Listing video vs. photography: where each one actually pays

Photos, by a wide margin. In the National Association of REALTORS® 2025 Home Buyers and Sellers Generational Trends Report, 83% of buyers who used the internet during their search rated listing photos "very useful" — the highest-rated website feature of any kind. Virtual tours came in at 41%. Videos came in at 29%. That is the single clearest answer available to the question this article asks, and it comes from the flagship survey in the industry: 5,390 responses from primary-residence buyers, a 3.2% adjusted response rate, giving a 95% confidence interval of plus or minus 1%. That report covers purchases made between July 2023 and June 2024. NAR's separate 2025 Profile of Home Buyers and Sellers, which surveys a later cohort — purchases from July 2024 to June 2025, 6,103 responses — asks the same question and puts photos at 81%, virtual tours at 38% and videos at 29%. The two reports are different surveys of different years, not competing estimates of the same one; the ranking is identical in both, and this article uses the Generational Trends figures throughout.

The floor plan result is the one worth stopping on. At 57%, buyers rated floor plans "very useful" at nearly twice the rate they gave video, and well ahead of virtual tours. Meanwhile, a study of 25,000 listings across leading U.S. portals between March and June 2021, reported by NAR, found only 16% included a floor plan, fewer than 30% used professional photography best practices, and 94% carried no virtual tour at all. Highly valued, rarely supplied, and cheap: that combination does not exist anywhere else on the list.

The supply side matches. In the same Generational Trends report, only 12% of recent sellers said their agent used video to market the home, and only 16% said a virtual tour was used, against 86% who reported an MLS listing and 61% a yard sign. (The later Profile cohort splits the video question differently, reporting 5% for video hosting websites and 20% for virtual tours, so compare like with like before quoting either.) Agents are not withholding video because they have not heard of it. They are allocating a limited marketing budget against the assets that carry the listing.

Agents report the same ranking. In NAR's 2025 Profile of Home Staging (1,266 responses, margin of error plus or minus 2.75%), 73% of buyers' agents said having photos available was more or much more important to their clients, compared with 48% for videos and 43% for virtual tours. Among sellers' agents the gap was wider still: 88% for photos versus 47% for videos. Survey data measures stated preference, not what moved a sale. The direction is nevertheless consistent across three independent instruments, which is more than most claims in this category can offer.

The association is well documented. The causal size is not, and the most-quoted study is thirteen years old. Redfin's 2013 analysis of homes listed and sold across 22 markets found listings shot with a DSLR sold for $3,400 to $11,200 more relative to list price, and homes with the sharpest 10% of photos sold at or above list price 44% of the time versus 13% for photos of average sharpness. It is a brokerage study rather than peer-reviewed work, it published no sample size, and it is almost always cited without a date.

It also has a selection problem nobody quoting it mentions. Agents who commission professional photography differ systematically from agents who do not: better listings, sharper pricing, more experience. The photographer gets credit for all of it, which makes the observed gap an upper bound on what the camera is worth rather than an estimate of it.

The cleanest quasi-experimental evidence on photography comes from an adjacent market. A difference-in-differences analysis in Management Science tracked 7,423 Airbnb properties over 16 months and found that properties with professionally shot verified images were associated with 8.98% higher occupancy than those using host-taken photos. Nightly rental is not home sale, and occupancy is not price. But the design compares the same properties before and after the switch, which removes most of the selection problem that weakens the real estate numbers.

The price band nobody quotes

Redfin's own release contains a finding the pages citing it consistently drop: the gap was concentrated in the mid-market. Homes around $400,000 with professional photos sold about 21 days faster, while homes above $1 million sold only four days faster and at prices similar to those with amateur photos. Per the California Association of REALTORS® May 2026 Home Sales and Price Report, Central Coast median sale prices were $968,000 in Monterey County, $1,254,500 in Santa Cruz County, $1,050,000 in San Luis Obispo County and $765,000 in San Benito County. Much of the region sits at or above the band where Redfin's measured premium thinned out.

That is not an argument against professional photography. Photos are the input every other channel consumes: the MLS record, every portal, the social post, the email and the thumbnail on the video itself all draw from one shoot. It is an argument against budgeting photography as though it buys a price premium at the top of the market. Above a million dollars, buy it as raw material.

