How to Build Accurate Local Contractor Cost Guides with BLS Labor Data

"Cost to replace roof [city]" is one of the highest-converting queries in the home services vertical. Homeowners searching this phrase are not browsing - they have a real problem and they are close to hiring someone. The issue is that most pages targeting this query use national averages and simply insert a city name into the title. That approach is detectable, and users know it when they see it.

Labor costs for identical home improvement work vary 40-60% between expensive metros like San Francisco and lower-cost markets like Memphis. A roofing contractor in the Bay Area earns roughly double what one earns in the mid-South - and that difference flows directly into the price a homeowner pays. BLS Occupational Employment and Wage Statistics (OEWS) data makes it possible to generate genuinely localized cost estimates at scale, with a defensible methodology you can cite and explain to readers.

This guide covers the exact data structure, API calls, multiplier calculations, and page templates needed to build cost guide pages that reflect real local labor markets.

Understanding BLS OEWS Data Structure

The Occupational Employment and Wage Statistics survey is conducted annually by the Bureau of Labor Statistics in partnership with state workforce agencies. It covers approximately 800 occupations across 580+ metropolitan and nonmetropolitan statistical areas. The survey collects wage data from about 1.1 million employer establishments each year, making it the most comprehensive occupational wage dataset available for the United States.

OEWS data is published each May, covering the reference period from the prior November survey. Each data record includes:

• Hourly mean wage (average)
• Hourly median wage (50th percentile)
• 10th percentile hourly wage
• 25th percentile hourly wage
• 75th percentile hourly wage
• 90th percentile hourly wage
• Employment estimate (number of workers in that occupation/area)

For home services content, the most useful occupations and their SOC codes are:

BLS Series ID Format

BLS series IDs for OEWS data follow this structure: OEU[area-code][occ-code] where the area code is a 7-digit BLS area code and the occupation code is a 6-digit SOC code (without the hyphen). For example, the series ID for roofers (47-2181) in the Denver-Aurora-Lakewood MSA (area code 0019820) would be OEUM0019820472181.

BLS area codes for metropolitan statistical areas are available at bls.gov/oes. The MSA codes generally correspond to Census CBSA codes, but there are exceptions - always cross-reference the BLS area code list directly rather than assuming CBSA code equivalence.

Fetching BLS Data via the API

The BLS Public Data API v2 accepts POST requests with a JSON body specifying series IDs and year range. Registration is optional but increases your daily limit from 25 to 500 queries per day.

POST https://api.bls.gov/publicAPI/v2/timeseries/data

{
  "seriesid": [
    "OEUM0019820472181",
    "OEUM0019820472111",
    "OEUM0019820472152"
  ],
  "startyear": "2022",
  "endyear": "2023",
  "registrationkey": "YOUR_KEY_HERE"
}

The API returns up to 50 series per call. Working out the math for a full collection run: 580 metros multiplied by 8 occupations equals 4,640 total series. Divided by 50 series per call, that requires 93 API calls. With a registered key and a 500-call daily limit, a full collection run takes just under one day. Plan for a 10-day collection window to allow for retries and partial failures.

Store each response with the publication vintage (May YYYY) alongside the wage values. This lets you display "Based on BLS OEWS May 2023 data" on the page, which adds credibility and helps readers understand the data currency. OEWS does not publish intra-year updates, so annual publication is the expected update cadence.

Handling Suppressed Values

BLS suppresses wage data when employment is too thin to publish reliably - often in smaller metros for specialized occupations. The API returns a value flag of W (withheld) in these cases. Your fallback strategy: use the state nonmetropolitan area data for the state containing the target MSA, then fall back to national data. Store the fallback tier so you can accurately disclose the data source on the page.

Calculating the Local Labor Multiplier

The local labor multiplier is the core calculation that converts a national average cost estimate into a locally-adjusted one. The formula is:

local_multiplier = local_median_wage / national_median_wage

Using roofers (SOC 47-2181) as an example with 2023 BLS OEWS data:

The multiplier only applies to the labor component of the total cost. Materials - shingles, underlayment, flashing, nails - are priced on national commodity markets and do not vary significantly by location. This is an important distinction to explain to readers: a contractor in Denver does not pay more for shingles than one in Memphis, but they do pay (and charge) more for the crew that installs them.

For each project type, identify the labor percentage of total cost. Apply the multiplier only to that fraction, then add back the non-adjusted materials component. This produces a more accurate localized estimate than simply multiplying the entire national average by the labor multiplier.

Cost Guide Page Template

The following structure works well for "cost to replace roof [city]" pages. Each section has a clear purpose that serves both users and search engines.

Quick Answer Box

Every cost guide page should open with a concise answer that can be featured as a rich result. The format:

"In [City], roof replacement typically costs $[min]-$[max], compared to the national average of $[national]. [City] roofing labor is [X]% [above/below] the national average based on BLS OEWS wage data."

This box should be marked up with ItemList or placed inside a FAQPage schema element to maximize rich result eligibility.

Recommended H2 Structure

• What Does Roof Replacement Cost in [City]? - cost range table by roof type (asphalt shingles, metal, tile, flat) with local labor multiplier applied to each. Include a note explaining the methodology.
• What Drives Roofing Costs in [City]? - local labor rate from BLS, permit cost from municipal data if available, materials note (national commodity pricing).
• Getting Roof Replacement Quotes in [City] - guidance on collecting 3 bids, what a written estimate should include, red flags like cash-only demands or pressure to skip permits.
• [City] vs National Average - comparison table showing cost by roof type for local vs national.
• Is It Worth Getting a Roof Permit in [City]? - brief explanation of permit requirements, links to local building department if available.

