Market entry strategy: the five routes into a new market and how to pick one

Expanding into a new market looks like a demand question and is almost always an operations question. Companies rarely fail abroad because nobody wanted the product. They fail because invoices could not be issued in the local currency, contracts arrived in the wrong language, support answered eight hours late, and a local competitor with a worse product had a salesperson in the room.

The route you take into a market determines which of those problems you inherit and which you hand to someone else.

The five routes

Direct entry

You sell into the market from where you already are. No local entity, no local staff, often no local language beyond a translated website.

It is the cheapest test and the right first step for most software companies. The limits appear quickly: procurement teams that will not sign with a foreign entity, data residency requirements, tax registration thresholds, and buyers who want someone in their timezone.

Partner or distributor

A local company sells your product, usually for a margin or a reseller discount. You get local credibility, an existing customer base and someone who understands how business is actually conducted there, without hiring.

What you give up is the customer relationship and most of the data. You will not know why deals are lost. Partners also prioritise whatever is easiest to sell that quarter, so a partnership with no enablement, no deal registration and thin margin produces a logo on a slide rather than revenue.

Joint venture

A shared entity with a local firm, with shared investment and shared governance. Common where regulation requires local ownership, and in markets where a foreign brand alone carries little weight.

The cost is decision speed. Every significant choice now needs two boards to agree. Worth it when the alternative is not being allowed to operate at all.

Acquisition

Buying a local company gives you customers, staff, local knowledge and revenue on day one. It is also the most expensive route and the one where most of the value is destroyed after signing, through integration failures rather than bad valuation.

Greenfield subsidiary

Your own entity, your own hires, your own brand. Full control, full cost, slowest path. Appropriate when the market is large enough to justify a permanent presence and when controlling the customer experience is part of the product.

Choosing between them

Three questions usually settle it.

How much does this sale depend on local knowledge? If buyers award contracts based on relationships built over years, regulation is dense, or the purchase requires navigating a public procurement process, a partner will outperform anything you can build in two years. If the product is bought after a self-serve trial, local knowledge matters far less.

How much can you spend before the first revenue? A subsidiary is eighteen to thirty months of cost before contribution. If that would strain the core business, the decision is already made.

How much do you need the customer relationship? Product-led companies that depend on usage data and in-product expansion lose something real when a partner owns the account. Companies selling a discrete implementation lose much less.

A common sequence that works: enter directly to test demand, sign a partner to accelerate once demand is proven, and build a local entity only when partner-driven revenue justifies it. Each step is reversible, which the reverse order is not.

The operational checklist people skip

Demand is usually the easy part. These are the items that quietly stall entries:

• Billing. Local currency, local payment methods, and an invoice format the buyer's finance team will accept.
• Tax and entity. VAT or GST registration thresholds, withholding tax on cross-border payments, permanent establishment risk once you have staff on the ground.
• Contracts. In the local language where expected, under a governing law the buyer will accept. A US-law contract is a real obstacle in parts of Europe and Asia.
• Data. Residency requirements, and a data processing agreement that satisfies the local regime.
• Support hours. Same-day response in the buyer's timezone, which usually means either a local hire or a deliberate follow-the-sun rota.
• Language. Not just the website. Onboarding, error messages, documentation and the first support reply.

Each of these can kill a deal at the final stage, after all the sales work is done, which is the most expensive place to discover them.

Localising the offer itself

Pricing rarely transfers unchanged. Willingness to pay differs, competitive baselines differ, and a price that reads as premium in one market reads as suspicious in another. Check the local competitive set before copying a price list across.

Positioning shifts too. The differentiator that wins in a mature market, where buyers already understand the category, is rarely the one that wins in a market where you first have to explain why the category exists.

Decide the exit condition first

Write down, before entering, what would make you stop. A number, a date, and a name.

"If we have not closed eight customers in this market within four quarters at a CAC within 50 percent of our home market, we withdraw and revisit in two years."

Without that written down in advance, an underperforming market survives on hope and sunk cost, absorbing budget and attention for years, because withdrawing becomes a personal defeat rather than a planned decision. The exit condition is what turns it back into a decision.