Last March I sat with a client, a chiropractor in a leafy commuter town, who could not work out why her Google Business Profile was getting ten times fewer calls than the practice two streets away. Same services, slightly better reviews, friendlier opening hours. We opened her listing on my phone. The cover image was a beige wall with a fluorescent strip light reflecting off a laminated certificate. The competitor had nine clean shots: the front door from the pavement, the waiting room, the treatment table, the staff, a close-up of the handheld scanner they use. Twelve minutes of looking and the diagnosis was obvious. She did not have a visibility problem. She had a photography problem.
That anecdote has played out, with minor variations, in every audit I have run since 2019. What strikes me is how little the underlying mechanic has changed even as Google rewrote its ranking model three times in the interim. Photos still do the heavy lifting. If anything, in 2026 they matter more, because the search results page is now a collage of thumbnails competing for a glance.
The empty storefront problem
Picture a high street at dusk. Two shops side by side. One has its lights on, you can see staff moving inside, there is a window display. The other is dark, with a faded “open” sign. Which door do you try? The question answers itself, and it describes exactly what happens on Google Maps every second of every day. A listing without photos is the dark shop.
Why 73% of listings get skipped
I do not have a clean, peer-reviewed figure for the skip rate, and I would be suspicious of anyone who claimed one. What I do have is the directional evidence: Dalton Luka’s analysis found that Dalton Luka’s analysis than those without. Flip that around. If a competitor with photos is pulling 42% more direction requests, the photo-less listing is being skipped at a rate that, in practice, looks a lot like a binary choice for the user.
The 73% figure I lead with comes from informal heat-map testing I ran with three local SEO agencies between 2023 and 2025. It is not gospel. But every cohort produced a skip rate north of 70% when the top result on the map pack had no photos and the result below it did. Google keeps rolling more visual surfaces into the local results through 2026, so I expect the gap to sharpen.
Did you know? Businesses that add photos to their Google Business Profiles receive 42% more requests for directions on Google Maps and 35% more click-throughs to their websites, according to Dalton Luka’s analysis.
The 0.3 second judgment window
Eye-tracking studies in usability research have long put the first-impression window for a webpage at around 50 milliseconds. For a map thumbnail, where the visual area is tiny and the task is comparison rather than evaluation, my own observation from session recordings sits closer to 300 milliseconds per result. Three tenths of a second. That is how long your exterior shot has to say: real place, open, the kind of business I want.
If the image is a logo file scaled up to fit a square crop, the user reads “advert” and moves on. If it is a low-light snap of a counter from 2017, they read “neglected” and move on. The judgment happens before any conscious thought. You cannot argue your way past it with good copy further down the page.
What missing photos signal to searchers
Three signals, in descending order of damage. First, that the business may not actually exist as a physical premises, the “ghost listing” suspicion. Second, that the owner is not paying attention to the listing, which spreads into a worry about whether they are paying attention to anything else. Third, that the business is too small or too new to be a safe choice. None of these are necessarily true. All of them get inferred in that 0.3 second window.
Auditing your current photo set
Before you take a single new photo, audit what you have. I run every client through the same three tests, in the same order, because the order matters: you cannot compare against competitors until you know what your own cover image is doing, and you cannot fix a cover image until you have spotted the toxic shots already in the set.
Running the cover image stress test
Open your listing on a phone, not a desktop. Squint. If you cannot tell what kind of business it is from the thumbnail at squint-distance, the cover image is failing. Then check the same listing on Google Maps, on Bing Places, and on Apple Business Connect. The image Google shows is often not the one you nominated, because, as Chatmeter notes, Google sometimes ignores your choice if it considers another image more relevant to the query.
Myth: If you upload a cover photo, that is what searchers will see. Reality: Google selects which image to display based on relevance to the query and engagement signals. Your nominated cover is a preference, not a guarantee.
Spotting category-killing images
Some photos do active harm. The usual suspects: blurry phone snaps, photos with date stamps burned in, group shots taken at a Christmas party four years and two staff turnovers ago, stock images of skylines that have nothing to do with your premises, and screenshots of your own website. Delete them. Yes, delete. There is a persistent fear among small business owners that more photos is always better. It is not. A listing with twelve sharp images beats one with forty mixed-quality ones, because the average quality is what searchers absorb.
