Addressing formats and challenges in the US market

As part of our global addressing blog series, we’re taking a detailed look into the top 10 e-commerce markets. We’ll be highlighting the unique addressing formats in each country, while revealing the nuances and idiosyncrasies that can bewilder even the most experienced data experts. First up – it’s the United States. 

A typical US address format 

number[ ]Thoroughfare{[ ]Sub-building information}
CITY[ ]STATE CODE[  ]postal code

For example:

2978 West Main Street # 12
MINNEAPOLIS MN  23876

Most streets in the USA are built in a grid system, and this is reflected in their naming. Many streets are numbered and many of them include directionals (north, east, south, west) in their names. 

Sub-building numbers are also often part of the address, with the type (suite, apartment, floor, etc.) often substituted with a hash symbol (#) to distinguish it from the building number. 

In some parts of the USA, street names and street types are in Spanish. In rural areas – where streets may not be named or where buildings are along highways – rural route addressing may be used in this way: 

RR 5 BOX 12 A

The complexities of addresses in the US

The USA has some good examples of streets sharing multiple names. This is common in most countries when a highway hits a populated area and gets a local name. But it happens in other cases too. Famously, Sixth Avenue in New York is also Avenue of the Americas. 

Interestingly, the postal code (called ZIP code) can commence with a zero. This can result in many organisations finding they have lost these initial digits if they store the postal code data in a numeric field. 

There are also examples of building numbers with a leading zero in the USA. In Portland, Oregon, 016 and 16 are two different buildings. In some cases, a differentiation is made between buildings with numbers written in digits, and those with numbers written in letters. For example, 10 Post Office Square and Ten Post Office Square could be different buildings. 

Addressing and numbering is the responsibility of counties (or their equivalent) in the USA. And there are several examples of variations from the so-called standard format described above. In Wisconsin, many addresses are based on direction and distance from a fixed point in this way: 

W156-N8480 Pilgrim Road

Though locals seem to be able to work these out, they can cause problems for visitors and delivery companies. Counties in Utah have opted for a similar system, with multiple fixed start points within each country, making the locations of many addresses even harder to interpret. 

In Queens, a borough of New York, streets are numbered rather than named, with different street types running in different directions. So, if you were on a street, you’d know it runs approximately north/south. But if you were on an avenue, it would be running east/west. Addresses are based on the closest cross street and then a building number. 

For example, 31-35 55th Street, indicates that the nearest cross street to this address is 31st Avenue, and that the building number is 35 on 55th Street. Confused? This jingle might help: 

In Queens, to find locations best
Avenues, roads and drives run west;
But ways to north and south ‘tis plain
Are street or place or even lane.

Wondering why one borough of New York has a different addressing system from the rest? It’s because the boroughs of New York are also counties (so New York City contains counties rather than being in one), and counties define addressing systems. 

How Loqate solves these challenges 

Let’s take a look at some of the ways our technology works to solve the complexities of US addresses: 

• We utilise lexicons to handle frequently used abbreviations for directional names in streets, as well as abbreviations for sub building fields. This ensures that we can correctly parse the input and match the words to our reference data, whether they are in abbreviated or full form. Our reference data also includes common abbreviations used in the USA, such as ‘N’ for North and ‘S’ for South.
• To cater for misspellings, aliases, and misspellings within aliases, we try to understand the intent of the input. Commination of the lexicon helps us to parse the input. But if there's no exact match, we do a pattern match. Here we’re using logic to find the closest likely match to what the user wants.
  
  
• Thanks to our providers, we receive data in various language options. Capable of recognising input in multiple languages, our engine then returns relevant addresses based on the input received. This capability also helps us to identify, replace and position address elements into the correct structure, so the customer is presented with a formatted, validated, accurate address.
  
  
• Fuzzy Logic is a crucial component in our efforts to suggest the closest and best match for addresses. We continuously review and update these rules based on feedback received from our customers. This allows us to improve the accuracy and relevance of our address suggestions.
  
  
• Zip codes hold significant importance in an address. They serve as a vital element, and we pay special attention to both the standard Zip code and the Zip+4 format. These additional four digits in the Zip+4 code provide a unique identifier for more precise location information. If a valid Zip+4 code is available, it confirms that the address is deliverable. 

We also have the capability to make compensations for typographical errors in addresses. This means that even if there are minor mistakes or typos in the input, our system can still make reasonable adjustments to ensure accurate address matching.

At our core, we strive to leverage Fuzzy Logic, prioritise Zip codes, and accommodate for common errors to enhance the accuracy and reliability of our address suggestions.

Why you need an addressing expert

This might all sound rather complicated. And it is. When you’re looking for a standard pattern, you’re bound to find a myriad of exceptions. Every country is unique in its addressing formats and each location has nuances that can surprise even the most experienced data experts.

Thankfully, we solve this problem by combining the richest globally curated data from multiple postal, geospatial and local sources with a sophisticated matching and verification engine, ensuring the most accurate address data is captured and returned. The result is verified address data, standardised, enriched, and structured to the most appropriate local format.

By doing this, we’re able to give our customers the unrivalled precision and reliability they need to make location data-driven decisions, deliver superior customer experiences, help prevent fraud and enable cross-border commerce. We do it globally and at scale.

We’re the global addressing data experts, so you don’t have to be.