Mods - if this isn't allowed, please remove. Disclosure up front: I built the tool (Sextaris) that produced this data. Posting because the methodology is reusable and the contrarian finding might save someone's geo-targeting budget.

Methodology

I joined the 2024 ACS 5-year Census income tables (table B19001 for the full income distribution, B19013 for medians) to the Google Ads postal-target geo-target table. 31,215 postal targets matched cleanly. 29,187 of them have a median household income figure attached.

Important caveat that determines the entire analysis: the ACS topcodes median household income is $250,000. Any ZIP whose true median is $260K, $400K, or $1M shows up as $250,001. So sorting "richest ZIPs" by raw median is statistically meaningless above the ceiling. There are 102 such top-coded ZIPs in the data.

What surprised me

1. 29.57M US households earn $150K+. Only 208,663 of them live in those 102 top-coded "$250K+" ZIPs. That's 0.7% of America's affluent households within the entire prestige tier.

2. The $100K-$150K median-income band holds 11.03M of those $150K+ households across 4,511 ZIPs. About 53x the buying-power scale of the prestige tier.

3. Hidden-scale leader: TX ZIP 77494 (Katy, west of Houston) has 21,294 households earning $150K+. That single ZIP holds more than 10% of the entire $250K+ ZIP combined population.

4. Precision-pocket leaders by $150K+ household share: 02071 MA at 91%, 10577 NY at 87%, 10282 NY at 87%, 07078 NJ at 83%. Tiny lists, very high concentration.

Enrichment layer

To make the data planning-grade rather than just "richest list", I added: IRS SOI tax composition (Schedule C share, asset income, wage share, AGI per return), Zillow ZHVI top-tier home values, Census Business Patterns establishment density NAICS-coded (investment advisors, private schools, country clubs, dentists, etc), ACS housing tenure and education. 1.84M normalized metric rows across 37,813 ZIPs, 28 signals total.

Then I scored every $250K+ ZIP against 7 archetype frames using percentile-based signals:

1. Asset Wealth: lives off dividends, interest, capital gains.
   
2. Executive Salary: wage-driven affluence, high bachelor+ density.
   
3. Luxury Housing: top-tier ZHVI + interior-design / remodeler density.
   
4. Wealth Services: local cluster of advisors + asset-heavy income.
   
5. Private School: private school + bachelor+ density.
   
6. Health / Lifestyle: concierge MD + country club + fitness density.
   
7. Business Owner: Schedule C share + total business establishment density.

Same income band, very different intent.

CPC overlay (proof-of-concept)

For each archetype, I pulled Google Ads keyword expansions per ZIP for archetype-relevant seed packs, then hydrated with national historical metrics.

Currently populated for 2 of 7 archetypes:

Private-School archetype: weighted CPC $1.99 across 60 keywords and 884K monthly searches. Top volume: "private schools near me" at $2.09 / 110K monthly. Cheapest affluent auction in the data.

Health/Lifestyle archetype: weighted CPC $4.24 across 20 keywords and 121K monthly searches. Top volume: "concierge medicine" at $4.68 / 40,500 monthly. Highest single CPC: "concierge md la" $5.72. Highest single CPC: "concierge md la" $5.72.

Caveats, Reddit will rightly ask about

ZCTA != USPS ZIP boundary. ZCTAs are Census approximations of ZIP service areas. A small share of postal codes won't match cleanly. Very rural ZCTAs sometimes cover multiple service areas.

ACS 5-year is a smoothed window. The 2024 release pools responses from 2020 to 2024. It lags the latest tax-year reality and damps year-over-year shifts.

IRS SOI masks ZIP cells with very low return counts. Some affluent but tiny ZIPs show no AGI breakdown. I leave them null rather than impute.

Archetype scores are percentile ranks within the $250K+ slice, not absolute thresholds. A ZIP scoring 99 on Asset Wealth means "asset-heaviest in the affluent slice", not "an asset-wealth ZIP by some industry standard".

The median-income heuristic is what the analysis pushes back against. If you've been using something else (HH count by income band, education, NAICS density), you're already past the issue. The atlas just makes it queryable.

Practical takeaway

Targeting "richest ZIPs" by median income means competing for less than 1% of America's $150K+ households. The hidden-scale and precision-pocket lists open a much wider surface, and the archetype frames tell you what to actually say to the people who live there.

Atlas + downloads

Live filters across all 31,215 postal targets, the 7 archetype descriptions, and Google Ads ZIP exports in three formats (postal-paste TXT, Editor CSV, full data CSV with all enrichment columns): https://www.sextaris.com/blog/richest-us-zip-codes-google-ads-income-targeting

If you're running geo-targeted campaigns and have used a heuristic better than median income for affluent intent, I would love to hear about it. The 7-archetype frame is one way to slice it, but I'm sure people on this sub have approaches I haven't tried.