US phone scam statistics 2026: Losses, robocalls and trends
US phone scam statistics 2026: $1.1 billion lost to scam calls in 2025, a $1,835 median loss, 3.8 billion robocalls in September, and who loses most.
In brief
People told the FTC they lost $1.1 billion in 2025 to scams that began with a phone call, and a phone scam takes the most per victim: a $1,835 median loss, higher than any other contact method. Robocalls are slowly falling, to an estimated 3.8 billion in September 2026, but phone-scam losses and complaints are not.
- Texts start more scams, with 411,438 reports in 2025, but calls cost more: 19% of phone-call reports involved a loss, against 9% for texts.
- Government impersonation is the costliest phone scam, at $421.5 million in 2025.
- People 80 and over lose money least often but most heavily, with a $2,000 median loss, and phone calls are their costliest contact method.
- Do Not Call robocall complaints rose 45.7% in fiscal 2025 even as robocall volume fell.
- 96% of the caller ID numbers in this year's FCC unwanted-call complaints were reported only once; this does not measure the effectiveness of blocking.
- In 47.6% of this year's FCC complaints that gave both numbers, the callback or promoted number was not the number shown on caller ID.
Figures current to November 2026, from 15 named sources. Every figure below links to its source. Download the data.
These US phone scam statistics cover reported losses, victim age groups, scam types, robocall estimates and unwanted-call complaints. The page combines completed-year figures with available 2026 data; each chart states its period and source so those totals are not confused.
The FTC and FBI count reports, the FCC records complaints, and YouMail estimates call volume. Those measures are different and should not be added together. Use the source links to check a figure and the CSV at the end to reuse the data.
- $1,835
Median loss when fraud began with a phone call, the highest of any contact method
2025 · Federal Trade Commission
- $2.9 billion
Reported to the FBI as lost to call-center fraud: tech support and government impersonation
The FBI says losses exceeded this.
How much money do people lose to phone scams?
People told the FTC they lost $1.1 billion in 2025 to fraud that began with a phone call, across 290,429 reports. That is a small slice of all reported fraud, which came to $15.9 billion from 3 million reports. It is also a slice that has grown every year: phone-call losses rose 39.5% from 2022 to 2025, while the number of phone-call reports hardly moved.
2022–2025
Reported losses to fraud that began with a phone call, 2022 to 2025
Phone-call losses rose every year, up 39.5% from 2022 to 2025, while the number of phone-call reports barely moved. Each report is costing more.
2025
US dollars
Year
Select a point to inspect
| Year | US dollars |
|---|---|
| 2022 | $794 million |
| 2023 | $851 million |
| 2024 | $948 million |
| 2025 | $1,108 million |
The typical loss is rising too. The median phone-call loss was $1,400 in 2022 and $1,835 in 2025, and it reached $1,997 in the first half of 2026. The share of phone-call reports that involved a loss climbed alongside it.
Scroll horizontally to compare all columns
The matched half-years tell the current story. From January to June 2026, people reported losing $665 million to fraud that began with a phone call across 162,991 reports, against $575 million across 156,136 reports in the same months of 2025. Losses rose 15.7% while reports rose 4.4%. The dashboard does not print a combined median for the first half of 2025; its median was $1,800 in the first quarter and $1,800 in the second, against $1,997 for the first half of 2026. Dashboard losses are rounded to whole millions.
How do phone scam losses compare with text and email scams?
Which channel is worst depends on what you count. Text messages start the most scams: 411,438 reports in 2025, against 290,429 for phone calls. Social media takes the most money in total, $2.1 billion. But a phone call is the most expensive way to be scammed. The median loss on fraud that began with a call was $1,835, higher than for any other contact method.
Median reported loss by how the scam first made contact, 2025
A scam that starts with a phone call takes more from the typical victim than any other kind: $1,835, nearly twice a text and six times social media.
| Category | US dollars |
|---|---|
| Phone call | $1,835 |
| Text | $1,000 |
| $844 | |
| Other | $550 |
| $415 | |
| Social media | $300 |
| Website or apps | $169 |
| Online ad or pop-up | $130 |
Calls also turn into losses far more often. 19% of phone-call reports included a money loss, against 9% of text reports. So phone calls produced 1.7x the losses of text messages, from fewer reports.
Bars compare median loss across rows.
Scroll horizontally to compare all columns
Which scams arrive by phone?
Impersonation dominates. Among scams that began with a phone call in 2025, callers posing as a government agency took the most money, followed by callers posing as a business.
