On a conference room wall in an office park outside Dallas, a printed map has 18 postal areas outlined in yellow highlighter and the words TOP TARGETS written across the top in marker. Nobody in the room can say which number produced the yellow. The list came from a vendor file 2 years ago, the person who ordered it has left, and a rep has been driving to those 18 areas every week since. Value can mean household income, business count, current revenue, average order size, or win rate, and those 5 measures rank the same city in 5 different orders.
The Postal Origin of ZIP Codes
ZIP codes were built to sort mail. They are collections of delivery routes assigned for postal convenience, which is why they overlap, why some cover a single building, and why a few describe no fixed area at all. They were never designed to describe a neighborhood, a market, or a customer.
That does not make them useless. They are the unit most business data arrives in, most customers can recite from memory, and most vendors will sell demographics against. The working approach is to treat the ZIP code as a container for data rather than as a description of a place, and to check the container before trusting what is inside it.
Choosing the Measure of Value
For a business selling to consumers, value usually tracks household income, home value, and household count. For a business selling to other businesses, income is nearly irrelevant, and the measure is the count of establishments in the target industry and size band. A wealthy residential ZIP code can hold almost no target companies at all.
Mixing the 2 measures is the common failure. A software company selling to dental practices does not care what the households in a postal area earn. It cares how many practices with 4 or more operatories are inside it, and those 2 numbers can point at opposite ends of the same county.
A third measure sits between them and is usually available for free: the company’s own history. Existing revenue per postal area is the only input that already reflects pricing, competition, product fit, and the sales team’s ability to close, and it should anchor the ranking that the external data then adjusts.
Ranking Before Drawing
The ranking step is arithmetic, and it should happen before anyone draws a line. Pull every ZIP code in the target region, attach the chosen measure to each, and sort. Then attach 2 more columns: current revenue from that ZIP code and current account count. The result usually splits into 4 groups, and each implies a different territory decision.
High value with high current revenue is a defend zone. High value with low current revenue is the growth target and the reason for the exercise. Low value with high revenue means the measure of value is wrong for this business and should be reconsidered. Low value with low revenue is deprioritized without guilt.
Turning a Ranking Into a Territory
A list of 20 ZIP codes becomes a territory only when the codes are contiguous, reachable, and sized to a week’s work. Grouped in a zip code mapper, the top ranked codes separate into adjacent clusters and strays 90 minutes from everything else.
Strays are the recurring problem. A high scoring ZIP code 2 hours from the cluster costs a full day per visit, and a rep assigned both will quietly stop making that drive. The choice is to attach it to a different territory, cover it by phone, or accept a lower ranked but adjacent code in its place. That choice is easy to make on a map and nearly impossible to make from a sorted list.
Concentration Is the Normal Case
Planning tends to assume value is spread evenly across a region, and it never is. A study of London’s office economy documents the impact of agglomeration on where output happens, finding that larger places are more productive and that a small part of the city accounts for a large share of its economic output.
Residential wealth skews harder still, and at the top the numbers stop resembling anything a territory model would predict. A 2025 ranking of the postal areas with the most expensive homes in the country put Miami Beach’s Fisher Island first at a $9.5 million median sale price, a 65% jump in a year, ahead of Atherton, California at $8.33 million after 8 straight years in first place.
For territory design that means the top 10 postal areas in a metro region often contain more target demand than the next 60 combined. A territory drawn to look balanced by area will be badly unbalanced by opportunity, and the rep assigned the geographically larger half will be the one who misses quota.
Direction of Travel in the Rankings
Rankings age, and the direction of travel matters as much as the current standing. Commercial real estate coverage in Greater Boston described retailers who flock to the suburbs, with suburban downtowns and open-air centers holding vacancy under 2% as residents who work from home shop and eat closer to where they live.
A territory plan built only on last year’s revenue map will chase the demand that used to exist. Adding a simple change column, revenue this year against revenue 2 years ago per ZIP code, separates the codes that are declining from the ones that are merely under-covered.
Cost Differences at the City Line
Adjacent ZIP codes often fall in different municipalities, and the difference shows up on the customer’s side of the transaction. Reporting in Philadelphia on the city’s very unfriendly business tax described a therapist whose Business Income and Receipts Tax bill rose above $5,700 in a single year, a cost that does not exist for the identical practice a few miles outside the city line.
For a rep, that translates into different objections, different pricing sensitivity, and sometimes different products on the same street. Marking municipal boundaries on the territory map keeps those differences visible instead of surprising.
Reviewing the Ranking Twice a Year
Twice a year, rebuild the ranking with fresh revenue data and check which postal areas moved into the top group, which fell out, and which of last cycle’s growth targets actually produced accounts.
That last question tends to embarrass the method, and it should. A high value area targeted for a year that produced nothing is either mis-measured, already owned by a competitor, or a place the product does not fit. Which points at something uncomfortable about the whole exercise. The ranking is a bet the company placed on one measure of value, and bets get checked. A company that treats its top 20 list as settled fact will defend it for years. A company that rechecks it every 6 months will be wrong for 6 months at a time.
