What is an editorial gap audit for AI fashion campaigns and how do you run one?

An editorial gap audit is a structured QA pass on your AI-generated fashion campaign that surfaces what's missing or off — shot-type gaps (back shots, still lifes, fragments, atmospherics), and drift gaps across geometry, fabric, fit, and brand mood. You run it by querying the agent for missing editorial coverage, then probe-testing one shot against four truth dimensions before batch generation.

Run the audit in five ordered steps, before you commit credits to the full shot list.

1. Query the agent for missing shot types. After your initial shot list is drafted, ask the invideo agent: "what's missing for a proper editorial?" This surfaces the categories a conventional lookbook brief routinely omits — back shots, product still lifes, cropped body fragments, reclined and seated poses, voyeur framing, empty atmospherics, and duo compositions. Product still lifes (garments and accessories with no human subject) are the "pause beats" that let a campaign breathe and are systematically missed by AI-generated lookbooks. Add the missing categories to the shot list before generation.

2. Build a four-dimension truth check per shot. For every garment in your campaign, write down what "correct" looks like across four axes the audit will score against: geometry (silhouette and pattern proportions match the actual garment), texture (fabric weight, weave, reflectivity render as the real material would), fit (drape and body proportions are plausible), and brand mood (lighting, set, and tone match the campaign's visual code). Upload reference angles — close-up, front, side, back, fabric detail, and on-model — so the agent has the ground truth to be audited against.

3. Run a single-shot probe before batching. Pick one full-frame shot that contains character, garment, set, skin, and pose, generate just that one, and audit it against the four dimensions. If geometry holds, fabric reads, fit is plausible, and the mood lands, the system is ready to scale. If any dimension fails, fix it at the system level (reference inputs, depth-structure rules, lighting direction) before generating anything else — skipping the probe propagates the same failure across the whole campaign.

4. Score each generated shot against three clip-evaluation questions. As shots come back, apply the same three-part check to each: does the weave/material read as the actual fabric, is the color accurate, and does the garment-body-environment interaction look physically plausible? Lock the ones that pass; regenerate the ones that fail with a targeted note (not a full re-prompt). Validate at three focal distances — close, mid, wide — so silhouette drift between distances doesn't slip through.

5. Audit coverage and connective tissue at the campaign level. Once the shot list is generated, check that every garment in the catalogue appears at least once, that coverage is balanced across all locations, and that graphic matches (visual echoes between product shape and environmental elements) tie sequences together. The invideo agent — an agentic video tool with all current image and video models routed inside — automatically balances catalogue coverage and location distribution if you ask it to, which is the easiest place to catch a missing-garment gap.

In one documented editorial campaign run this way, the audit pass surfaced enough additional shot types (still lifes, fragments, back shots, atmospherics) to expand a standard lookbook brief into 40 editorial stills and 30 motion clips across two models and five locations — produced in 3–4 hours for roughly 630 credits (~$150), rejected clips included. The audit is the step that turns a lookbook into an editorial.

Watch some of these to see what works for you:

Push past the lookbook. Jam with the agent to see what you're missing for a 'proper editorial'.

— invideo's creative team