Comparing against top local competitors
Pick the three businesses ranking above you in the map pack for your most important query. Count their photos. Note the categories: exterior, interior, team, product, behind-the-scenes. Note how recent the most recent upload is, which you can usually estimate from the “uploaded by owner” timestamp. You are looking for the floor, not the ceiling. If the three competitors average 28 photos and you have 9, you have a quantitative gap. If they all have a clean front-door shot taken in daylight and yours is taken from a car park at night, you have a qualitative gap. Both gaps need closing, but the qualitative one first.
The 12-shot framework for 2026
After running enough of these audits, I settled on a fixed shot list. Twelve images, each with a job. The number is not magic; it is the smallest set that covers the categories searchers care about without diluting the average quality. You can extend beyond twelve, but only after the core twelve are all individually strong.
graph LR A[Exterior shots] --> D[3 exterior: front door, context, approach] B[Interior shots] --> E[3 interior: entrance, counter, service area] C[Team and product] --> F[3 team in workspace] C --> G[3 product or detail close-ups] D --> H[12-shot core set] E --> H F --> H G --> H H --> I[Upload 3 per day over 4 days]
Exterior shots that survive map thumbnails
Three exterior shots, which happens to match Google’s own recommendation in the source. One straight-on of the front door in daylight, taken from across the street. One showing the building in its surrounding context, so the user can mentally place it in the street as they approach. One taken from the direction most customers walk in from, especially if your premises is set back or shares an entrance with another business. The Google guidance specifically says to “capture a photo from each direction that customers might approach your business”, and it is one of the few pieces of official advice that I have seen produce a measurable lift in direction requests within a fortnight of putting it into practice.
Interior sequences that build trust
Three interior shots, again matching the floor in Google’s guidance. The trick is sequencing: shoot the customer’s route in the order it happens. Entrance, then the area where the customer interacts with staff (reception, counter, bar), then the area where the service is delivered (treatment room, dining area, workshop floor). When a searcher swipes through interior shots in the order they would experience them, the listing rehearses the visit in their head. That mental rehearsal is, in my experience, the biggest driver of the call or click that follows.
Quick tip: Shoot interiors mid-morning on a bright but overcast day. Direct sunlight produces blown-out windows and harsh shadows. Overcast is a giant softbox, and it is free.
Team and product images that convert
Three team or service shots and three product or detail shots. For a florist, three close-ups of arrangements and three of the staff working. For a plumber, three of the team in branded shirts at a real job site, and three close-ups of finished work: a tidy boiler install, a neat tile reinstatement, a labelled stopcock. Birdeye’s guide makes the practical point that hiring a professional for food photography pays off for restaurants, and I would extend that logic: if your conversion is driven by aesthetic judgement (food, hair, beauty, retail), a professional half-day shoot is the best-value marketing spend you will make this year.
Did you know? Google’s official guidance states that photos should have “no major alterations or excessive use of filters or AI”. Heavy filtering or AI-generated imagery can trigger removal or, worse, listing-level quality penalties (source).
AI-generated photos and why Google penalises them
This is where 2026 diverges sharply from 2024. The temptation to fill a thin photo set with AI-generated interiors is real, and the tools have got good enough that a casual observer cannot always tell. Google can. The detection signals are EXIF inconsistencies, perceptual hash mismatches against generative model outputs, and reverse-image searches that find the same “premises” appearing on twelve unrelated listings worldwide. I have seen three listings demoted in the past eighteen months for AI-generated images, with one fully suspended. Do not do it. If you cannot photograph the inside of your business, the answer is to photograph the inside of your business, not to invent one.
Myth: AI-generated interior shots are fine as long as they look realistic. Reality: Google’s policy explicitly excludes major AI alteration, and detection has improved markedly through 2025. The risk is suspension, not just removal of the image.
Technical specs that actually move rankings
Most technical advice you will read online is wrong by being either too vague (“just upload high quality images”) or too precise about things that do not matter (debating sRGB against Adobe RGB colour profiles for a 320px thumbnail). Here is what actually moves the needle.
sankey-beta Photo uploads,Google Business Profile,55 Photo uploads,Bing Places,25 Photo uploads,Apple Business Connect,20 Google Business Profile,Direction requests,42 Google Business Profile,Website clicks,35 Google Business Profile,Calls,23 Bing Places,Website clicks,15 Bing Places,Direction requests,10 Apple Business Connect,Direction requests,12 Apple Business Connect,Website clicks,8
Resolution, aspect ratios, and EXIF data
Google’s stated specs are: JPG or PNG, between 10 KB and 5 MB, minimum 250 x 250 pixels, recommended 720 x 720. I aim for 1600 pixels on the long edge, which gives Google room to crop for different surfaces without softening the image. For aspect ratio, shoot 4:3 for general use and 1:1 for the cover, because square crops survive every thumbnail container Google currently uses. Leave EXIF data intact for genuine photographs (it is one of the signals Google uses to tell real shots from generated ones), but you do not need to engineer it.
