There's a piece of advice that circulates on every freelancer forum: fill out your profile completely, and that includes your rates. Be transparent. Save everyone time. Look professional. And for a lot of translators in high-supply pairs, that's reasonable — if a hundred people can do your job, a public number is just how you get shortlisted before someone else does.
But I stopped publishing a fixed rate years ago, and the longer I work in a thin-supply language, the more convinced I am that the standard advice quietly works against people like me. There's a live discussion on ProZ about exactly this — experienced freelancers deliberately leaving rates off their profiles because, as one put it, some clients surprise you by offering more than you'd have naturally stated. That's not a trick. It's an honest observation about how information asymmetry actually flows in this business, and who it usually favors.
A public rate is a ceiling wearing the costume of a floor
When you post "$0.12 per word," you think you're setting a minimum. You're not. You're setting a maximum. No project manager has ever emailed to say, "I see your rate is twelve cents, but this job is genuinely worth eighteen, so we'll pay that." The number you publish becomes the number you get, and often the starting point for a negotiation downward.
The published ranges bear this out. The figures floating around the industry right now put freelance work somewhere between $0.10 and $0.30 per word, with hourly rates from $20 to $75 — a spread so wide it's almost useless as guidance. That range isn't measuring skill. It's measuring positioning. The person at $0.30 and the person at $0.10 may be equally competent; what differs is whether they anchored the conversation or let the market anchor it for them.
Here's the part that matters for a rare pair. The value of a piece of Turkmen work is set far more by the client's situation than by my keystrokes. A pharmaceutical company with a regulatory deadline in Ashgabat and no other qualified vendor is not price-shopping the same way a content mill filling a language grid is. If I've publicly committed to one number, I've thrown away my ability to read the room. I've told the pharma client and the content mill that I'm worth exactly the same to both, which is false in a way that costs me real money.
Scarcity changes the math, but only if you let it
I've written before about being a single-vendor language — the strange position of sometimes being the only working Turkmen translator a given agency has ever found. That scarcity is leverage, and a published rate throws the leverage in the bin. If you're genuinely hard to replace, the appropriate response to "what's your rate?" is a question of your own: what's the document, what's the deadline, what's it for. Legal filings, oil-and-gas safety documentation, and medical content aren't priced the same, and they shouldn't be, because the consequence of getting them wrong isn't the same and the pool of people who can get them right is even smaller.
This is also where the AI conversation lands, correctly, in favor of the specialist. General translation is being automated, and the honest response to that isn't panic — it's to be clearly, demonstrably in the category machines can't own. Regulated, high-liability, low-resource work. But you only capture the premium that specialization earns if your pricing can move with the job. A flat public rate flattens exactly the distinctions your expertise is supposed to charge for.
None of this is an argument for being cagey or for pulling numbers out of the air. When a serious client asks, I answer within the same business day, clearly, with a rate tied to the specific work. What I decline to do is broadcast a single figure to the entire internet and then live inside it for the next three years. Withholding the number from the profile isn't opacity. It's refusing to negotiate against myself before the conversation has even started.
What to do instead
Three things have served me better than a published rate.
First, quote per project when you can. I've argued elsewhere that the per-word model is the wrong instrument for the next decade, and pricing on outcome rather than volume is easier when you haven't already boxed yourself in with a public unit price.
Second, build inflation into your model as a matter of routine, not crisis. The industry is finally saying this out loud — inflation affects your living and business costs whether or not your rate moves — and it's far easier to nudge a private, per-client number upward each year than to publicly revise a figure that clients have bookmarked.
Third, treat direct clients and agencies as different pricing universes, because they are. The end client who comes to you without an intermediary has margin the agency has already spent. If your public rate is calibrated for agency work — and most are — you've pre-discounted every direct inquiry that will ever reach you.
Transparency is a virtue. But transparency about your rate to strangers isn't the same as fairness to yourself. Keep the number in your pocket, and take it out when you know what it's for.