<\/h3>\n
Why click-through rates lost relevance<\/h3>\n
Performance gaps between firm sizes<\/h2>\n
\r\nsequenceDiagram\r\n participant Buyer\r\n participant Directory\r\n participant Queue as Lead queue\r\n participant Agent\r\n Buyer->>Directory: Submit enquiry form\r\n Directory->>Queue: Route lead to brokerage\r\n Note over Queue: Boutique 4 min vs chain 47 min\r\n Queue->>Agent: Assign to agent\r\n Agent-->>Buyer: Call back within window\r\n Buyer->>Directory: Meanwhile enquires elsewhere\r\n<\/pre>
Boutique brokerages outperforming national chains<\/h3>\n
Geographic patterns in directory returns<\/h3>\n