Reported losses by scam type, where the first contact was a phone call, 2025
Callers pretending to be the government took the most: $421.5 million, almost twice the next type, callers pretending to be a business.
| Category | US dollars |
|---|---|
| Government imposters | $421.5 million |
| Business imposters | $218.1 million |
| Investments | $150.2 million |
| Prizes, sweepstakes and lotteries | $102.5 million |
| Tech support | $43 million |
| Job scams | $15.7 million |
| Online shopping | $13.4 million |
| Romance | $13.2 million |
| Credit and debt counseling | $12.3 million |
| Fake checks | $9.2 million |
Scroll horizontally to compare all columns
Government impersonation is growing well beyond the phone. Across all contact methods, government-imposter reports nearly doubled from 2022 to 2025, while the median loss fell from $1,500 to $645: more attempts, each taking less. The FTC puts all imposter losses at $3.5 billion in 2025, including about $920 million lost to government impersonators.
Scroll horizontally to compare all columns
The FBI's view of call-center fraud
The FBI sees the same pattern. Its Internet Crime Complaint Center groups tech-support and government-impersonation scams as call-center fraud, and counted more than 80,000 such complaints in 2025 with losses above $2.9 billion. Government-impersonation losses reported to the FBI rose 96.7% in a single year.
Scroll horizontally to compare all columns
Who loses the most to phone scams?
Age changes the picture more than anything else. Across all fraud in 2025, people 80 and over were the least likely to report losing money, at 22.2% of their reports, but had the highest median loss, $2,000. People in their twenties lost money more often and lost far less, a median of $305.
Bars compare median loss across rows.
Scroll horizontally to compare all columns
The phone matters most at the oldest ages. The FTC found that for people 80 and over, phone calls cost more than any other contact method, while social media cost the most for every younger age group. For people 60 and over, the median loss on fraud that began with a call was $3,000 in 2025. Among older adults who lost $10,000 or more to a business or government imposter, 41% said a phone call came first.
Older adults and phone scams
- $2,000
Median fraud loss for people 80 and over, the highest of any age group
2025 · Federal Trade Commission
- $3,000
Median loss for people 60 and over when fraud began with a phone call
2025 · Federal Trade Commission
- $7.7 billion
Losses reported to the FBI by people 60 and over
The FBI's numbers for people 60 and over are larger again: 201,266 complaints and $7.7 billion in losses in 2025, up 59% from 2024. 12,444 people in that age group reported losing more than $100,000.
How many robocalls do Americans get?
YouMail, which makes a call-blocking app, estimates that 3.8 billion robocalls reached US phones in September 2026. That is about 128 million a day and 11.7 for every person. Its estimate for all of 2025 is 52.5 billion.
Jan 2025–Sep 2026
Robocalls placed to US phones each month, January 2025 to September 2026 (YouMail estimate)
Volume peaked in April 2025 and has drifted lower since. September 2026 was about 3.8 billion calls, roughly 128 million a day.
Count
Sep 2026
Month
Select a point to inspect
| Month | Count |
|---|---|
| Jan 2025 | 4.7 billion |
| Feb 2025 | 4.5 billion |
| Mar 2025 | 4.8 billion |
| Apr 2025 | 5 billion |
| May 2025 | 4.8 billion |
| Jun 2025 | 4.5 billion |
| Jul 2025 | 4.3 billion |
| Aug 2025 | 4.2 billion |
| Sep 2025 | 4.1 billion |
| Oct 2025 | 3.7 billion |
| Nov 2025 | 3.9 billion |
| Dec 2025 | 4.1 billion |
| Jan 2026 | 3.9 billion |
| Feb 2026 | 3.8 billion |
| Mar 2026 | 4.2 billion |
| Apr 2026 | 4.2 billion |
| May 2026 | 4.1 billion |
| Jun 2026 | 4.3 billion |
| Jul 2026 | 4.3 billion |
| Aug 2026 | 3.9 billion |
| Sep 2026 | 3.8 billion |
Volume is falling, slowly. January to September 2026 ran 10.4% below the same months of 2025, and the latest twelve months add up to 48.3 billion. The peak year in the index was 2019, at 58.5 billion.
2019–2025
Robocalls placed to US phones each year, 2019 to 2025 (YouMail estimate)
After the 2019 peak and a dip in 2020, annual volume has held between about 50 and 55 billion a year since 2021. The decline is real but slow.
2025
Count
Year
Select a point to inspect
| Year | Count |
|---|---|
| 2019 | 58.5 billion |
| 2020 | 45.9 billion |
| 2021 | 50.5 billion |
| 2022 | 50.4 billion |
| 2023 | 55 billion |
| 2024 | 52.8 billion |
| 2025 | 52.5 billion |
Complaints show what those calls look like up close. Of the unwanted-call complaints filed with the FCC from January to October 2026 that named a call type, 47.5% described a prerecorded voice. The rest named a live caller or a text message. That split describes complaints people chose to file, not all robocalls, and complaints that named no type are left out.
