Why StoryPress Doesn't Count Your Languages

This month we're shipping our native CMS. Buried in it is a feature that is barely a feature: every story can carry language variants — field-level overrides where only a headline needs translating, full variant rows where a market needs its own page. Add as many languages as your business can use. The meter doesn't move.

In most of the CMS market, that last sentence is the strange one. Webflow bills per locale. Enterprise platforms tier their plans by locale count. Whole pricing pages rest on the premise that a language is a billable unit, like a gigabyte. We think that premise died a while ago and nobody sent flowers. Consider this the eulogy.

A meter is a claim about cost

Every meter on a pricing page makes a claim: this thing costs us money in proportion to your use of it, so we pass it through. Metering bandwidth makes that claim honestly — serving traffic costs real money, and more traffic costs more. Metering storage: same. These are the boring, defensible meters, and StoryPress uses both.

Metering languages makes the same claim, and the claim is false. A language variant in a modern CMS is rows in a database and cached documents on an edge network. The fortieth language costs the platform roughly what the second did: approximately nothing. A meter attached to a unit with no marginal cost behind it isn't a meter. It's a toll booth on a road somebody already paid for.

Locale fees made sense once

In fairness, the toll booth used to guard a real bridge. Professional human translation still runs about $0.15–$0.30 per word for most business content (Smartling's 2026 rate guide), and before machine translation was any good, that was the only kind there was. At those rates, a 20,000-word site in ten languages was a mid-five-figure capital project before hosting even entered the picture. On the platform side, locale support was hand-built infrastructure: separate render paths, duplicated templates, translation-workflow tooling some team had to keep alive. Charging per locale mirrored genuine cost on both sides of the transaction.

That's the world these meters were born into. Then both of its load-bearing costs collapsed.

AI translation collapsed the cost of creating variants. What was twenty cents a word is now an API call plus an editorial review pass — you should absolutely still have a human read the output, but reviewing is an order of magnitude cheaper than translating. Meanwhile, edge economics collapsed the cost of serving variants. A localized page is a cached document on the node nearest the reader. It doesn't need a bigger server in proportion to how many languages you publish.

Both cost curves fell off a cliff. The meters, curiously, did not. They survived because they bill beautifully: a per-locale fee scales with your success rather than the platform's costs, renews monthly, and is invisible to whoever approved the original plan. A meter like that doesn't get removed by accident. It has to be removed on purpose.

What the meter actually taxes

Here's the perverse part: what per-locale pricing discourages is exactly what the data says you should do. CSA Research surveyed 8,709 consumers across 29 countries and found that 76% prefer to buy products with information in their own language — and 40% won't buy from a website in another language at all (Can't Read, Won't Buy — B2C, CSA Research, 2020). Shopify reports that brands see an average 13% conversion lift when a storefront is shown in the buyer's local language (Shopify, as of July 2026). Languages aren't decoration. For a small site, they're some of the cheapest revenue growth on the menu.

Now put that growth behind a meter. On Webflow, the Localization Essential add-on runs $9 per locale per month and caps at three extra locales; Advanced runs $29 per locale per month and is capped too (as of May 2026, per ClearBrand's Webflow localization guide — Webflow's add-on pricing has shifted more than once, so check the live page). Five locales on Advanced is $145 a month before a single word is translated, on top of the site plan. So teams ration. They ship the German site and shelve the Dutch one — not because Dutch customers aren't worth having, but because the meter turned them into a line item.

A cap on languages is a cap on your addressable market, sold back to you as a plan feature. We keep a running index of who charges what per language in our CMS language pricing index.

The industry is starting to say it out loud

This isn't just us grumbling into our terminals. Today — July 12, 2026 — Netlify deleted seat-based pricing entirely: Pro went from $20 per seat to a flat $20 a month with unlimited seats, with usage rates adjusted so most bills stay flat or drop (“The end of seats,” Netlify, July 12, 2026). Their reasoning, verbatim: “Seat pricing made sense for us when software was written by a handful of developers. It makes far less sense when anyone, including any AI agent, can build software.”

Swap “seat” for “locale” and “developer” for “translator,” and the sentence still scans. Seats and locales are the same species: artificial units that once tracked a real constraint and now track only billing ambition. The whole genus is headed the same way — we made the longer argument in The meter is dying — but it's pleasant to watch a platform of Netlify's size arrive at the same place in public.

What we built instead

The native CMS treats language as a property of content, not an add-on with its own invoice. Every story supports two kinds of localization. Field-level translations override just the fields that differ — a headline, a CTA — while everything else flows through from the default language. Full language variants are separate rows with their own body, their own SEO metadata, their own everything, for markets that need a genuinely different page. Hreflang tags, localized sitemaps, and per-language structured data are wired up automatically, because that's plumbing, not product.

There is no per-locale fee. There is no cap on variant count. Not “generous limits.” No cap. If your business needs forty languages, the CMS's entire response is: fine.

It also composes with the rest of the platform. StoryPress ships an MCP server, so an AI agent can edit your site conversationally — which means “translate this launch post into the six languages we sell in” is one instruction, not six invoices.

What we do meter

We are not claiming there are no meters. That claim is always a lie with a delay on it. A forty-language site stores more content than a one-language site and serves more traffic, and storage and bandwidth cost real money at every scale, ours included. So those are metered — plainly labeled, with sensible caps you can read without a magnifying glass or a solutions engineer.

The rule we hold ourselves to is simple: meter resources, never units. If it appears on our infrastructure bill, it can appear on yours, in proportion. If it doesn't — languages, locales, seats — it can't. StoryPress is $5 a month, billed $60 yearly, and that's the whole pricing page.

The short version

• StoryPress's native CMS puts language variants on every story: field-level overrides and full variant rows.

• There is no per-locale fee and no cap on the number of language variants.

• StoryPress meters real resources only — bandwidth and storage — with plainly-labeled caps. $5/month, billed $60/yearly.

• Professional human translation still runs about $0.15–$0.30 per word (Smartling, 2026); AI translation reduced variant creation to an API call plus review.

• 76% of consumers prefer buying in their own language; 40% won't buy in another language (CSA Research, 8,709 consumers, 29 countries, 2020).

• Shopify reports an average ~13% conversion lift from showing a storefront in the buyer's language (Shopify, as of July 2026).

• Webflow Localization: $9/locale/month (Essential, capped at three extra locales) or $29/locale/month (Advanced, also capped), as of May 2026 (ClearBrand).

• Netlify ended seat-based pricing on July 12, 2026 — Pro is now $20/month with unlimited seats (Netlify blog).

If you're pricing out a multilingual site right now, run the math with the meter and without it — our CMS language pricing index has current numbers, and The meter is dying has the longer argument. And if you'd rather just publish in every language your customers speak, get started. We won't be counting.