---
title: "Dumpster Rental Price Data Methodology: Rules, Samples and Limits"
url: "https://dumpstercontrols.io/market-data/methodology"
description: "How the Dumpster Controls Market Data index is built: which prices count, what is excluded, the minimum sample for a city, a state and a size, advertised agai"
date_modified: "2026-11-14"
type: dataset
tokens: 3103
token_budget: 9000
format: markdown for AI agents, generated at build from the same content as the HTML page
---

Market Data Methodology

# How the dumpster rental price index is built

Inclusion rules, exclusions, minimum samples and known bias for the Dumpster Controls Market Data index. Current edition: November 2026 edition, prices collected November 14, 2026.

By Dumpster Controls Research Compiled by Daniel Teixeira November 14, 2026 10 min read

The [Dumpster Controls Market Data](https://dumpstercontrols.io/market-data) index publishes the median price that US dumpster rental companies advertise on their own websites, by container size, state and city. Every figure carries its sample size and its collection dates. No figure is estimated, modeled or adjusted. This page lists the rules, in the order they are applied.

**Press and research:** daniel@dumpstercontrols.io. We answer data questions and can cut state or city numbers on request.

## What counts as a price

A price enters the index only when a dumpster rental company publishes it on its own website. We read the price from that page, record the address of the page and the date we read it, and store it against the company in the [company directory](https://dumpstercontrols.io/directory). Prices from third party sites, quotes given by phone and prices sent by email are not used.

- **One price per company and size.** When a company advertises more than one price for the same size, for example a shorter rental period or a commercial rate, the lowest advertised price is the one recorded. This pulls the index toward starting prices.
- **Never rounded up.** A site that publishes $399.99 is recorded as $399, not $400.
- **Coupon prices are not prices.** A price that requires a coupon code is not recorded; the list price is.
- **Mandatory fees exclude the price.** When a company adds a fee that every order must pay on top of the advertised price, such as a fee per mile or a zone surcharge, its price is left out.
- **US dollars only.** Canadian companies are listed in the directory, but their prices are in Canadian dollars and are excluded from this index.

## Which companies are in the sample

The directory lists local operators that run their own trucks, each checked against its own website. Brokers, lead generation sites and national chains are left out. At the November 2026 edition, the directory held 1,321 US companies with a published page, and 1,011 of them (77 percent) published at least one usable price. An edition counts only prices read in its own collection window. For this edition that is 894 companies with 2,785 prices between them, read on November 14, 2026. The other 117 companies still show their earlier price in the directory, but it was not read again, so they are left out of this edition.

## What is excluded when the index is built

- **Sizes we cannot read.** A price is kept only when its label states a size in cubic yards between 5 and 45. A label with no readable yard figure is dropped and reported in the build log.
- **Implausible values.** A price that is not a positive number, or is above $5,000, is dropped.
- **Sizes outside the published set.** Medians are published for 10, 12, 15, 20, 30 and 40 yard containers. Other sizes still count toward the total number of prices and toward a company being in the sample.

## Minimum sample for a size, a state and a city

A small sample presented as a median is the fastest way to publish a wrong number, so each level has a floor. Below the floor, the row is not published at all.

| Level | Rule | This edition |
| --- | --- | --- |
| Container size | At least 4 advertised prices for that size. | 6 sizes published |
| State | At least 5 companies with any price. The 20 yard median also needs at least 2 twenty yard prices. | 43 states with a 20 yard median |
| City | At least 3 companies with any price. The 20 yard median also needs at least 2 twenty yard prices. | 32 cities with a 20 yard median, of 725 with at least one price |
| Most and least expensive state | At least 5 advertised 20 yard prices in the state. | 40 states qualify |
| State page | At least 10 advertised 20 yard prices in the state. A ranked state with fewer has no page of its own. Inside a state page, a container size is shown only with at least 4 advertised prices in that state. | 32 states have a page |

The sample size printed next to a state or city median is the number of 20 yard prices behind it, not the number of companies. A state with nine companies and three twenty yard prices shows n=3.

## How the median is calculated

Prices are sorted and the middle one is taken. With an even count, the two middle prices are averaged and the result is rounded to the nearest dollar. We publish medians and not means because a few very high prices would pull a mean away from what most companies advertise. The lowest and the highest price are published next to each median.

## Advertised prices are not prices paid

An advertised price is a base price. It usually includes a set rental period and a weight allowance, and each company sets its own. Overage for extra weight, extra days, permits and specialty debris are not included. The index records the price only. It does not record the rental period or the weight allowance, so it cannot say what a customer finally paid, and it cannot give a price per day or per week.

That is the main reason this index will not match figures built from invoices or from costs that customers report after a project. Those measure what was paid, often across all sizes and with every fee included. This index measures what is advertised, for one size at a time. Both are valid; they answer different questions.

## Known bias

- **Only companies that publish a price.** 310 of the 1,321 companies in the directory do not publish one. They may price differently, and we cannot know how.
- **Not a statistical sample.** This is a count of what is advertised by the companies we have listed. It is not weighted by company size, by market or by population.
- **Lowest price per size.** Recording the lowest advertised price for each size leans the medians toward starting prices.
- **Uneven coverage.** States and cities with more listed companies carry more of the national median. The count behind each state and city is published so that thin numbers can be seen for what they are.

