Short answer: judge every gallery CRM against twelve points — art data model, proposals, multi-location, client portal, API, data ownership, EU/GDPR hosting, roles, audit trail, reporting, migration, true cost. For each point there is one question to ask the vendor and one red flag that should end the conversation. This page lists all twelve, ends with a copyable checklist, and works with any vendor — including us.
The honest way to choose an art gallery CRM is to score every candidate against the same twelve points and let demonstrations — not brochures — decide. Each point below has the question to put to the vendor, word for word, and the red flag that means trouble. The method is vendor-neutral: it works on Artlogic, on vertical suites, on an Odoo build, and on us. Where our own platform answers a point, we say so and name the module — check us the same way you check everyone else.
The twelve points
1. An art-specific data model
A gallery system must hold provenance, certificates, editions and condition as structured records on the artwork — not as text in a notes field. Everything else in the system depends on this foundation.
- Ask: "Show me one artwork where I can see its provenance, its certificate and its edition children — in one screen."
- Red flag: the artwork is a generic "product" row and provenance lives in a free-text note.
2. Proposal generation from the catalogue
Offers are a gallery's daily output. They should assemble from real records — images, prices, availability — and become a branded PDF without leaving the system.
- Ask: "Assemble a three-work offer from my catalogue and generate the branded PDF, live, during this demo."
- Red flag: "You export to Word and lay it out yourself" — or the PDF feature is promised for a future release.
3. Multi-location stock
Works travel: two spaces, a storage facility, a fair booth. The system must know where every piece is now and record the movement.
- Ask: "Show me where all works are located right now, and move one from the gallery to the fair — with a trace."
- Red flag: "Location" is a single text field typed by hand on each artwork record.
4. A client-facing portal
Collectors increasingly expect a private page for their offer, not a PDF thread. The portal should present branded offer pages the gallery controls.
- Ask: "Can a collector open a private page with their offer — and can I see when they viewed it?"
- Red flag: the portal is "coming next year" — anything on a roadmap instead of in production does not count.
5. An open API
Your CRM must talk to your website, your accounting, your tools. The strongest sign of a real API is that the vendor's own product runs on it.
- Ask: "Which API runs your own interface, and can we call the same one? Show me the documentation."
- Red flag: the API exists only on the most expensive tier — or exists only as a slide.
6. Data and code ownership
Your catalogue is your gallery's capital. It should live in a standard format you can leave with — and the platform should not hold it hostage.
- Ask: "In what format does our data physically live, and what exactly do we receive if we leave?"
- Red flag: a proprietary database format, or exports offered as a paid extra per request.
7. Hosting and GDPR
Collector records are personal data. Residency, backups and access discipline are legal questions, not features — EU hosting deserves its own scrutiny.
- Ask: "Where exactly are the servers, where are the backups, and who inside your company can access our data?"
- Red flag: "the cloud" as an answer — no region, no backup schedule, no named access policy.
8. Roles and permissions
Assistants, registrars, partners and directors should not share one login. Permissions should be granular enough to hide prices from some eyes and nothing else from others.
- Ask: "Create an assistant profile that sees the stock but not the purchase prices — and prove it by logging in."
- Red flag: all-or-nothing access, with fine-grained roles "planned".
9. A full audit trail
When a price changes or a record disappears, "who did what, when" must be answerable in seconds. This is what makes multi-user work trustworthy.
- Ask: "Show me who changed this price, when, and what the previous value was."
- Red flag: history kept on a few record types only — or none at all.
10. Reporting and dashboards
Every gallery reviews the same few numbers monthly: sales, stock ageing, pipeline, fair performance. Those views should be configurable without a developer.
- Ask: "Build the two dashboard views my gallery reviews every month — during the demo, without writing code."
- Red flag: fixed reports; every change is a billable custom project.
11. Migration — the honest scoping
Moving from Excel or another system is a cleaning project, not a miracle. A serious vendor scopes it from a sample of your real file before promising anything.
- Ask: "Send us your current export — we will scope the migration from it and validate the import with your team."
- Red flag: migration priced as a large fixed project before anyone has looked at your data.
