Ecommerce product pages
- Square and portrait crops from one master
- Responsive widths in modern formats such as WebP and AVIF
- A full-resolution zoom image
Digital Asset Management
Digital asset management software development
Product photos, campaign images and video end up in shared drives with no record of who shot them, what the licence allows or when it runs out. Agnotic Technologies builds digital asset management software that keeps metadata and rights with every file, from upload to the storefront.
IPTC and XMP metadataRights and expiry checksRenditions and CDNAI tags people approve
Trusted by global innovators
Digital asset management
What is digital asset management software?
Digital asset management software is the library where a business stores its images and video with the metadata, usage rights and approved versions each channel needs. Agnotic Technologies builds custom digital asset management software for brands, retailers and publishers, with embedded IPTC and XMP metadata, rights checks, renditions and links to your PIM and CMS.
The hard part is rights, not storage. A photo licensed for one country, one year and print only must never reach a global web campaign, and the software has to know that without anyone remembering it.
This page covers managing media. Creating images and video with generative AI is on our media generation page, and clinical imaging such as DICOM studies is on our medical imaging PACS page.
Rights and metadata table
Which rights and metadata fields does each asset need?
The field table behind the upload form and every publishing check. Field names follow the IPTC Photo Metadata Standard and are written into the file as XMP, so they travel with it.
| Asset field | Metadata standard field | Who fills it | What the system blocks when it is missing |
|---|---|---|---|
| Creator | IPTC Creator (dc:creator) | Photographer or agency at upload | Publishing to any channel |
| Credit and copyright | IPTC Credit Line (photoshop:Credit) and Copyright Notice (dc:rights) | Rights team or agency | Use on channels that must show a credit |
| Licence terms and licensor | IPTC Rights Usage Terms (xmpRights:UsageTerms) and Licensor (plus:Licensor) | Rights team, from the contract | Any use until a licence is linked |
| Territory, channels and end date | Licence record in the DAM, linked from the asset | Rights team | Use outside the licensed markets or channels, and any new use after the end date |
| Model and property releases | IPTC Model Release Status and Property Release Status | Producer or photographer | Use of images showing people or private property without a release on file |
| Source type | IPTC Digital Source Type | Uploader, or the generation tool | Publishing AI-generated images where your policy requires a label |
| Alt text | IPTC Alt Text (Accessibility) | Content editor, from an AI draft they review | Publishing to web channels |
Territory, channel and end date usually live in the DAM's licence record rather than in the file, because one licence covers many assets and can change after they are downloaded.
Asset lifecycle
How does an asset move from upload to takedown?
Six stages, each with an owner and a record, so nothing is published without its rights in place.
Ingest
Files arrive from photographers, agencies and studios by upload, watched folder or API, and a perceptual hash catches near duplicates before they are stored.
Describe
Embedded IPTC and XMP fields are read, required fields are checked, and AI suggests keywords and alt text for a person to accept or edit.
Clear rights
The asset is linked to its licence, with territory, channels and end date, and any releases are attached.
Approve
Brand or legal reviewers approve the asset for named channels.
Publish
Renditions are generated and delivered through the CDN, the PIM or the CMS, and every place the asset is used is recorded.
Expire and take down
Before a licence ends, owners get a list of live uses; on the end date, delivery URLs stop serving and the asset moves to an archive.
AI tagging
How should AI tagging work in a DAM?
AI makes a large library searchable. People still own anything that affects rights, accessibility or the brand.
- AI suggests keywords, objects, colors and scenes, and a librarian accepts or edits them
- AI drafts alt text, and an editor approves it before web use
- Faces are never named automatically unless your policy and the law allow it
- Suggested tags are stored apart from approved ones, so search can use either
- Low-confidence suggestions go to a review queue
- The model and version are recorded with every suggestion
Renditions
Which renditions does each channel need?
Renditions are generated from one master file and cached at the CDN, so an image is edited once and stays consistent everywhere.
Marketplaces and partners
- Each marketplace's size and background rules
- Feeds that send the approved version only
- Withdrawal when the licence or the product ends
Social and campaigns
- Aspect ratios per placement
- Video cut-downs with captions
- Expiry tied to the campaign licence
Print and press
- Full-resolution master with color conversion for print
- Embedded credit and copyright
- Every download logged per user
What do we build into digital asset management software?
Modules we scope one by one, so the first release can cover one collection and one channel with rights checks already in place.
Ingest and deduplication
Upload, watched folders and API ingest, with checksums for exact copies and perceptual hashing for near duplicates.
Metadata schema and editor
IPTC Core and Extension fields plus your own taxonomy, read from and written back to each file as XMP.
Rights and licence manager
Licences with territory, channels and end dates, releases attached, and checks at every download and publish.
