SG&A Ratio Benchmarks by Industry

SG&A ratio is simple math: total SG&A divided by total revenue, times 100. It tells you how many cents of every dollar a company brings in get spent on overhead. On the income statement, it sits right after gross profit: revenue minus cost of goods sold gives you gross profit, and gross profit minus SG&A gives you operating income. Two companies can post the same gross margin and still end up with wildly different operating income, and SG&A is usually the reason why.

Selling costs, general costs, and administrative costs (think commissions and advertising on one end, rent and insurance in the middle, legal and IT on the other) don't get sorted the same way from one industry to the next. A cost that lands in SG&A at a software company might get buried in cost of goods sold at a factory. So SG&A benchmark data tends to get published sector by sector, not as one giant table covering every business. Any cost a company shifts into COGS simply disappears from the SG&A line, so an industry that packs more expense into production costs will show a lower ratio almost by default. The number reflects how a company does its bookkeeping almost as much as it reflects how the company actually spends money. Which means comparing a mining company's SG&A ratio to a software company's ratio is a bit like comparing a shark to a bicycle. Both move, but that's about where the similarity ends.

The sector-level spread: from single digits in mining to a substantial share in services

The gap between the leanest and the most overhead-heavy sectors is enormous, and Schonfeld's sector summary lays it out with median SG&A-to-sales ratios alongside 10th-to-90th-percentile ranges for every major group. Mining and extraction is at the bottom, with a median in the high single digits and 10th-percentile companies running ratios in the low single digits. Construction runs a bit higher, with a median in the low-to-mid teens and a fairly tight spread around it. Wholesale trade is in the high teens, with its leanest performers again down in the low single digits. Transportation, communications, and utilities post a median in the mid-teens, with the leanest companies in that group also down near the low single digits. Agriculture, forestry, and fishing come in with a median roughly in the high 20s, but the spread is wide, stretching from the low single digits at the 10th percentile to well over double the median at the top. Retail trade is in the high 20s too, but its distribution is tighter than manufacturing's. Finance, insurance, and real estate run from the high 20s up into the low 30s. Manufacturing posts a median in the low 30s but carries one of the widest spreads of any sector, with the leanest companies in the high single digits and the heaviest spenders running at multiples of the median. Services top the chart with the highest median of any sector, and even its leanest performers at the 10th percentile are still in the low-to-mid teens, well above the leanest companies anywhere else.

Read these as medians, not targets. The percentile range tells you more than the single number does. Mining is the clearest example: the gap between its 10th and 90th percentile stretches from the very low single digits all the way to the mid-70s. A mining company at the high end of that range isn't necessarily doing anything wrong, and one at the low end isn't automatically lean and mean. Where a company sits in that spread depends heavily on its size and its specific corner of the sector, so the median by itself is an incomplete yardstick. Schonfeld's data accounts for some of this by splitting companies into three revenue bands, because SG&A habits shift with company size regardless of industry.

Mining, construction, and wholesale trade's leanest ratios

The lowest-ratio sectors share one trait: almost all of their money goes into cost of goods sold or capital spending, leaving a thin slice for selling and administrative costs.

Mining and extraction make money on volume and commodity price, not on convincing a buyer to pick one company's iron ore over another's. There's no sales floor for copper. The cost to pull the material out of the ground dominates the budget and sits in COGS, not SG&A.

Construction works the same way. Labor, materials, and subcontractors are tied directly to each project and get classified in COGS. What's left for SG&A is estimating, bidding, and general management, a thin layer on top of a much larger pile of direct project costs.

Wholesale trade runs on thin margins and high volume, and the distributor's value comes from logistics and availability, not brand-building or a persuasive sales pitch. The Schonfeld report on retail and wholesale trade covers 266 publicly traded retail companies and 119 wholesale companies across durable and non-durable goods, and it shows wholesale medians sitting well below retail medians in that same study.

If a wholesale or construction company's ratio climbs toward the top of its sector's range, that doesn't mean it's investing in growth. In models built around cost discipline rather than persuasion, if the ratio rises it more often means overhead is creeping into functions that were never supposed to carry much weight.

Manufacturing's wide internal range

Manufacturing's median is in the low 30s, but the range around that number is among the widest of any sector tracked, stretching from the high single digits at the 10th percentile to multiples of the median at the top. That spread alone should tell you not to trust the headline figure unless you know exactly which corner of manufacturing you're standing in.

The sector's variety bakes in the reason. Commodity-adjacent businesses, petroleum refining, primary metals, paper, load most of their costs into COGS and run lean, much like mining does. Branded consumer goods makers, apparel, food, chemicals, need real marketing and sales muscle to win shelf space and loyalty, so their ratios climb. Instrument and machinery makers often carry heavy engineering teams, technical sales support, and after-sale service, and all of that tends to land in SG&A, not production cost.