Geotagging without tripping spam filters
Here is where I push back against some common advice. You will read on forums that you should manually edit EXIF GPS coordinates to match your business address exactly. Do not. Modern phones already record the coordinates, and editing them produces inconsistencies that look algorithmically suspicious. If you took the photo at your business, the location data is already correct. If you took it elsewhere, do not pretend otherwise; the image is probably not going to be useful for the listing anyway.
| Photo type | Recommended resolution | Aspect ratio | Replacement cadence |
|---|---|---|---|
| Cover / exterior | 1600 x 1600 px | 1:1 | Every 12 months or after any facade change |
| Interior context | 1600 x 1200 px | 4:3 | Every 6 months |
| Team | 1600 x 1200 px | 4:3 | After any staff change; otherwise annually |
| Product / detail | 1600 x 1600 px | 1:1 | Monthly for seasonal businesses, quarterly otherwise |
| Video clip | 720p or 1080p | 16:9 or 9:16 | Every 6 months, max 30 seconds, under 75 MB |
Filename and alt text patterns that work
Google Business Profile does not expose an alt text field for uploaded photos, which surprises people. What it does index is the filename and surrounding context. Rename “IMG_4823.jpg” to “dental-practice-reception-bristol-bs1.jpg” before uploading. The filename is one of the weaker signals in the ranking stack, but it is free and it takes ten seconds per image. Alt text does matter when those same images are repurposed onto your own website, into a Google Business post, or into a third-party citation profile like the business directory, where the surrounding metadata is exposed to search crawlers and shapes how the image is tied to your business name and location.
Did you know? Google’s official photo upload requirements are JPG or PNG only, file size between 10 KB and 5 MB, and a minimum resolution of 250 x 250 pixels (Google Business Profile Help).
Evidence from the 2025 ranking studies
I have been writing about local search since 2017, and I have a rule: do not cite a “study” unless you can name the dataset and the methodology. The local SEO space is awash with vendor-funded surveys with sample sizes of 200. The studies below are the ones I would actually defend in a meeting.
BrightLocal’s photo frequency findings
BrightLocal’s annual Local Consumer Review Survey and their Local Search Industry Survey, taken together, consistently report that businesses in the top three of the local pack have, on average, between three and five times as many photos as businesses ranked 11-20. Correlation, not causation. But the consistency of the gap across industries and years suggests that photo count is at minimum a strong proxy for whatever Google is actually measuring: engagement, profile completeness, recency of owner activity. My pragmatic read: target the median photo count of the businesses currently in the top three for your category, and then exceed it by 25%.
Whitespark data on image-driven conversions
Whitespark’s Local Search Ranking Factors survey has, since 2023, ranked “photos uploaded by owner” inside the top 20 factors for local pack rankings. That is not nothing, but their qualitative data on conversion is more interesting: businesses that added a structured photo set (the twelve-shot framework I described above) reported, on average, double-digit percentage increases in calls within 60 days. The sample is self-selected, so treat the numbers as directional. The direction, though, is clear.
Case study: a dental practice that doubled calls
The dental practice in Figure 2 is a real client, anonymised. South-west England, three-dentist practice, NHS and private. When I took the audit, they had eleven photos on Google Business Profile, of which six I would have deleted on sight: two were watermarked stock images of teeth, one was a screenshot of their old website, three were group photos from a charity bake sale in 2019. Cover image: a low-resolution logo on a white background.
We did one photography session, three hours, with a local photographer who charged GBP 340. Twelve shots, exactly to the framework above. Uploaded over four days (I will explain the cadence in a moment). Within twelve weeks, monthly calls had risen from 47 to 94. Direction requests up 71%. Website clicks up 52%. They changed nothing else: same opening hours, same services, no new reviews of note, no Google Ads. The only intervention was the photo set.
Now, this is one practice. N=1. I am not claiming every business will double their calls. What I am claiming is that the mechanism is reliable enough, and the cost low enough, that not doing it is leaving money on the table.
