They read your pricing page in English. They can't negotiate it in English.
Your prospects read English well enough to reach a demo. Speaking is a different skill, and it fails in the funnel stages where the buyer has to push back on you. Here is where that happens and how to run those calls in a language your team doesn't speak.
Your prospects almost certainly read English well enough to get through your site. Negotiating in it is a different skill. Reading is usually the stronger one, which means the language barrier in your funnel doesn't show up where you're watching for it.
It shows up late, in the stages where the buyer has to push back.
Why does your funnel make language look solved?
Your inbound signal comes from reading. Someone finds a comparison post, skims the pricing page, and books a demo. Every step is comprehension at their own pace, with the option to reread.
Then they join a call and everything inverts. Now it's production rather than comprehension — in real time, in front of colleagues, on a topic where being vague about money is embarrassing.
EF's English Proficiency Index carries the detail that matters here: across the countries it measures, speaking is usually the weakest of the four skills. People read better than they talk.
Japan's country page shows the split plainly: reading sits well above speaking. That gap is the one your funnel inherits.
One caveat before you lean on that. EF measures general adult English, not B2B buyers, and the people on your calls likely score above their national average. So don't transfer the level. Transfer the shape — wherever a population reads better than it speaks, your funnel inherits that asymmetry.
And here's what makes it hard to catch. A strained call doesn't look like a failed call. It looks like a polite one. Short answers. Agreement. No hard questions. Then the thread goes quiet and your CRM records "no response" against a deal that died because the buyer never got to say what was actually blocking it.
Which funnel stage actually breaks?
No survey measures this stage by stage. What follows is our read, sorted by how much the buyer has to produce rather than receive:
The pattern is simple. Pressure rises where the buyer stops receiving information and starts producing it. Which is also where your deal is won.
Two stages deserve a closer look.
Discovery is worse than negotiation, and it comes first. You need the buyer to describe how their team works, who else touches the decision, and what they tried before. That's long-form and full of internal vocabulary they have only ever used in their own language. Under strain, people compress. You get "we use spreadsheets" instead of the four-minute answer that would have told you which feature closes this deal.
Multi-threading collapses. Often exactly one person on the buying side is comfortable in English, so they become the only channel. Your champion. Every sales methodology warns against single-threading a deal, and here language is doing it to you automatically. The economic buyer and the security reviewer never appear, because the meeting defaults to English and they'd rather not.
The fix isn't a better script. It's removing the requirement that they perform in English — which is what live interpretation tools like Newey are for.
How do you run the call in their language?
For a call where your AE speaks English and the buyer's team speaks Japanese or German or Portuguese:
- Say it in the calendar invite. One line: we'll use live interpretation so everyone can speak their own language. Highest return on this list, and it costs nothing. It tells the buyer they can bring the colleague who doesn't speak English — exactly the multi-threading you couldn't get before.
- Open the split view. Meeting interpretation mode splits your screen and runs both directions at once. It reads the other side from a shared browser tab, so use desktop Chrome or Edge and join the meeting in a tab rather than the app.
- Screen-share it. The interpretation isn't private to you. Share the split view and the buyer's side reads their half live, with nothing to install and no bot joining. Everyone reads the same text, so whoever notices a drifting term can fix it.
- Load a glossary first. Your product name and category acronyms are what generic translation gets wrong, and they're what you say most often. A glossary pins them to your exact wording. Build it once, reuse it.
- Let the write-up happen. Save the session and AI notes produces a meeting document — who spoke to whom, decisions, numbers, follow-ups — in your language regardless of what was spoken.
That last step matters more than it looks. If a rep covered a language nobody else in the company spoke, you had no way to inspect the work. Deal reviews ran on that one person's summary, and recording the call didn't help because nobody could listen to it. Now the transcript and the write-up are readable by anyone on your team. An unauditable channel becomes a normal one.
What does this change about staffing a market?
The usual model is one AE per language region, hired locally, ramping for months. The Bridge Group's 2026 research across 158 B2B companies puts average ramp at 6.2 months — the longest they've recorded.
That's an expensive way to test a hypothesis. And the specific thing you lose is your strongest asset: the AE who has run four hundred discovery calls on your product and can hear a bad-fit deal in ninety seconds. That skill isn't in your enablement deck. It's judgement, and it doesn't transfer to a new hire in a new market.
Running the calls with your existing team inverts the order. Validate the market first, with your best people on the calls. Then hire against evidence.
When is a local AE still the right answer?
Live interpretation solves the conversation. It does not solve the market. Hire when:
- Pipeline has to come from local networks. If leads come from a regional conference or word of mouth in a language you don't operate in, no translation layer generates them. The most common reason to hire, and the most underrated.
- The sale needs someone in the room. On-site pilots, in-person security reviews, procurement meetings held face to face.
- Procurement is documentation-heavy and regulated. Public sector, healthcare, financial services. Binding artefacts are written and reviewed locally.
- Support has to run in-language. Selling in a language you can't support in is a churn problem, just deferred.
- You already know the market works. Once there's real revenue, hiring stops being a bet.
The clean line: if language is the only thing between you and the conversation, don't hire yet. If the market is what's between you, hire. We worked through that trade-off in sell into a market before you hire.
FAQ
Does this work with Zoom, Google Meet, and Teams?
Yes, and the same way in all of them, because it doesn't integrate with any of them. Newey runs in desktop Chrome or Edge, captures the meeting tab's audio, and you screen-share the interpretation view back into the call. No bot to admit, no admin to enable it. One requirement: join the meeting in a browser tab rather than the desktop app, since the tab is where the capture reads from.
Won't the interpretation lag make the call awkward?
There's about a second of delay, and people adapt within a few minutes. The bigger adjustment is speaking in complete sentences and pausing at the end of a thought. That's good sales practice anyway.
What about product names and jargon?
That's what the glossary is for, and it matters more in SaaS than almost anywhere, because your vocabulary is invented. Put your product name, feature names, and category acronyms in a set before the first call.
Can our sales manager review calls they can't understand?
Yes, and that's the practical unlock. Save the session and generate notes, and both the transcript and the write-up land in your manager's language. Deal reviews stop depending on one person's account.
Is meeting audio stored on your servers?
Live captured audio goes from your browser straight to the speech engine and does not pass through our servers. Session records are saved to your account only if you choose to save them. Details are in our privacy policy.
Should we still localize the product and the website?
Yes, and they're separate problems. Interpretation covers the live conversation. Your site, docs, and onboarding emails are read asynchronously and want proper localization.
Start with the calls you're already losing
You probably don't need a new market to test this. You have deals in your pipeline right now where an English-speaking champion is the only person you've ever spoken to, and the rest of the buying committee has never appeared.
Invite them. Tell them they can speak their own language.
Newey runs live translated captions in your browser and two-way meeting interpretation in desktop Chrome or Edge, in 60 languages, in whatever app your call happens to be in.