Monetization Slots

Embed affiliate and lead generation placeholders as semantic HTML elements that can be populated by a templating layer without modifying the page structure:

<div class="affiliate-slot" data-type="home-depot"></div>
<div class="leadgen-slot" data-type="angi"></div>

This pattern keeps monetization elements separate from content, making it possible to swap affiliate programs or lead gen partners without regenerating pages.

The 5 Core Home Improvement Cost Guides

Not all home improvement categories have the same labor intensity. Understanding the labor-to-materials ratio for each project type is essential for applying BLS wage data accurately. These five project types represent the highest search volume in the home services space:

Exterior painting deserves special attention: at 70% labor, it is the most labor-sensitive of the five project types. A painter in San Francisco earns roughly 2.2x the national median, which means a $4,000 national average job could realistically cost $6,200-$7,000 in the Bay Area. This kind of genuine localization is what separates useful cost guide pages from national average pages with a city name pasted in.

For kitchen remodels, calculate a weighted average multiplier across the three trades involved. A typical mid-range kitchen remodel splits labor roughly as: 50% carpentry/cabinetry, 30% electrical, 20% plumbing. Calculate each trade multiplier separately, apply the weights, and use the resulting blended multiplier against the 40% labor component of the total project cost.

Handling Cities Without MSA-Level BLS Data

BLS OEWS covers 580+ metropolitan statistical areas, but the United States has approximately 30,000 incorporated cities and towns. The vast majority of smaller cities and rural communities do not have direct MSA-level OEWS data. A workable fallback strategy has three tiers:

1. MSA-level data - use directly if the city falls within a covered MSA. The Census CBSA delineation file maps every county (and therefore every city) to its MSA. Cities in covered MSAs should use MSA-level wage data, even if the city itself is in a suburb or exurb of the named metro center.
2. State-level OEWS data - BLS publishes state-level OEWS separately from metro data. For cities in uncovered areas, use the statewide median wage for the relevant occupation. Statewide data is available for all 50 states and Washington D.C.
3. Census Region average - if state data is also suppressed for a given occupation, fall back to Census Region averages (Northeast, Midwest, South, West). BLS publishes regional data alongside state and metro data.

Store the fallback tier alongside each wage value in your database. When generating page content, surface this information to readers: "Labor cost estimate based on [State] statewide BLS data - this city does not have direct metropolitan-area wage data available." This transparency builds trust and protects you if a user disputes the estimate.

Avoid silently using national averages for small cities without disclosing the fallback. If your page claims to show "roofing costs in [small town]" but is using national average labor rates, that is no more accurate than the pages you are trying to outcompete. The fallback tier system gives you genuine localization for the majority of your city list while being honest about data limitations for the remainder.

Combining BLS Data with Other Sources for ROI Sections

Cost guide pages rank well. But pages with an ROI angle - "is this project worth it in [city]?" - earn backlinks and social shares that pure cost pages do not. Adding a return-on-investment section to each cost guide requires combining BLS labor cost data with two additional government sources.

The formula uses three inputs:

• FHFA HPI (House Price Index) - provides the market appreciation trend for the metro. Available from fhfa.gov as quarterly CSV downloads.
• Census ACS B25077 - median home value by ZIP code. The most granular government home value data available, updated annually.
• BLS-adjusted project cost - the locally-adjusted cost from your multiplier calculation.

The ROI section frame for a roof replacement page:

"In [City], the median home value is $[Census B25077]. A full roof replacement at $[BLS-adjusted cost] represents [Z]% of home value. Nationally, NAR data shows that homes with replaced roofs sell for 7-10% more than comparable homes with aging roofs - and in a market where the median is $X, that represents $Y in potential resale value."

This framing answers the question homeowners actually have: "I know it costs money - but will it pay off?" It also ties together three separate government data sources in one section, creating content that aggregator sites and real estate blogs are likely to link to.

See our guide on using FHFA HPI data for city-level pages for the full methodology on pulling and calculating home value appreciation by metro.

Schema Markup for Cost Guide Pages

Cost guide pages benefit from several schema types stacked together in a single JSON-LD block. Using multiple schema types is valid and encouraged by Google - each one enables different rich result formats.

Recommended Schema Stack

• Article - basic document schema with datePublished and dateModified. Use dateModified to reflect quarterly data updates. A recently-modified date signals freshness to crawlers.
• FAQPage - wrap the "Is it worth getting a permit?" and similar sections as FAQ items. These can appear as expandable rich results in SERPs, significantly increasing click-through rates for informational queries.
• HowTo - if your page includes a "how to get quotes" section with numbered steps, wrap it in HowTo schema. This enables a step-by-step rich result format.
• BreadcrumbList - essential for site structure signals. Use the pattern: Home > Blog > Cost Guides > [City] Roof Replacement Cost.

For pages with specific cost figures, consider adding a priceRange property to the Article schema. While this is not a standard Article property, structured data parsers at major search engines recognize it and may use it to populate local service result cards.

One important note on schema accuracy: only include specific dollar figures in schema markup if those figures appear in the visible page content. Schema claims that are not substantiated by on-page content can trigger manual actions. The cost range in schema should match the cost range visible in the page's Quick Answer Box.

For a complete picture of the government data sources that power localized content like this, see our guide to the best free government APIs for local SEO content.