
Multi-Tenant Architecture: 3 Models & 8 Best Practices
Learn how multi-tenant architecture works, compare database isolation models, and explore six components & eight best practices for shared infrastructure.
Price localization explained: cosmetic vs. market-based pricing, real examples from Spotify and Microsoft, how to set prices by country, and the stack it needs.
A $49 credit pack is an easy yes for a team in Zurich and a hard sell for a team in São Paulo, even though both teams get the same product. When you charge everyone the same price, you're either leaving money on the table in one market or pricing yourself out of the other.
Price localization is how companies deal with that, and it's also where pricing strategy runs straight into engineering, since every regional price has to exist in the catalog, the entitlement layer, billing, and checkout at the same time.
Stigg ships price localization as part of its product catalog, so we see the decisions teams make about regional prices, and what breaks when those prices don't line up across systems.
By the end, you’ll know how the two types of price localization differ, how Spotify, Microsoft, and OpenAI handle regional prices, and what your catalog, entitlements, and billing need to keep those prices in sync.
Price localization is a pricing strategy where the price a customer pays depends on where they are, adjusted for local purchasing power, competition, and payment habits, and usually shown in the local currency.
It's a different thing from localizing your product's language. Translating your app and website is content localization, while price localization changes the number on the pricing page and what the customer gets for it.
Three factors usually drive a localized price:
Converting your USD price into euros at today's exchange rate is the starting point, but on its own it only changes how the price is displayed. The price itself still reflects what customers in your home market will pay.
The two types of price localization are cosmetic localization, which converts one global price into local currency, and market-based localization, which sets a different price for each market.
| Dimension | Cosmetic | Market-based |
|---|---|---|
| What changes | The currency shown | The price itself |
| Example | $99 shown as about £79 | $99 in the US, $49 in India |
| Pricing strategy | One global price | A price per market |
| Engineering effort | Low | Medium to high |
| Catalog model | One global plan | A plan variant per region |
| Works best for | Markets with similar purchasing power | Markets with big economic gaps |
Cosmetic localization changes how a price is displayed while the underlying price stays the same, so a $99 monthly plan shows up as roughly £79 for a UK visitor based on the current exchange rate.
It's the simplest version and usually the first step for companies expanding abroad, since seeing a familiar currency at checkout removes friction.
To implement, you need to detect the visitor's location, show prices in the matching currency, and let the billing provider handle conversion, which most modern billing systems do natively.
Where cosmetic localization falls short is in markets with very different purchasing power. A price converted from USD can still be far too high for customers in Southeast Asia or Latin America, even though it looks local.
Market-based localization sets the price for each region based on what customers there are willing to pay, using local purchasing power, competitor pricing, and price sensitivity.
That takes more infrastructure. You need a product catalog that supports regional prices, plan and entitlement rules that can vary by region, multi-currency billing and invoicing, and regional payment methods, and every one of those has to agree on which plan a customer is on.
For AI products, there's an extra wrinkle, because a credit package that's fairly priced in one market can be badly mispriced in another while the underlying token costs stay the same everywhere.
Spotify, Microsoft, and OpenAI show three different approaches to price localization, from full market-based pricing with regional packaging to USD pricing with local currency billing in some countries.
Spotify Premium is $12.99 a month for Premium Individual in the US, while the UK page lists £12.99 and the Brazil page lists R$23.90, which is about $4.75 in October 2026.
India goes further. The India page leads with Premium Standard at ₹139 a month (or about $1.50), with audio up to about 320kbps. Lossless sits in a separate ₹299 Platinum plan, while the US Premium Individual plan includes it.
That's market-based pricing plus localized packaging, since the plan itself changes along with the price.
Microsoft 365 Personal costs $9.99 a month in the US and ₹689 a month in India (roughly $7.30), each shown in local currency on the country's own buy page. The plan is the same, and the price point is set for each market.
OpenAI takes a lighter approach. According to its help center, ChatGPT Go is available in every country ChatGPT supports, all purchases are in USD, and local currency billing is offered in a limited set of countries.
That's closer to cosmetic localization, with a lower-priced plan doing some of the work a regional price would otherwise do.
Disclaimer: Prices are subject to change without notice. Always visit the official company websites for the most up-to-date pricing information.
The main benefits of price localization are higher conversion in target markets, better positioning against local competitors, access to local payment methods, cleaner multi-currency reporting, and a lower-risk place to test pricing.
Teams that localize usually have international traction, a broad market, signs of a price mismatch, and the capacity to run regional prices. Four questions help decide:
Teams that answer yes to the first three and no to the fourth often start with cosmetic localization while the infrastructure catches up.
To set localized prices, price your biggest market first, benchmark other countries against a purchasing power index, factor in local costs and value, and then test and adjust.
Your largest market gives you the most data and the most revenue at stake, so most teams settle that price before deriving the others from it. Every regional price ends up as a ratio of this one, which means a mispriced home market spreads the error everywhere.
Purchasing power parity data, such as the World Bank's PPP conversion factors, shows how far a unit of currency goes in each country.
Many teams build their own index around their buyer instead, like local developer salaries for a developer tool, because it tracks willingness to pay for that specific product more closely.
The idea is to keep the price-to-index ratio roughly the same across markets, so the product costs a similar share of what your buyer earns or spends everywhere.
Some costs don't shrink with purchasing power. Fixed per-transaction payment fees and, for AI products, model and compute costs don’t shrink with the price, so a price set purely from a PPP index can slip below your cost to serve.
Value can differ by market too. Local competitors, content or feature availability, and how much a capability is worth in that market can all justify a higher or lower price than the index suggests.
Roll the new price out to one or a few countries, watch conversion, churn, and revenue per customer, and then adjust. Value-based pricing applies here as much as anywhere, since the test tells you what customers in that market think the product is worth.
