Digital asset management with AI: the last mile

That gap is the real digital asset management problem. Not the folder on your server.

The library is the easy part

Digital asset management is the system that stores your visual content, keeps track of versions, and makes everything findable. It should be able to answer four questions:

• •Which image is the right one, and who approved it?
• •What am I allowed to use it for, where, and for how long?
• •How do I find it without knowing the file name?
• •How does it reach the people outside my team who need it?

Almost every article on this subject covers the first three. The fourth one is where brands that sell through others actually lose. We will come back to that.

What AI genuinely does well now

A few things have become real rather than promised:

Metadata on the way in. Images arrive with tags and a description instead of waiting for someone to add them later, which in practice never happens.

Search by what you can see. You look for “tap in a dark bathroom” rather than for IMG_4471_final_v2.jpg.

Derived variants. One approved image becomes the set of formats your webshop, your marketplaces and your social channels need, without anyone cropping by hand.

Flagging what breaks the rules. Wrong colour, wrong logo, wrong crop, caught before it goes out.

What we have learned from doing this at volume: treat AI as an assistant, not as the owner of your library. A wrong label on two thousand images is much faster to create than to undo. Keep a person in charge at the moments that matter: approval, usage rights, and the moment an image leaves the building.

What AI does not fix

AI makes your library searchable. It does not make it trustworthy. For that you still need three things that have nothing to do with AI.

One source of truth. Everything outside it is a copy. Without that agreement you keep having the same argument about which folder is the real one.

A status on every asset. Draft, approved, or retired. A retired image should not quietly disappear; it should be visibly retired, so the people still using it notice.

Provenance. Which assets were AI-generated, from what reference material, and who signed them off. This is becoming a compliance question as well as a housekeeping one: the transparency obligations in Article 50 of the EU AI Act apply from 2 August 2026 to AI-generated content. If you sell into the European Union, that applies to you whether or not you are based there.

None of this requires software. It requires an afternoon and a decision.

The last mile

Here is the part that gets skipped.

If you sell through dealers, retailers and marketplaces, your real asset library is not your server. It is the sum of every product page your product appears on, and you usually have no view of it at all.

What we run into at almost every brand that sells through a network:

• •The WeTransfer link from the product launch expired, so the dealer is still using last year’s images.
• •A two-year-old shared folder sits with fifty sales partners, and nobody knows which versions are in it.
• •A dealer cannot find the image, so they take one from Google, with the old logo or a different colour.
• •The lifestyle photography you invested in never reaches the dealer page. Only the packshot on white does.

Selling across borders adds another layer. A German dealer needs shots with German packaging. A Danish retailer has its own image specification. One marketplace crops square, another insists on a white background, and a third compresses everything you upload. Every market drifts a little further from what you approved, quietly, and nobody reports it because from their side nothing is broken.

So our position is simple: production and management are not finished until distribution is solved. An image your dealer does not use has no value, no matter how well your library is organised.

Two examples from our own work

Both are Dutch manufacturers, and both say something about sequence rather than about one good picture. You can see more of them on our use cases page.

Xenz | Beterbad Sanitair makes baths and taps. We produced print-ready bathroom scenes from their technical drawings, within three days. The three days got the attention, but the part that mattered came afterwards: a repeatable route so that every new product goes through the same steps and comes out in the same style.

DHK Kozijnen sells windows and doors. Instead of showing products in isolation, we placed them in realistic Dutch homes. The images worked, but they only worked at scale once the entire range had the same treatment and looked the same everywhere it appeared.

In both cases the value was not in the first image. It was in the tenth, the fiftieth and the two hundredth looking like it.

A sequence that works

You do not need a six-month programme. A workable order:

• •Start with your best sellers, not with your archive. The archive can wait, possibly forever, and that is fine.
• •Define what “complete” means. Per product: which images you want as a minimum, and in which formats per channel and per market.
• •Name one source. Everything else is a copy with an expiry date on it.
• •Write the brand rules before you generate. Colours, materials, light, setting, what is in and what is out.
• •Fix the last mile. Give your partners one door that always holds the current set, instead of links sent by email.

Step five returns the most and gets postponed the most.

Where LensLab fits

We built LensLab as one layer rather than three separate tools, for exactly this reason. You can see how it works in more detail.

Create. Our creators produce your visuals with AI, with human quality control all the way through to delivery. If you have your own designer, they can work in our tooling directly.

Manage. Everything lands in a central content library, with brand consistency based on your Brand DNA.

Distribute. Through the distribution module you make your visual material centrally available to dealers, retailers and other sales partners, so they are always working with the current set. Social channels connect directly, so images go out from LensLab itself.

To be straight about what is not there yet: Shopify and WooCommerce integrations are planned for November 2026. Insight into which images are actually being used is in development. And we are not a classic enterprise DAM with hundreds of integrations and a year-long implementation. We are built around product visuals for brands that sell through a network.

In closing

Digital asset management sounds like a tidying job, and partly it is. But the return is not a tidy folder. It is the confidence that the image you approved is the image your customer sees on a dealer’s site in a country you have never visited.

Curious what that looks like for your range? Try LensLab for free with five credits, or have a look at what we made for other brands first.