January to October 2026
Which unwanted calls do people complain about most?
Prerecorded robocalls made up 47% of January to October 2026 unwanted-call complaints that named a call type, live callers 31% and texts 22%.
Prerecorded voice (robocall)
50,497
Live caller
32,649
Text message
23,267
| Category | Complaints | Share |
|---|---|---|
| Prerecorded voice (robocall) | 50,497 | 47% |
| Live caller | 32,649 | 31% |
| Text message | 23,267 | 22% |
When do scam calls come?
DeepSearch calculated the two findings below from the FCC's public unwanted-call complaint records, using the snapshot and filing period shown in each finding. The filing period selects the records; the reported call time describes what happened within that sample. Each citation includes the calculation, scope, limitation and source so those details travel with the number. Percentages are rounded.
Reported calls cluster in daytime and evening hours
93% of FCC unwanted-call complaints with a usable call date and time describe calls from 8 a.m. to before 9 p.m., in the complainant's local time.
US unwanted-call complaints filed with the FCC; records without a usable call date and time are excluded. Filing period: January to October 2026.
Calculation: 91,748 / 98,591 × 100 = 93% (rounded). Count complaints whose reported call hour falls in the daytime/evening interval, then divide by all timed complaints.
Limitations: Self-reported times do not establish whether a call was fraudulent or lawful. Calls people do not notice, including overnight calls, may be underreported.
January to October 2026
What time of day do unwanted calls come?
93% of timed FCC unwanted-call complaints filed from January to October 2026 describe calls from 8 a.m. to before 9 p.m. local time.
12 p.m.
Complaints · 10%
Hour the call came, complainant's local time
Select a point to inspect
| Hour the call came, complainant's local time | Complaints | Share |
|---|---|---|
| 12 a.m. | 390 | 0.4% |
| 1 a.m. | 245 | 0.2% |
| 2 a.m. | 156 | 0.2% |
| 3 a.m. | 152 | 0.2% |
| 4 a.m. | 139 | 0.1% |
| 5 a.m. | 300 | 0.3% |
| 6 a.m. | 1,301 | 1.3% |
| 7 a.m. | 1,664 | 1.7% |
| 8 a.m. | 5,268 | 5.3% |
| 9 a.m. | 8,917 | 9% |
| 10 a.m. | 9,842 | 10% |
| 11 a.m. | 9,878 | 10% |
| 12 p.m. | 10,087 | 10% |
| 1 p.m. | 10,004 | 10% |
| 2 p.m. | 9,087 | 9.2% |
| 3 p.m. | 8,399 | 8.5% |
| 4 p.m. | 6,942 | 7% |
| 5 p.m. | 5,337 | 5.4% |
| 6 p.m. | 3,789 | 3.8% |
| 7 p.m. | 2,576 | 2.6% |
| 8 p.m. | 1,622 | 1.6% |
| 9 p.m. | 1,476 | 1.5% |
| 10 p.m. | 515 | 0.5% |
| 11 p.m. | 505 | 0.5% |
Weekdays dominate the timed complaint sample
91% of FCC unwanted-call complaints with a usable call date and time describe calls on Monday through Friday.
US unwanted-call complaints filed with the FCC; weekdays are taken from the reported call date, not the filing date. Filing period: January to October 2026.
Calculation: 89,986 / 98,591 × 100 = 91% (rounded). Add the Monday-through-Friday complaint counts and divide by all complaints with a usable call date and time.
Limitations: This is a share of self-selected complaints, not the chance of receiving a scam call. The pattern alone does not establish how call centers operate.
Sunday's share of the timed sample was 2.9%.
January to October 2026
Which days do unwanted calls come?
Workdays: 91% of the unwanted calls people reported to the FCC from January to October 2026 came Monday to Friday, and an average weekday got 6 times as many as a Sunday.
Thursday
Complaints · 19%
Day the call came
Select a point to inspect
| Day the call came | Complaints | Share |
|---|---|---|
| Monday | 16,997 | 17% |
| Tuesday | 18,209 | 18% |
| Wednesday | 18,778 | 19% |
| Thursday | 19,083 | 19% |
| Friday | 16,919 | 17% |
| Saturday | 5,728 | 5.8% |
| Sunday | 2,877 | 2.9% |
These patterns describe voluntarily reported call times. They do not establish why calls were placed at those times, how call centers operate, or whether a particular call was fraudulent or lawful. Calls people do not notice, including overnight calls, may also be reported less often.