Renditions and CDN delivery
Crops, sizes and formats generated from one master and cached at the CDN, with signed or expiring URLs where needed.
AI tagging with review queues
Suggested keywords and alt text that people approve, with confidence thresholds per tag type.
Search and collections
Faceted search on metadata, rights and tags, saved searches and shared collections for campaigns.
PIM, CMS and ecommerce connectors
Approved assets linked to products in the PIM and pages in the CMS, with usage recorded both ways.
Content provenance
C2PA Content Credentials read on ingest and kept with the asset, so its recorded history is not lost.
Who it is for
Who do we build digital asset management software for?
Teams whose images and video are worth more when they can be found, trusted and reused.
Retailers and ecommerce teams
Product imagery for large catalogues, linked to the PIM and sent to the storefront and marketplaces in the right sizes.
Consumer brands
Campaign assets with licences per market, so agencies and regional teams use only what is cleared.
Publishers and media teams
Photo and video archives with credits, rights and syndication terms that travel with each file.
Manufacturers and distributors
Product photos, diagrams and documents shared with dealers through a portal.
Standards
Which metadata and provenance standards does a DAM use?
IPTC Photo MetadataXMP (ISO 16684-1)ExifC2PAWCAG 2.2
Facts and sources
Facts behind digital asset management software
The standards behind the field table, AI labelling and provenance features, each linked to its primary source.
The IPTC Photo Metadata Standard defines IPTC Core and IPTC Extension properties, implemented with Adobe's XMP technology, now an ISO standard, for data about the people, locations and products shown in an image, dates and identifiers of its creation, and rights information.
Source: IPTC Photo Metadata Standard (IPTC)The IPTC Photo Metadata Standard 2021.1 added the Alt Text (Accessibility) and Extended Description (Accessibility) properties, and the 2025.1 version, published in November 2025, added four properties for AI-generated content: AI Prompt Information, AI Prompt Writer Name, AI System Used and AI System Version Used.
Source: IPTC Photo Metadata Standard, what is new by version (IPTC)In the C2PA technical specification, statements about an asset's creation, edits and capture device are assertions; they are wrapped into a digitally signed claim and bound into a C2PA Manifest, and the set of manifests in the asset's Content Credential is its provenance data.
Source: Content Credentials: C2PA Technical Specification 2.4 (C2PA)WCAG 2.2 success criterion 1.1.1, Non-text Content (Level A), requires all non-text content presented to the user to have a text alternative that serves the equivalent purpose, with listed exceptions.
Source: Understanding SC 1.1.1 Non-text Content (W3C)
Scope and cost
What drives the cost of digital asset management software?
We do not publish prices. These are the things that change the size of a build, and we price them on the scoping call.
Library size and file types
How many assets, and whether video, layered design files and 3D sit beside photos.
Migration
Moving files and metadata out of shared drives or an existing DAM, with fields mapped and cleaned.
Rights model
How licences, territories, channels and releases work in your business today.
Connectors
The PIM, CMS, ecommerce platform and creative tools the DAM exchanges files with.
AI tagging
Whether AI suggestions are in the first release, and the review queues they need.
Delivery
CDN setup, rendition volume and video transcoding.
Engagement
How does a DAM engagement work?
We start from a sample of your real library, not an empty schema.
1.
Scoping call
We sample your library, licences and channels, and agree the metadata schema the first release needs.
Scope2.
Fixed-scope first release
Ingest, metadata, rights checks and delivery to one channel, with one collection migrated end to end.
Build3.
Scale and support
More collections, channels, connectors and AI tagging, with support after launch.
ScaleFrequently Asked Questions
Buy a packaged DAM when its metadata, rights handling and connectors fit how you work. Build when your rights rules, product catalogue links or delivery needs would force a packaged system out of shape, or when the DAM is part of a product you sell.
Yes, where the file format supports it. We write IPTC fields into the file as XMP on download, so creator, credit, copyright and licence details travel with the image outside the DAM.
Every licence has an end date. Before it arrives, owners see a list of live uses; on the date, new downloads are blocked, CDN URLs for the asset stop serving and the asset moves to an archive.
Yes. AI can suggest keywords, objects and alt text across the backlog, stored as suggestions. Librarians approve them in batches, starting with the collections people search most.
Usually. We export the files and any metadata the current system holds, map fields to the new schema, flag assets with missing rights data and migrate in batches your team can check.
Assets are linked to product records in the PIM by SKU or product ID, and the storefront pulls approved renditions from the CDN. When an asset changes, every linked product gets the new version.
Yes. Video is transcoded into the formats each channel needs, with thumbnails, caption files and the same rights and expiry checks as images.