12. True cost — the three-year number
Licence, seats, storage, image tiers, support, migration, exit: the real price is the sum over three years, not the monthly headline. Our cost guide walks the full computation.
- Ask: "What does this cost in year three — with our team size, our storage, our support needs and our exit included?"
- Red flag: every essential feature is a separate unlock; the base price is a foot in the door.
The checklist, copyable
- Art-specific data model — provenance, certificates, editions as structured records
- Proposal generation — branded PDFs assembled from the catalogue, live in demo
- Multi-location stock — real locations and traced movements, not a text field
- Client portal — private offer pages in production, not on a roadmap
- Open API — the same one the vendor's own product runs on, documented
- Data and code ownership — standard format, full export, no lock-in
- EU/GDPR hosting — named region, backup schedule, named access policy
- Roles and permissions — granular, proven by a live login
- Audit trail — who changed what, when, previous value
- Reporting and dashboards — configurable without a developer
- Migration — scoped from your real file, validated with your team
- True cost — the year-three total including seats, storage, support and exit
How Art Gallery CRM answers these twelve
For transparency — and because a checklist should be applied to its author — our answers, each mapped to a module running in production:
- Data model: artists, artworks, price history per deal, opportunity timelines artgallery_custom_fields
- Proposals: multi-artwork proposals and branded server-side PDFs art_proposal artwork_proposal_pdf
- Portal: client-facing offer pages custom_mail_templates
- Multi-location: multi-warehouse stock, inter-company transfers, multi-currency
- API: the secured REST API that powers our own React frontend custom_api_2FA
- Ownership: standard PostgreSQL, documented schema, full dumps on request, open-source Odoo 15 base
- Hosting: private EU hosting, GDPR compliant, daily off-site backups
- Roles: granular roles and permission profiles base_user_role base_user_role_profile
- Audit trail: every create, edit and delete logged with user and timestamp auditlog
- Reporting: configurable dashboards and XLSX reporting ks_dashboard_ninja
- Migration: imported during onboarding, validated with your team before go-live
- Cost: quoted on your number of locations and users — no per-seat metering
Verify all of it live in a demo — that is the point of the method. If you are still mapping the market, start with the six-category 2026 comparison; if you are leaving a specific tool, see the honest Artlogic alternatives review and the spreadsheet-to-CRM migration guide.
FAQ
What should I look for in an art gallery CRM?
Twelve things: an art-specific data model (provenance, certificates, editions), PDF proposal generation from the catalogue, multi-location stock, a client-facing portal, an open API, data and code ownership, GDPR-compliant EU hosting, roles and permissions, a full audit trail, reporting dashboards, a migration plan, and the real three-year cost. For each point, ask the vendor for a live demonstration on your own data — and walk away if the answer is a roadmap.
Which questions should I ask a CRM vendor?
One per criterion, always asking for a demonstration rather than a statement: show one artwork's provenance, certificates and editions; assemble an offer and generate the PDF live; move a work between two locations; show who changed a price and when; state where the servers and backups physically are; and quote the total year-three cost including seats, storage, support and exit. Vague or deferred answers are themselves the answer.
Do galleries need a custom CRM?
Not necessarily — they need a CRM that fits how a gallery actually works. A generic CRM fails on art-specific jobs: provenance, editions, certificate records, offers assembled from the catalogue. That fit can come from a vertical SaaS, or from a platform purpose-built on a customisable base such as Odoo. What matters is not custom for its own sake, but whether the twelve points in this checklist are covered in production today, not on a roadmap.
How long does it take to evaluate a gallery CRM properly?
A focused afternoon per shortlisted vendor is enough: run your three most repeated tasks — usually sending an offer, checking availability and updating a record — through a trial with a real artwork and a real collector, and ask the twelve questions. Most galleries can move from shortlist to decision in two to three weeks without rushing the trial.
Next reads: the six-category 2026 comparison, Artlogic alternatives reviewed honestly, spreadsheets vs a gallery CRM, and the full module-by-module comparison.
Every artwork, every collector, one workspace
Bring the checklist to the demo — hold us to all twelve points.
Book a private demoor write to [email protected]