How many photos is enough

More than most listings carry, and not infinitely more. In the 53,063-listing dataset analysed in NBER Working Paper 33204, the average listing carried 14 photographs and the average time on market was 32 days. The more useful finding comes from a peer-reviewed study in The Journal of Real Estate Finance and Economics, which found additional photographs are associated with higher sale prices but simultaneously longer marketing duration, both effects rising at a decreasing rate for interior shots. Photographs help buyers self-select, which can mean fewer showings and better-qualified ones. Cover every room, the exterior and the floor plan properly. Twelve angles of the same countertop buys neither price nor speed.

Nobody has published an isolated effect estimate for a linear listing video. That sentence is the honest answer. The large studies routinely presented as video evidence measured navigable virtual tours, which is a different product, consumed differently and rated differently by buyers.

The distinction is not pedantic. A virtual tour is navigable: the viewer chooses where to go, as in a 3D dollhouse or a panoramic walkthrough. A listing video is linear: the creator chooses the path and the pacing. Buyers rate them 41% and 29% "very useful" respectively, so they are clearly not interchangeable in a buyer's mind. Most industry statistics blur the two, and that blur is where inflated claims are manufactured.

What the tour studies found

NBER Working Paper 33204 (Soleymanian and Qian, 2024) applied causal machine learning and instrumental variables to 53,063 MLS listing observations and reported an estimated virtual-tour effect on sale price of about 1% on average — an identification strategy applied to observational data, not a trial. The authors also found that effect has declined over time, particularly post-COVID, as tours moved from novelty to standard equipment. The same paper found tours less impactful for highly differentiated properties, and more beneficial in competitive markets and for less experienced agents unfamiliar with the local market.

A second large study points at time rather than price. Research in Information Systems Research, summarised by UT Dallas in 2025, examined nearly 43,000 properties listed in 2018 and 2019 on a major Chinese real estate platform. Homes with a non-immersive VR tour spent an average of 19 days on the market versus 34 days without, while the tours showed no association with the final selling price. A different country, a different platform and pre-pandemic data, so it travels poorly to a Central Coast listing — but it is a larger sample than anything the vendor literature offers.

The best-documented vendor study is Matterport's 2020 research post, which is unusual in publishing its methods. Its matched comparison of 700 listings, 350 with a 3D tour and 350 without, in a Southern U.S. metro between March 2018 and July 2019, reported that listings with a tour sold 20% faster and for 4.8% more. The headline on that same post advertises "up to 31% faster", a figure drawn from the best of four markets in a separate regression. The tightly matched study returns the smaller number, nobody has re-run either analysis on 2024 to 2026 listings, and the NBER work suggests the effect has shrunk since.

Why every raw comparison overstates the case

The NBER dataset contains the confound in plain sight. Listings that offered virtual tours also had significantly more photos and longer descriptions than listings that did not, so a with-versus-without comparison credits the tour for the photography, the copywriting and the agent behind both. Any statistic reporting a raw gap between listings that had video and listings that did not is measuring effort, not format.

It could not be sourced. The claim appears across dozens of vendor and agency pages, almost always attributed to the National Association of REALTORS®. The string "403" does not appear in NAR's 2025 Home Buyers and Sellers Generational Trends Report or in its 2025 Profile of Home Staging, and a domain-restricted search of nar.realtor surfaces no such figure. The same is true of the companion claims that listings with video receive 157% more website traffic, that 51% of homebuyers use YouTube in their search, and that 85% of buyers and sellers want to work with an agent who uses video.

What NAR does publish runs the other way. Only 12% of sellers reported their agent used video at all, and only 29% of buyers rate videos "very useful". A statistic implying video quadruples inquiries would be difficult to reconcile with an industry that has declined to adopt it.

This matters beyond citation hygiene. The numbers you plan against decide what you skip. An agent budgeting off "403% more inquiries" books the video and skips the floor plan, which 57% of buyers rate "very useful" and only 16% of scanned listings carry. The false number overstates video, and in doing so it reorders everything underneath it.

Every competing comparison table on this topic fills its cells with adjectives — "emotional connection", "shareability", "lead qualification" — which is a way of appearing to answer the question without answering it. This one uses numbers where numbers exist and says "no estimate published" where they do not. The gaps are as informative as the figures.

Not the audience the industry assumes. The premise underneath nearly every video pitch is that video reaches younger buyers. NAR's data on where buyers actually went during their search says close to the opposite: 45% of buyers aged 70 to 78 and 43% of those aged 60 to 69 used an online video site during their home search, versus 25% of buyers aged 26 to 34.