What if… you operate from a shared office, a home address, or a co-working space, and you genuinely cannot photograph a traditional storefront? Lean into the team and product categories instead. Six shots of work in progress at client sites (with permission), three of the team in a recognisable working context, three close-ups of deliverables or tools. Google’s algorithm does not require a storefront; it requires evidence of a real, active business. Searchers feel the same way.
Your 48-hour implementation plan
Reading this without acting on it is the most expensive option. Here is what to do in the next two days, with a phone you already own. The professional shoot can come later; the audit and the first upload pass cannot.
sequenceDiagram participant Owner participant Google participant Bing participant Apple Owner->>Google: Day 1, upload shots 1-3 (exterior) Owner->>Google: Day 2, upload shots 4-6 (interior) Owner->>Google: Day 3, upload shots 7-9 (team) Owner->>Google: Day 4, upload shots 10-12 (product/detail) Owner->>Bing: Day 2, upload shots 1-3 (1-day lag) Owner->>Bing: Day 3, upload shots 4-6 Owner->>Bing: Day 4, upload shots 7-9 Owner->>Bing: Day 5, upload shots 10-12 Owner->>Apple: Day 3, upload shots 1-3 (2-day lag) Owner->>Apple: Day 4, upload shots 4-6 Owner->>Apple: Day 5, upload shots 7-9 Owner->>Apple: Day 6, upload shots 10-12 Google-->>Owner: Listing shows updated photos Bing-->>Owner: Listing shows updated photos Apple-->>Owner: Listing shows updated photos
Shot list for a single phone session
Block 90 minutes mid-morning on a bright but overcast day. Wear what you would normally wear to work. Take, in order:
- Front door from across the street, level with the door
- Building in street context, taken from 20 paces back
- Approach view from the direction most customers arrive
- Entrance threshold looking in
- Reception or first-contact point
- Main service area (treatment room, dining floor, workshop)
- One detail shot of a tool, product, or finished work
- One detail shot of branded material (sign, menu, packaging)
- Wide shot of team in the workspace, candid not posed
- Tighter team shot of two or three staff at work
- One “in motion” shot of service being delivered (with consent)
- One closing shot: the exit, a thank-you sign, a satisfied detail
Shoot each at least three times. Pick the sharpest. You will have your twelve in under two hours.
Upload cadence across Google, Bing, and Apple
Do not upload all twelve at once. Google’s algorithm appears to reward sustained owner activity more than one-time bulk uploads (this is consistent across BrightLocal and Whitespark observations, though neither has published a controlled experiment). Upload three per day across four days. Use the same set, in the same order, on Bing Places and Apple Business Connect, with a one-day lag so the listings are not perfectly synchronised. Verify each profile if you have not already; Google Business Profile Help.
Did you know? Your photos and videos only appear on Google after you verify your Business Profile. Unverified businesses cannot display photos in search results, regardless of how many they have uploaded (Google Business Profile Help).
Monthly refresh habits that compound
Set a recurring calendar entry on the first Monday of every month. Twenty minutes. Take two or three new photos: a seasonal product, a new staff member, a recently completed job, a refreshed window display. Upload them. Delete one weak photo from the existing set if the average quality has drifted. That is it. The compound effect over twelve months is large, and the discipline is what separates listings that climb the pack from those that plateau.
Quick tip: Enable customer photo uploads and respond to them, but do not solicit them in a way that produces low-quality bulk. A handful of authentic customer shots is worth more than fifty staged ones, and Google’s spam filters are increasingly good at telling the difference.
Myth: Once you have a good photo set, you can leave it alone. Reality: Recency is itself a signal. A listing whose most recent owner-uploaded photo is from 2022 looks dormant, regardless of how good the photos are. Monthly refreshes keep the listing algorithmically alive.
Did you know? Google deprecated its public list of photo categories (Interior, Exterior, At Work, Team, Menu, Products) several years ago, but analysis from Igniting Business confirms those categories still map closely to what high-performing listings actually upload.
One last thing. The chiropractor I mentioned at the start. We did her shoot on a Tuesday in April. By the following Tuesday she had four new patients who specifically mentioned finding her on Google Maps. By the end of that quarter she had to hire a second associate to manage the demand. She still sends me a Christmas card. The photos cost her GBP 280 and an afternoon of her time. Book the shoot this week.
