Spam call statistics: robocalls, complaints and phone fraud, with sources

The YouMail Robocall Index estimates that 52,477,649,500 robocalls were placed in the U.S. in calendar year 2025, and the FTC logged 1,601,611 robocall complaints in its fiscal year 2025. Those two numbers come from different systems and count different things, so each figure below is printed with its publisher, its period and how it was counted.

All of it is public data from YouMail, the FTC, the FCC, Hiya, Pew Research Center and one TNS case study, and none of it comes from calls to VettedCalls customers. Each figure was checked against the publisher's own page or report on October 10, 2026, and the last section lists every source.

The headline figures

• 52,477,649,500 robocalls in the U.S. in 2025. YouMail Robocall Index estimate, calendar year 2025.
• 3,840,711,700 robocalls in September 2026, about 128.0 million a day. YouMail estimate, one calendar month.
• 1,601,611 robocall complaints. FTC Do Not Call Data Book, fiscal year 2025 (October 1, 2024 to September 30, 2025), up from 1,099,229 in fiscal year 2024.
• About $16 billion reported lost to fraud, through every contact method. FTC, calendar year 2025.
• Phone call was 11% of 2025 fraud reports that indicated a loss, counting only reports that named a contact method other than a write-in "other". FTC, calendar year 2025; a share of reports, not of dollars.
• $948 million reported lost to phone call fraud, with a median loss of $1,500. FTC Consumer Sentinel Data Book, calendar year 2024.
• 27% of 2025 FCC robocall complaints were Debt Relief/Loan Assistance, the largest category. FCC, calendar year 2025.
• 9.9 unwanted calls a week for Americans. Hiya's State of the Call 2026 survey, released March 2, 2026.
• Eight in ten Americans say they don't generally answer their cellphone when an unknown number calls. Pew Research Center survey, July 13 to 19, 2020.

How many robocalls: the YouMail Robocall Index

YouMail, which also sells spam-protection apps, publishes a monthly estimate of U.S. robocalls. It says the index works by extrapolating the data collected from the many tens of millions of calls made each month to YouMail users, so every count in this section is an estimate.

YouMail also splits robocalls by what they are. For July 2026 it reported these shares, which add to 99% as YouMail prints them:

YouMail states that scams and telemarketing together were 46% of July 2026 robocalls. The same table shows why a robocall count is not a spam count: YouMail put 36% of the month's robocalls in alerts and reminders and another 15% in payment reminders.

Complaints: the FTC and the FCC

FTC Do Not Call Data Book, fiscal year 2025

The FTC counts the complaints people file about unwanted calls by fiscal year, and fiscal year 2025 runs from October 1, 2024 to September 30, 2025. The FTC is plain about what this is: "Statistical data on complaints is based on unverified complaints reported by consumers, not on a consumer survey."

The FTC says overall complaints rose in fiscal year 2025 and that unwanted calls remain about 48% lower than in fiscal year 2021, when it received approximately five million reports. Of the 2,618,077 complaints in fiscal year 2025, 1,601,611 were about robocalls, 802,144 were about live callers and 214,322 did not report a call type. Robocall complaints were up from 1,099,229 in fiscal year 2024, a rise of 45.7%, and robocalls were 61.2% of the fiscal year 2025 total (both our arithmetic).

The registry itself had about 258.5 million active registrations on September 30, 2025 (258,515,050 in the data book's chart; its text rounds the same figure to 259 million), and more than 4.7 million numbers were added during fiscal year 2025. Per 100,000 people, New Hampshire has the most active registrations, followed by Connecticut, Vermont, Massachusetts and Kansas.

What the complaints were about

Complaints that don't report a topic are left out of the topic chart, so these counts are a subset of the totals above. The FTC names reducing debt, imposters, and medical and prescription issues as the topics consumers most often reported.

Where people complained most

These are complaints per 100,000 people in fiscal year 2025, across all call types. The rates use 2023 U.S. Census population estimates.

FCC complaint categories

The FCC publishes the share of robocall complaints in each category. The category is the one the consumer picked on the complaint form, and FCC staff don't adjust it. For calendar year 2025 the five largest were these, about 51% of all robocall complaints the FCC received that year:

The FCC also keeps a page that is replaced each month. Its August 2026 edition, updated September 10, 2026, lists Loans/Financing as the largest category at 26%. Check the month on that page before you quote it.

On volume, an FCC fact sheet dated November 25, 2025 puts the illegal-robocall complaints the Commission received at "approximately 120,000 last year alone", without naming the year. Its footnote gives approximately 193,000 for 2019, 157,000 for 2020, 164,000 for 2021, and 119,000 for 2022 and for 2023. These come from the FCC's complaint system. The FTC's figures come from a different one, so don't add them together or set one against the other.

Fraud reported by phone: FTC Consumer Sentinel

Consumer Sentinel counts fraud reports by calendar year. The FTC says its data book is based on unverified reports filed by consumers and is not based on a consumer survey.

Calendar year 2025

People reported losing about $16 billion to fraud in 2025, the highest on record and about 25% more than in 2024, and $3.5 billion of the total was reported lost to imposter scams. Both totals cover every contact method.

For 2025 the FTC also ranks contact methods by share of fraud reports that indicate a loss. This is a share of reports, not of dollars, and it leaves out reports with no contact method and methods classed as "other". The FTC did not collect reports during the 2025 government shutdown.