Testing well depends on your system. Regional prices need to live in configuration with versioned plans, so a test can go live in Brazil without touching existing customers in the US, which is the job of plan versioning.
Price localization is shaped by purchasing power, competition, payment methods, currency stability, and tax rules, and each one turns into requirements for product, billing, and checkout.
Purchasing power is the main driver of willingness to pay. A $50 monthly plan might be a rounding error for a team in one country and a serious line item in another, and when pricing ignores that, adoption drops even when the product is good.
Local competitors often set the ceiling. If a strong regional player charges much less, pricing far above it limits adoption regardless of quality, so teams need to adjust regional prices without a code deploy every time the market moves.
In many markets, a buyer who’s fine with the price still can’t finish checkout if their usual payment method isn’t there:
In markets with high inflation or volatile currencies, some companies keep pricing in USD for stability. Your billing system needs to support both local-currency and USD pricing by region.
VAT in the European Union, GST in Australia, and local invoicing rules elsewhere all affect the final price and how it's shown, so billing and invoicing have to handle region-specific tax and compliance.
Early-stage companies usually start with cosmetic localization, growth-stage companies add market-based prices, and late-stage companies run regional pricing as an ongoing operation.
Early-stage teams usually begin with local currency display in a few high-traction markets. The US, UK, Canada, and the EU usually cover most early international demand.
The goal is less checkout friction, with geo-detection for currency, common regional payment methods, and conversion tracking by region.
Full market-based localization this early tends to create overhead you can't justify while product-market fit is still moving.
Once international revenue is meaningful, localization shifts from currency display to pricing strategy, with price sensitivity research by region, competitor monitoring, and regular price adjustments.
Iteration speed is what decides whether this works. According to SaaStr’s analysis of Paddle’s dataset of $36 billion in ARR, companies that update pricing every three months see a 103% jump in ARPU over three years, and regional pricing experiments that need an engineering cycle rarely get run that often.
At scale, localization becomes an operating function: regional price floors and ceilings, adjustments as currencies move, and region-specific taxes and payment methods across many markets at once.
Enterprise customers add their own requirements here, like per-team allocations and per-workspace limits that have to resolve against the correct regional plan on every request, along with an audit trail finance can use to reconcile revenue across markets.
Price localization usually breaks in the infrastructure. Showing a price in euros is straightforward, and the hard part is keeping one regional price consistent across every system that touches it.
A single regional pricing change touches several systems at once:
Someone also has to decide which country a customer belongs to.
Stigg's price localization docs leave that to your app, through GeoIP, the billing address, or a customer's chosen currency, and recommend applying the same logic on the public pricing page and in-app so prices don't change unexpectedly once a billing address is entered.
Each customer also has only one billing currency, so once they’re on a paid plan, every subscription they hold bills in that currency.
Without a single source of truth, each system fails in its own way:
A centralized catalog fixes this by making the plan definition the source that billing, entitlements, the pricing page, and checkout all read from.
Teams that localize efficiently keep regional prices, plan limits, feature access, and currency settings in one catalog, so a regional price change becomes a configuration edit. The pricing and packaging guide covers how to structure that catalog in the first place.
Enforcement speed still applies, since regional limits get checked on every request.
Stigg's Sidecar answers entitlement checks from an in-memory cache by default (Redis-backed persistence is optional), fetches from the Edge API at around 100 ms on a miss, and returns your configured defaults if that call times out.
The tools that support price localization are billing platforms, Merchant of Record services, product catalog and entitlement systems, and multi-currency accounting tools.
Billing platforms such as Stripe and Zuora handle multi-currency billing, invoicing, tax calculation, and payment processing. Stigg also partners with Airwallex, which handles multi-currency processing and collection for invoices Stigg generates.
Stripe's Adaptive Pricing covers the cosmetic side for Checkout and Payment Links, switched on from the Dashboard.
Customers pay in their local currency in more than 150 countries, Stripe picks the presentment currency and calculates the localized price, and the exchange rate is guaranteed for 24 hours.
A Merchant of Record becomes the legal seller, taking on VAT and GST collection, tax filings, and cross-border compliance, so you don't have to register for tax in every jurisdiction. Paddle is a well-known example.
This layer decides which plans exist, what each includes, which limits apply in which region, and which price applies to which market. Stigg added price localization to its catalog so teams can set different price points, local currencies, and even different packaging per country.
Accounting tools handle revenue recognition across currencies, exchange-rate adjustments, and MRR reporting, so revenue metrics don't swing with exchange rates when no prices changed.
Most of the work in regional pricing is getting each price into the catalog, entitlements, billing, the pricing page, and checkout at the same time, then changing it again next quarter.
Stigg is the usage runtime for AI products. It sits between your product and your billing stack and works alongside the billing you already run.
With Stigg, teams can:
If you're planning your first regional prices and want to talk through how they'd map onto your stack, talk to a human at Stigg, or start with the AI quickstart in the docs.
Price localization is a pricing strategy where the price a customer pays depends on their location, adjusted for local purchasing power, competition, and payment habits. It usually includes showing prices in the local currency.
Spotify is a well-known example of location-based pricing. Premium Individual is $12.99 a month in the US, while the India page leads with a Premium Standard plan at ₹139 a month.
Cosmetic localization converts one global price into local currency using exchange rates. Market-based localization sets a different price for each region based on purchasing power and competition.
Price localization costs depend on how pricing is modeled in your product. With a centralized catalog, a regional price is a configuration change, while scattered pricing logic means updates across billing, entitlements, and checkout. Translation is a separate content localization cost.
Price localization affects entitlements when plans or feature limits differ by region. The entitlement system has to resolve the correct regional plan and its limits for each customer, on every check.