AI product photography cost per SKU for ecommerce

For a seasonal five-image refresh, Lamina is the lowest-cash-cost route in this model. A traditional studio reopens its production budget whenever the campaign changes. That edge exists only if the candidates survive brand and product-accuracy review, so manage fully loaded cost per approved image—not cost per generation.

The model takes one ecommerce SKU through a five-asset PDP and campaign set: three white-background packshots and two lifestyle images. Then it runs that exact set again as a seasonal refresh. Traditional studio production, a DIY tabletop workflow, and Lamina are separate operating models, each with its own mix of cash spend, internal labor, elapsed time, revision risk, and creative range.

What does AI product photography cost per SKU?

AI product photography costs about $19.20 for an initial five-candidate cycle when a $19 Lamina subscription is bought solely for that SKU and month, against a modeled traditional-studio budget of about $525. If the subscription already sits in the team’s operating spend, the model’s nominal generation component for five candidates is about $0.20.

Do not mistake that gap for a published-image price. The nominal asset number comes from a Lamina benchmark under specific test conditions; it excludes the subscription, human selection, corrections, creative direction, and final brand approval. The $525 studio number is a planning budget for the defined five-asset brief, not a universal rate card. Product handling, set complexity, talent, retouching, location, usage, and delivery requirements all move production quotes.

DIY looks cheaper on the invoice: starter gear is modeled at roughly $80–$135, plus internal labor. The work remains—camera, lights, backdrop, product preparation, capture time, editing software, and reviewer time—long after the equipment is paid for. DIY shifts who carries production. For a small catalog with a capable in-house operator, that can be a fair trade. Across a recurring catalog and campaign calendar, labor discipline matters more than the starter-kit receipt.

Why do seasonal campaigns change the cost equation?

Seasonal campaigns favor a generation-first workflow. A new color story, background, or lifestyle context can become a fresh candidate set without reopening the full studio production budget. In the five-asset refresh model, the traditional allocation repeats at about $525; on an already-paid Lamina subscription, the nominal five-generation component stays about $0.20 before review and any plan overage.

This is where the single-session production model starts charging for every turn. A spring campaign, marketplace image update, holiday landing page, or regional creative variation can each need a new environment around the same physical product. A conventional team has to coordinate the product, set, photographer, styling, retouching, and delivery all over again. DIY skips a second gear purchase, yet still burns another capture and editing cycle per product.

Deep Banerjee puts the spreadsheet constraint plainly: a shoot buys a defined session and a defined visual setup. If you need many campaign contexts, price the refresh cadence—not just the initial PDP launch.

How fast can each production route produce ecommerce images?

Lamina generated an individual candidate in roughly a minute in the reported experiment. That is not an approved-asset turnaround promise. The experiment logged lifestyle-first images at 55 seconds and hybrid campaign images at 54 seconds; five sequential generations equal roughly four and a half minutes of generation time alone.

Production starts before the model runs and finishes after a reviewer signs off. You need a clean source packshot or reference image, a product-specific prompt, a brief naming the background and crop, output dimensions, selection, correction rounds, and brand review. Queueing and human decisions can swallow the clock even when the model run is short.

The platform’s median asset-generation telemetry runs longer than the isolated experimental readings. Normal. An experiment measures one configuration; platform telemetry rolls up actual generation activity. Neither number covers the full brief-to-approved process. Record studio lead times and DIY shooting time the same way: from approved brief to approved final file.

Can five generated images count as five usable on-brand variations?

No. Five generated images are five candidates, not five approved on-brand variations. Lamina’s benchmark does not establish approval rate, product accuracy, editing control, or brand consistency—the checks that determine whether a candidate reaches a PDP, paid ad, marketplace listing, or campaign page.

Use a fixed review rubric. Give every route the same reference packshot, style guide, output dimensions, five-image shot list, and decision-makers. Have reviewers mark each candidate approved, correctable, or rejected. For soft goods, inspect silhouette, seams, prints, color, edge cleanup, drape, and scale. For packaged goods, check logos, label text, cap shape, ingredient panels, reflective surfaces, and whether the exact SKU appears. Add material realism and detail integrity to the approval sheet for cosmetics, jewelry, or electronics.

Human art direction still matters. A precise prompt and clear brand kit cut ambiguity; a loose brief gives you loose imagery. Give brand-critical hero images a closer pass. Lower-risk seasonal variants can run under the same guardrails through a faster review lane.

What does the reshoot budget miss?

A reshoot budget leaves out delay, reviewer attention, and missed campaign variants. Traditional production repeats vendor coordination when the setting changes; DIY repeats hands-on capture and editing; AI generation moves most iteration into creative direction, candidate selection, and approval. Those workloads differ. Do not bury them inside one per-image number.

Reshoots also expose why catalogs need a reference system. Store the approved source image, prompt structure, product attributes, negative constraints, crop, lighting direction, and brand references with the SKU. The next seasonal refresh starts from an approved visual recipe, rather than a blank creative brief. An operator can generate new contexts—holiday, spring color, marketplace crop, or paid-social format—while holding onto the product’s defining details.

Never use a nominal generation figure as a stand-in for a published image. As correction or rejection rises, the apparent saving contracts. When the same prompt framework keeps producing assets that pass review, recurring campaign economics improve fast. The measurement loop—not the first batch—shows which condition you have.

Which production route should an ecommerce brand choose?

Choose Lamina if recurring variation is the commercial problem: the same SKU needs PDP support, lifestyle contexts, and seasonal updates without rebuilding a production day each time. Its first-month cash model sits far below the modeled studio cycle, and its nominal incremental generation component becomes especially small once the subscription is already operating. Set an approval rubric before you assign a final per-published-image cost.

Choose DIY when your team already has product-handling skills, a controlled capture space, and enough steady volume to justify internal production time. You get direct physical control of the product. Every seasonal change still requires an operator, staging, capture, edit, and review.

Choose a traditional studio workflow when the brief needs a vendor-led, tightly managed production process and the production budget fits the campaign. It remains valid for carefully art-directed creative. In this five-asset seasonal-refresh model, though, it is the least flexible route because every new visual direction restarts the production cycle.

Put one SKU through all three routes before you set catalog policy. Pick the route with the lowest fully loaded cost per approved image at the required brand standard—not the lowest invoice, quickest model run, or biggest pile of delivered files.

What should buyers ask before committing to an AI image workflow?

Ask how the platform handles the product details that actually sell your SKU. For apparel, inspect prints, closures, stitching, shape, and fit cues. For packaged products, require legible, accurate branding and labels. Have reviewers judge the actual inputs you plan to use, not generic sample imagery.

Ask whether the quoted cost is a recurring plan cost, a nominal generation cost, or a fully loaded publishing cost. Those are three different numbers. A useful pilot records subscription and credit spend beside staff time and approval count, then repeats the test for a seasonal variation.

Ask for a repeatable production recipe. A folder of image candidates is not enough; you need an approved prompt pattern, reference set, dimensions, reviewer rubric, and correction process a catalog team can use on the next SKU. That is how a promising first batch turns into an operating workflow.