How to choose a translation management system for your agency

Choosing a translation management system is less about feature checklists and more about fit: does it match how your agency actually runs work, and can you move onto it without losing a week? Here is a vendor-neutral way to evaluate the options.

Start from your workflow, not the feature list

Every tool demos well. The question is whether it models your reality: the way jobs move from quote to delivery, how you price by word and language, how margins are tracked, and who needs to see what. If a system forces you to bend your workflow to its assumptions, that friction never goes away.

Capabilities that actually matter

• A real job pipeline with states you recognise, and costs and margins that travel with the work.
• A CRM that separates the clients you invoice from the vendors you pay, each with their own rates and currencies.
• File handling that ties every source and deliverable to its job, with a change history.
• Quality assurance that can catch factual errors — numbers, dates, names — not just typos.
• Roles and permissions granular enough that clients and vendors can be given a limited view without seeing your internal workspace.
• Clear, honest data isolation between tenants, ideally enforced at the database level rather than in application code alone.

Ask the migration question early

The cost of switching is mostly the cost of moving your data. Ask exactly how your existing clients, vendors, and history come across, how long it takes, and who does the work. A vendor that migrates you — taking your current tool’s exports or spreadsheets and standing your account up for you — removes the single biggest reason agencies stay stuck on software they have outgrown.

Traps to avoid

• Paying per seat for people who never log in, such as clients and vendors.
• Quality features that only reformat text instead of verifying it against the source.
• Data-residency claims that are vaguer than your clients require — get the actual region in writing.
• A pricing model that charges for capabilities that do not yet exist.