Calendar year 2024

The 2024 Data Book is the newest source of phone-only dollar figures:

• Phone call: $948 million reported lost, from 284,659 reports, with a median loss of $1,500.
• Text: $470 million reported lost, with a median loss of $1,000.
• Phone calls were the second most commonly reported contact method for fraud in 2024, after email.
• Phone call was 19% of the 1,509,002 fraud reports that had a contact method identified, out of 2,600,678 fraud reports in all.

This 2024 share and the 2025 shares in the table above are measured on different bases: the 2024 share is out of all fraud reports with a contact method, and the 2025 shares are out of reports that indicate a loss, with "other" and no-method reports removed. Neither year tells you whether phone fraud rose or fell.

How people answer unknown calls: Hiya and Pew

Hiya, State of the Call 2026

Hiya, which sells call-protection and branded-calling products, released its report on March 2, 2026. It is based on a survey of over 12,000 consumers in the U.S., UK, Canada, France, Germany and Spain, so it records what people say, not call logs. From its press release:

• American consumers receive an average of 9.9 unwanted calls a week. Across all six markets the average is 7.4.
• Nearly half of Americans, 48%, say phone spam is getting worse.
• One in four Americans say they have received a deepfake voice call in the past 12 months.
• Seniors (55 and older) are losing an average of $1,298 to phone scams, which Hiya says is triple the losses of younger adults. The release doesn't say what that average is taken over.

Hiya's report page is headlined "86% of unknown calls go unanswered." The page names no market and no method, and the full report sits behind a form, so we quote it as Hiya's figure and don't read more into it.

Pew Research Center, July 2020

Pew surveyed 10,211 U.S. adults from July 13 to 19, 2020 on its American Trends Panel. What it found about unknown numbers:

• "Eight-in-ten Americans say they don't generally answer their cellphone when an unknown number calls."
• 19% of U.S. adults overall say they generally answer the phone for an unknown number.
• 67% say their general practice is to not answer an unknown number but to check a voicemail if one is left.
• 14% say they generally ignore any voicemail left after not answering.

That survey is from 2020, so use it as a baseline and say the year when you cite it. Hiya's and Pew's numbers come from different surveys, years and wording, and shouldn't be set side by side.

What the numbers mean for a small-business phone

This part is our reading of the data. Two things in the data above are true at once. Plenty of robocalls are scams or sales pitches: YouMail put scams at 27% of July 2026 robocalls and telemarketing at 19%. And in Pew's 2020 survey, eight in ten Americans said they don't generally answer their cellphone when an unknown number calls.

For a business that runs on a cell phone, both facts bite. A customer calling you for the first time is an unknown number on your screen, and the calls you place to customers can arrive with a label on theirs. We go through the labels in our Potential Spam and Spam Risk guides, and Spam Likely covers what a business can do about its own number. Here is one company's case: a TNS case study (TNS sells branded calling) reports that 91% of the 3,092 calls placed from one unnamed large U.S. credit union's local outbound line were flagged as spam. That's a single line at a single company, not a national rate.

On the incoming side, here is what VettedCalls does with an unknown number. You keep your cell number and forward it to a VettedCalls number we set up for you, so we pick up every call first. An unfamiliar caller is greeted in your business name and asked who's calling and what they need, and an AI reads the reply. A caller who isn't a priority call then hears a press 1 or press 2 choice, and a caller who stayed silent or sounds like spam gets the same one. Anyone who doesn't press a key hears the blocked-caller message and the call ends. If that caller stayed silent or sounded like spam, the number gets one flag on a spam list shared across every VettedCalls customer, and at 10 counted flags it's blocked for all of them. Saved contacts skip screening.

Screening and the shared spam list are part of Pro, the plan you can buy today. Our call screening page has every step in order, and pricing has the plan details.

Two limits. VettedCalls handles incoming calls only, so we can't change how a customer's carrier labels the calls you place. And every call we answer counts toward your voice minutes, blocked spam included; a blocked call usually uses one. We don't print a block rate here. Every number on this page is public data, and none of it says how many unwanted calls your own line will get.

Read this before you quote a number

• Check the period. FTC Do Not Call figures are fiscal years (fiscal year 2025 is October 1, 2024 to September 30, 2025). FTC Consumer Sentinel figures are calendar years. YouMail gives calendar months and years. The FCC's 2025 shares are a calendar year, and its monthly page covers one month.
• Complaints are not counts of calls. They are what people chose to report, and the FTC says they aren't a survey.
• YouMail's counts are estimates, extrapolated from calls to its users.
• Don't combine the FTC and the FCC. The FTC's 1,601,611 robocall complaints in fiscal year 2025 and the FCC's yearly counts (approximately 119,000 to 193,000 from 2019 to 2023) come from separate complaint systems.
• Don't compare the FTC's 2024 and 2025 phone shares. They rest on different bases, as explained above.
• Credit vendor figures to the vendor. Hiya and TNS sell call-protection or branded-calling products, so name them with every figure you take from them.
• Name the year on old surveys. Pew's figures are from July 2020.
• Monthly pages change. The YouMail index and the FCC's monthly page are replaced as new months arrive, so quote the month.

When you cite a figure, credit the publisher and the period, for example "YouMail's estimate for calendar year 2025", and link the publisher's page from the table below.

Where these numbers come from

Every row links the publisher's own page, and each one was checked on October 10, 2026.