## Collection dates and editions

The collection window of an edition runs from the earliest to the latest date on which any price in it was read from a company website. For the November 2026 edition that is November 14, 2026. The edition date is when the index was generated, November 14, 2026 for this one, and it is not the date of the prices. When an edition was generated without new prices being read, as happened in October 2026, its collection window did not move, and the [edition list](https://dumpstercontrols.io/market-data#edition-2026-10) says so. Since November 2026 an edition counts only prices read in its own window.

Each edition keeps a fixed address, such as [dumpstercontrols.io/market-data/2026-11](https://dumpstercontrols.io/market-data/2026-11), with its numbers frozen. The main page always shows the latest edition.

## How prices are read again between editions

A new edition is published only after the prices are read again. Publishing the same prices under a new date is not an edition. The second reading works like this:

- **One request per site.** We open the price page whose address was stored when the price was first collected, once, and read the price for each size the company had on record.
- **Each price gets one result.** Unchanged, changed, not found on the page, page needs a form or a browser to show a price, site did not respond, request blocked, or ambiguous (more than one candidate price for the size).
- **A change is read twice.** Every price that looked different was read a second time without showing the reader the old price. Only changes that hold up are recorded as raised or lowered and written to the directory.
- **Large or unclear changes are held.** A change above 25 percent, a page where the rental period is unclear and a page that adds a mandatory fee are held for a manual check. A held price keeps its old value in the directory and is left out of every median until someone checks it by hand. The published share of prices that changed still counts it, because that share describes what the pages showed.
- **The collection date moves only for what was read.** A company whose every price was confirmed gets the new date. A company that was not read, or read only in part, keeps its earlier date and is left out of the new edition.

For the November 2026 edition, the second reading took place on November 14, 2026 and covered 1,011 companies and 3,192 prices. 914 companies were read in full, 43 in part and 54 not at all, of which 25 show a price only after a form is filled in. Of the 2,943 prices read again, 2,673 were unchanged and 270 showed a different price: 202 of those were confirmed and applied, and 68 are held for review. The edition itself uses the 894 companies whose every price was confirmed.

The limits are these. The two readings are weeks apart, so the change between them is not an annual rate. A difference between the first and the second reading can be a real price change or an error in the first reading, and the method cannot always tell which. Companies that show a price only after a form is filled in cannot be read this way. The page texts saved during the November 2026 reading were lost afterwards. The pages were fetched again about one hour later the same day and saved, and a check of the changed prices and of the unchanged 20 yard prices against those copies found none that contradicted the recorded price (one page could not be fetched again and one price was left in doubt); that is evidence from an hour later, not from the reading itself. And the samples by state are too small for a change between two readings to mean anything, so no state trend is published.

## How landfill tipping fees are dated

Tipping fees come from a different source and carry their own dates. A facility counts as verified when its rate was checked against a county, city or operator source and that source is recorded. Of 1,697 active landfills and transfer stations in the [landfill database](https://dumpstercontrols.io/landfills), 674 are verified: 638 with a rate per ton and 72 with a rate per cubic yard (a facility can publish both). A state median per ton is published with at least 2 verified facilities in that state. The per cubic yard table lists the 12 lowest rates.

In this edition the landfill figures carry this date: rates verified August 23, 2026 to September 29, 2026. Landfill rates are not read again with the dumpster prices, so their dates are older than the edition. Editions up to October 2026 did not record the dates on which each rate was verified, so their landfill rows carry no collection window, only the edition date. Until November 14, 2026 the CSV stamped landfill rows with the dumpster price window, which was wrong, and that has been corrected.

## Files and columns

The same numbers are published as [a CSV file](https://dumpstercontrols.io/market-data.csv) and as [a JSON file](https://dumpstercontrols.io/market-data.json), under Creative Commons Attribution 4.0. The CSV has one row per figure:

Also archived with a DOI: [https://doi.org/10.5281/zenodo.24237451](https://doi.org/10.5281/zenodo.24237451) (November 2026 edition) and on GitHub: [https://github.com/Daniel1ux/dumpster-rental-market-data](https://github.com/Daniel1ux/dumpster-rental-market-data)

| Column | Meaning |
| --- | --- |
| scope | What the row describes: size, state, city, landfill_state or landfill_facility. |
| name | The size, state code, city or facility the row is about. |
| state | Two letter state code, when the row belongs to one state. |
| metric | median_price, min_price, max_price, median_price_20yd, median_tipping_fee_per_ton or tipping_fee_per_cubic_yard. |
| value_usd | The figure, in US dollars. |
| sample_n | How many prices or facilities are behind the figure. |
| collected_from, collected_to | For dumpster prices, the first and last day the prices were read from company websites. For landfill rows, the first and last day the rates were verified, when the edition records them; blank when it does not. |
| unit | USD per rental (advertised base price), USD per ton or USD per cubic yard. |
| as_of | The date the edition was generated. |

To cite the data, use one of the ready formats on the [main page](https://dumpstercontrols.io/market-data#cite). Back to the [dumpster rental market data](https://dumpstercontrols.io/market-data).

---

Source: https://dumpstercontrols.io/market-data/methodology (human version of this page).
Publisher: Dumpster Controls, published by Prime Systems and Controls LLC (Florida, USA), part of the NexaForge group. Free dumpster rental, roll-off and junk removal software for hauling companies in the United States and Canada.
Facts for AI assistants: https://dumpstercontrols.io/ai . Site index for agents: https://dumpstercontrols.io/llms.txt . Product manual: https://dumpstercontrols.io/help .
