The Accounts You Cannot Open in Your New Market

What nobody budgets for is the fortnight after, when the company legally exists and still cannot operate. The delivery platform will not finish onboarding. The local marketplace rejects the seller application. The ad account sits unverified. Each one stops at the same screen, asking for a phone number in the country you have just incorporated in, and rejecting the one you have.

Why the number is the thing that blocks you

When you enter a number, the platform runs a lookup that returns the country that issued it, the carrier, and the line type — mobile, landline, or internet-based. That result is judged before any message is sent. A number from the wrong country fails on the country field. An internet calling number fails on the line type, because those are cheap to create in bulk and therefore useless as evidence of anything.

What passes cleanly is a real carrier line issued in the market you are entering. Teams with staff on the ground buy a local prepaid SIM in the first week; teams without them use services that receive SMS on a local number to clear the same gate. The requirement is identical either way, and it is not negotiable by argument — the check is automated and there is nobody to appeal to.

The founder's usual first move is to use their existing number and hope the platform is relaxed about it. Sometimes it is. More often the account opens in a degraded state: limited features, held payouts, or a verification request that surfaces three weeks later at the worst possible moment.

Incorporating and operating are different problems

Cross-border expansion advice tends to stop at the legal entity, as if the company becomes operational the moment the register updates. In practice a company is a bundle of accounts — payments, logistics, advertising, marketplaces, local SaaS, sometimes a utility or a mobile contract — and each of those has its own idea of what counts as a legitimate local business.

Most of them settle on the phone number as the cheap proxy. It is not a good proxy. It is simply the one signal that is hard to fake at scale and takes two seconds to check, which is why it has quietly become the gate that decides whether your second market opens this month or next quarter. Nobody designed it as the deciding factor. It ended up there because it was the easiest check to automate.

What actually asks

Marketplaces and delivery platforms. Seller and partner onboarding almost always requires a local contact number, and the number often has to match the country of the business address you gave. Mismatches route the application into manual review, which is a queue measured in weeks.

Advertising and business accounts. Platforms tighten verification for accounts that spend money, especially new ones operating in a country the account has no history in. A number from a third country is one of the strongest signals that pushes an account into extra scrutiny.

Local payment and invoicing tools. Regional providers are the ones most likely to hard-require a domestic number, because their user base is domestic and they have no reason to build for anything else.

Anything with a courier or a technician. Logistics, equipment rental, office services. These are unglamorous and they will stall an office opening for a fortnight if the delivery company cannot send a text to a number that answers in the right country.

The workarounds founders try

Three come up repeatedly, and all three have the same failure mode.

The first is to route everything through a local contractor or an agency, using their number. It works until the relationship ends and you discover that account recovery, and sometimes the account itself, belongs to someone who no longer works with you. Treat any account opened on someone else's number as a loan, not an asset.

The second is an internet calling number, which is the option that looks cleanest and fails most reliably. The lookup identifies the line type and the code is never sent. The screen still says it was, which is why founders spend an afternoon convinced they have a bug rather than a policy.

The third is a free public code-display page. Those numbers are shared, already used on the platform you are trying to join, and readable by anyone who opens the same page. For a company account carrying payment credentials this is not a shortcut, it is an incident waiting for a date.

What a local number does not get you

This is the part worth being blunt about, because the opposite claim is common and wrong.

Passing a phone check clears one gate. Everything downstream that establishes who you are stays exactly where it was. Company registration documents, a director's identity verification, proof of address, VAT registration, beneficial ownership declarations, bank account opening under anti-money-laundering rules — none of that moves because a text message arrived. Regulated finance in particular is designed to be hard here, and it should be.

Nor does a number from a market entitle you to operate in it. If a platform's terms require a registered local entity, holding a local number while having no entity is a terms violation, and the enforcement, when it comes, tends to take the account and everything attached to it. The number solves an operational annoyance for a company that is genuinely entering a market. It does not manufacture the entitlement to be there.

Sequence it before the entity, not after

The fix is scheduling rather than technology.

Write down every account the new market needs before you file anything — payments, logistics, marketplaces, advertising, local tools. It is usually somewhere between fifteen and thirty, and the list itself is the useful artefact, because half of them will turn out to have a country requirement nobody mentioned.

Decide who owns the number. It should belong to the company, sit with a role rather than a person, and be recorded next to the accounts that depend on it. The alternative is discovering during a funding round that a critical account recovers to a phone that left with a former employee.

Then get the number working before the entity is finished, not after. The registration timeline is fixed and known. The account onboarding timeline is the one that surprises people, and it only starts once you can receive a code.

The short version

The legal part of entering a new market is the part with a documented process and a professional who does it for a living. The part that actually delays launches is a series of automated checks that ask for a local phone number and do not explain themselves when the answer is wrong.

It is a small dependency with a disproportionate ability to hold up a quarter. Handle it in the same week you brief the lawyer, and it never becomes a story worth telling.