How many Do Not Call complaints are reported?
Robocall volume and robocall complaints are moving in opposite directions. Do Not Call complaints to the FTC rose to 2.6 million in fiscal 2025, and complaints about robocalls jumped 45.7% to 1.6 million, or 61.2% of the total. Complaints are still about 48% below fiscal 2021.
FY2021–FY2025
Do Not Call complaints filed with the FTC, fiscal 2021 to 2025
Complaints fell by more than half after fiscal 2021, then turned up again in fiscal 2025, driven by robocalls.
FY2025
Count
Fiscal year
Select a point to inspect
| Fiscal year | Count |
|---|---|
| FY2021 | 5 million |
| FY2022 | 3.1 million |
| FY2023 | 2.1 million |
| FY2024 | 2.1 million |
| FY2025 | 2.6 million |
The robocalls people complain about are mostly selling debt relief.
Do Not Call robocall complaints by what the call was about, fiscal 2025
Debt-relief robocalls drew more complaints than imposters and medical calls combined.
| Category | Count |
|---|---|
| Reducing debt | 389,560 |
| Imposters | 152,764 |
| Medical and prescriptions | 129,076 |
| Energy, solar and utilities | 17,963 |
| Warranties and protection plans | 10,813 |
| Home improvement and cleaning | 9,478 |
| Computer and tech support | 6,198 |
| Lotteries, prizes and sweepstakes | 4,637 |
| Home security and alarms | 3,711 |
| Money-making opportunities | 3,675 |
| Vacation and timeshares | 3,153 |
Scroll horizontally to compare all columns
The registry keeps growing, to 258.5 million active numbers. Registration only binds callers who follow the law; scammers ignore it.
The FCC hears the same rise. It received 125,199 unwanted-call complaints in 2025, up from a low in 2023.
Scroll horizontally to compare all columns
Complaint rates vary widely by state. From January to October 2026, Colorado filed 11,154 unwanted-call complaints per 100,000 residents, the highest rate of any state and more than three times the national rate of 3,740.4. Rates also reflect how well people in a state know they can complain to the FCC, so a high rate shows reporting as well as calls received.
January to October 2026
Which states report the most unwanted calls per person?
Colorado filed 111.5 unwanted-call complaints per 100,000 residents from January to October 2026, the highest rate in the country; the national rate was 37.4.
| Category | Complaints per 100,000 residents |
|---|---|
| Colorado | 111.5 |
| Arizona | 54.1 |
| District of Columbia | 54.1 |
| Florida | 50.1 |
| Virginia | 49.5 |
| Wyoming | 47.1 |
| Washington | 46.5 |
| North Carolina | 45.7 |
| Tennessee | 45.6 |
| Maryland | 44.3 |
| South Carolina | 42.4 |
| Pennsylvania | 42.2 |
| Massachusetts | 40.3 |
| Georgia | 40.0 |
| Nevada | 37.3 |
How common is caller ID spoofing?
Caller ID is easy to fake. The largest illegal robocall operation the FCC has investigated placed more than 5 billion auto-warranty calls in three months of 2021, displaying 1,051,461 different caller IDs. About 75% of the caller IDs in related complaints used the consumer's own area code, a neighboring one, or one in the same state, a practice the FCC calls neighbor spoofing. The FCC fined the operation $299,997,000, its largest robocall penalty.
One spoofing operation, in the FCC's numbers
- 5 billion
Robocalls that operation placed in three months
The FCC says more than five billion.
January to March 2021 · Federal Communications Commission
- 1,051,461
Different caller IDs that one operation displayed
January to March 2021 · Federal Communications Commission
- $299,997,000
The largest robocall penalty the FCC has imposed, against an auto-warranty operation
August 2023 · Federal Communications Commission
This year's complaints show how little a caller ID number is worth. DeepSearch calculated the two findings below from the FCC's public unwanted-call complaint records; each citation includes the calculation, scope, limitation and source so those details travel with the number.
Most displayed caller IDs appear only once
96% of distinct displayed caller ID numbers in the FCC unwanted-call complaint sample appeared in exactly one complaint.
US unwanted-call complaints filed with the FCC that contain a usable displayed caller ID. Filing period: January to October 2026.
Calculation: 52,308 / 54,378 × 100 = 96% (rounded). Group complaints by displayed caller ID, then divide the number of IDs appearing once by all distinct IDs in the sample.
Limitations: This measures recurrence in complaints, not all calls, unique callers or the effectiveness of number blocking. Caller IDs can be spoofed.
January to October 2026
How often do caller IDs recur in FCC complaints?