Stated preference runs slightly the other way, and stays remarkably flat: 31% of buyers aged 26 to 34 rated videos "very useful", versus 33% of those aged 35 to 44, 25% of those aged 60 to 69 and 21% of those aged 79 to 99. So younger buyers rate video marginally higher while older buyers use video sites considerably more. Neither pattern supports targeting listing video at a young audience, and the spread across five decades of buyers is twelve percentage points. Age is not the variable that decides this.

Market conditions are. NBER Working Paper 33204 found tours more beneficial in competitive markets, where buyers face high search costs, and less impactful for highly differentiated properties that already stand out. Sellers were also less likely to offer tours in hot markets, where scarce inventory does the work instead. California's statewide median time to sell a single-family home was 22 days in May 2026, unchanged year over year. On a typical Central Coast listing that argues for restraint, and for concentrating motion assets on properties that genuinely need explaining: unusual layouts, acreage, view lots, and homes competing against many close substitutes.

Video's defensible case is the feed rather than the listing page. In NAR's 2025 REALTORS® Technology Survey (1,241 responses, margin of error plus or minus 2.78%), 39% of REALTORS® named social media as the technology delivering the highest number of quality leads, ahead of CRM at 23%, their local MLS at 17%, and listing syndication portals at 9%. Video is the native format of that channel. Judged as agent-brand marketing rather than listing collateral, it competes on different ground, and the property becomes the subject matter rather than the product.

Three constraints sit outside the production quote, and all three are cheaper to solve before the shoot than after it.

Distribution

Listing syndication decides which assets buyers actually encounter. Photos travel through the MLS out to consumer portals reliably; video handling varies by portal and by MLS rules. Confirm how your own MLS and the portals it feeds treat a video file before commissioning one, because a video that lives only on your website competes for attention you then have to buy separately.

Search eligibility

Google's video SEO documentation sets specific conditions: the video's watch page must be indexed, must be performing well in Search, and must have the video as the page's primary content before the video is eligible for video results. VideoObject structured data or a video sitemap is recommended rather than required; it helps Google find the video and lets you control the thumbnail URL, name and description that appear. The consequence is blunt. A listing video embedded at the bottom of a property page, where the primary content is the property, generally will not surface as a video. If search visibility is part of the business case, the video needs its own indexable page and the technical work that goes with it.

Compliance

Aerial imagery shot to help market a property is a commercial operation, and 52% of REALTORS® report using drone photography or video, against 38% using virtual tours and 4% using VR for immersive walkthroughs. That makes drone work the most widely adopted of the three, and the most common compliance gap in listing marketing. 14 CFR Part 107 requires a remote pilot certificate (§107.12) and a minimum age of 16 (§107.61), visual line of sight at all times (§107.31), and caps flight at 400 feet above ground level and 100 mph (§107.51). Hobbyist rules do not apply regardless of who owns the aircraft. Ask any vendor for a certificate number, and check controlled airspace around Monterey Regional and Salinas Municipal.

Measurement

Half of REALTORS® spend under $250 a month on lead generation in total: 24% spent less than $50 monthly over the past year and 27% spent between $50 and $250. Against that baseline, a commissioned video is rarely a marginal purchase; for most agents it competes directly with the rest of the year's marketing. If it is going to be bought, it needs a destination that can be tracked to an inquiry rather than a view count on a platform reporting impressions back to itself. Platform analytics measure the platform; your own site measures the business, which is why real estate marketing is better served by a tracked landing destination than by another follower count.

Photography, then a floor plan, then a navigable tour, then video — and the second step is the one most people skip.

The staging rung has the clearest cost anchors in the category. NAR's 2025 Profile of Home Staging reports a median of $1,500 spent on a professional staging service, versus $500 when the seller's agent staged the home personally. It sits high in the sequence for a mechanical reason: staging changes what the camera records, so it has to happen first or the photography budget partly buys pictures of the wrong thing.

The reality check is the total. Half of REALTORS® spend under $250 a month on lead generation, so most listing budgets never reach rung four in a single cycle, and that is a defensible outcome rather than a failure. Where the fundamentals are covered already, video is a reasonable next purchase, particularly for a property that needs explaining or an agent building a social presence that outlives the listing.

One caveat on the sequence itself. It is an inference drawn from preference data, deployment data and effect estimates of varying quality, and no trial has tested this ordering against another. It is the most defensible ordering the published evidence supports, which is a materially different claim from proof. Anyone selling you the opposite ordering should be asked which study they are reading.