Size adds another layer on top of sub-industry. Larger manufacturers typically see their ratio compress through scale, because they spread fixed costs over more revenue. Smaller manufacturers carry the same fixed administrative costs over a smaller base, which pushes their ratio up even within the same sub-industry. A manufacturer who benchmarks against the sector median without accounting for sub-industry and size is comparing against an average built from companies that may look nothing like their own.

Retail trade's mid-range ratio and the cost of physical presence and brand

Retail trade's median is in the high 20s, meaningfully above wholesale despite the two sectors sharing much of the same supply chain. The Schonfeld retail and wholesale report covers 266 publicly traded companies across eight retail sub-industries: building materials, hardware, garden, and mobile homes; general merchandise stores; food stores; automotive dealers and gasoline service stations; apparel and accessory stores; home furniture, furnishings, and equipment stores; eating and drinking places; and miscellaneous retail.

Retail and wholesale ratios diverge based on what each one is actually selling. Wholesale moves goods efficiently from point A to point B. Retail sells an experience on top of the goods, which costs money. Food stores and automotive dealers tend to sit at the lower end of retail's range, because their margins are thin and cost control is how they survive. Apparel and home furnishings sit higher, because brand image and the store environment itself are part of what the customer is paying for, not just an afterthought.

Rent, store staffing, and consumer marketing drive the difference. These costs scale with how many stores a company operates and how big its brand ambitions are, not with how many units move through the register. That makes them semi-fixed costs that are hard to cut quickly when sales slow down, which is part of why retail's ratio sits where it does even in a rough year.

Finance, insurance, and real estate: high medians driven by relationship-intensive selling and compliance

Finance, insurance, and real estate companies post medians that run between the high 20s and low 30s. That reflects two forces at once: winning customer trust costs more in industries where relationships matter more than a catchy ad campaign, and legal and compliance staff add unavoidable overhead. The Schonfeld report on this sector covers a large sample of publicly traded companies across seven sub-industries: depository institutions, non-depository credit institutions, security and commodity brokers, dealers, exchanges, and services, insurance carriers, insurance agents, brokers and service, real estate, and holding and other investment offices.

Insurance carriers and agents carry heavy selling costs through agent networks and broker commissions, plus administrative overhead to manage claims. Banks and depository institutions run branch networks, compliance teams, and relationship managers almost entirely through SG&A, and since money itself has no cost of goods sold in the traditional sense, a larger share of total spending appears in that single line than it would for a company making a physical product. Real estate and investment firms vary enormously by business model: companies that own property outright can run leaner, while brokerage and advisory firms often run much heavier.

A bank's SG&A ratio and a factory's SG&A ratio aren't measuring the same slice of the business. Financial firms have no traditional cost of goods sold to divide against, so by definition they route a bigger share of total spending through SG&A, and a high ratio here doesn't mean what it would in a goods-producing industry. Size matters here too: small financial firms carry compliance and legal costs that barely shrink no matter how small their revenue is, while large institutions spread that same regulatory burden across a much bigger base.

Transportation, communications, and utilities: infrastructure costs compress the ratio despite heavy fixed overhead

Transportation, communications, and utilities companies post a median in the mid-teens, lower than their actual overhead might suggest. Their biggest costs, the network itself, the fleet, the fuel, the depreciation on all of it, get classified as production or capital cost.

What's left in the SG&A line covers customer billing, service functions, regulatory affairs, executive management, and marketing, a fairly thin layer sitting on top of a much larger pile of infrastructure spending recorded elsewhere on the income statement. The companies at the 10th percentile of this sector, often utilities with a captive, rate-regulated customer base, barely spend on sales and marketing at all, since customers have nowhere else to go.

A low ratio in this sector shouldn't read as a model to copy for a business with no fleet, no pipeline, and no power plant sitting on its balance sheet. The lean number here comes from classification, not from frugality. A company with real production costs and real capital assets will always compress its SG&A ratio relative to a company whose only real asset is its people, which is exactly where the next sector picks up.

Services: the highest median of any sector

Services post the highest median of any sector Schonfeld tracks, and for a straightforward reason: the product and the overhead are often the same thing, people. A factory can point to a machine and say "that's production cost," and a services firm often can't make that same clean separation. The Schonfeld services report covers a large sample of publicly traded companies across a wide range of sub-industries within the sector.

In professional and knowledge-based firms, the people delivering the service are frequently the same people selling it and managing the client relationship. Their pay gets split between cost of goods sold and SG&A depending on how each company chooses to classify it, and that inconsistency is a big part of why the spread within this sector is so wide, with the 90th percentile reaching well above the median while the 10th percentile is still in the low-to-mid teens.

Software and SaaS companies usually get classified inside this sector too, and they push the high end of the range even further up. Their cost of goods sold, mostly hosting and delivery-related support, stays relatively small, while their sales, marketing, and customer success teams sit almost entirely in SG&A. That combination produces some of the highest SG&A ratios of any business model tracked, showing that a single benchmark number means nothing without knowing the structure behind it.