96% of the 54,378 caller ID numbers in unwanted-call complaints to the FCC from January to October 2026 were reported only once, and the most-reported number drew 42.
Reported once
52,308
Reported 2 to 4 times
1,950
Reported 5 or more times
120
| Category | Caller ID numbers | Share |
|---|---|---|
| Reported once | 52,308 | 96% |
| Reported 2 to 4 times | 1,950 | 3.6% |
| Reported 5 or more times | 120 | 0.2% |
That finding describes reported caller IDs; it does not measure how effective number blocking is. A displayed number can be spoofed, so it should not be treated as proof of who called. Often it is not even the number the call wants you to use:
The number on screen is often not the number to call back
47.6% of FCC unwanted-call complaints that gave both a displayed caller ID number and a callback or promoted number listed two different numbers.
US unwanted-call complaints filed with the FCC that contain both a usable displayed caller ID and a callback or promoted number. Filing period: January to October 2026.
Calculation: 15,489 / 32,547 × 100 = 47.6% (rounded). Compare the displayed caller ID with the callback or promoted number in the same complaint, then divide the complaints where they differ by all complaints giving both.
Limitations: A different callback number is not proof of spoofing: legitimate callers also route callbacks to other numbers. It shows the displayed number alone does not identify who a callback would reach.
The numbers that do recur are short-lived in complaints. Of the 2,070 caller ID numbers reported more than once this year, 54.6% drew their first and last complaints within a week of each other, and the median span between first and last filing was 5 days. Those spans use filing dates, which lag the calls themselves, so they bound how long a number stayed in reported use rather than dating its last call. The pattern is not new: the share reported only once was 95.9% in 2023, 93.9% in 2024 and 96.2% in 2025, counted the same way within each year.
2023 to January to October 2026
Are spam callers reusing numbers more than before?
No: caller IDs reported only once made up between 93.9% and 96.2% of distinct numbers in FCC complaints every year from 2023 to January to October 2026.
2026 YTD
Share of of caller ID numbers reported once
Year the complaint was filed
Select a point to inspect
| Year the complaint was filed | Share of of caller ID numbers reported once |
|---|---|
| 2023 | 95.9% |
| 2024 | 93.9% |
| 2025 | 96.2% |
| 2026 YTD | 96.2% |
How local the fake looks depends on where you live. In Colorado, 68.6% of complaints that gave a caller ID showed a Colorado area code, against 40.6% nationally. Colorado's pattern changed during the year: its complaints stating a caller ID averaged 113 a month in January and February, then 517 a month from March on, while the in-state share rose from 37.6% to 70.3%. The records cannot say whether one campaign began targeting Colorado numbers or more people there started reporting. An in-state number is no reason to pick up.
January to October 2026
Which states see the most in-state caller IDs?
In Colorado, 68.6% of unwanted-call complaints from January to October 2026 that gave a caller ID showed a Colorado area code, the highest share of any state; nationally it was 40.6%.
| Category | Share of complaints with a caller ID |
|---|---|
| Colorado | 68.6% |
| Illinois | 56.3% |
| Texas | 54.1% |
| Iowa | 50% |
| Oklahoma | 48.9% |
| Wisconsin | 46.2% |
| Ohio | 46% |
| Arkansas | 44.2% |
| Minnesota | 41.5% |
| Missouri | 41.5% |
| Oregon | 41.4% |
| Connecticut | 40.9% |
| California | 40.7% |
| Washington | 40.1% |
| Massachusetts | 40% |
Not every reported number pretends to be local. In 16.9% of this year's complaints with a matched caller ID area code, the displayed number was toll-free, and the most-reported area codes after one Colorado code are toll-free codes. A toll-free number names no place and proves nothing about who placed the call.
January to October 2026
Which caller ID area codes draw the most complaints?
Caller IDs with the 303 (Colorado) area code drew 2,303 unwanted-call complaints from January to October 2026, more than any other.
| Category | Complaints |
|---|---|
| 303 · Colorado | 2,303 |
| 833 · toll-free | 2,086 |
| 855 · toll-free | 1,561 |
| 877 · toll-free | 1,516 |
| 866 · toll-free | 1,380 |
| 844 · toll-free | 1,317 |
| 888 · toll-free | 1,284 |
| 202 · District of Columbia | 955 |
| 630 · Illinois | 678 |
| 502 · Kentucky | 619 |
| 509 · Washington | 520 |
| 804 · Virginia | 448 |
| 270 · Kentucky | 403 |
| 720 · Colorado | 384 |
| 470 · Georgia | 371 |
Tracing calls back to their source is slow work. In 2024 the industry's traceback group started 3,606 tracebacks of suspected unlawful robocalls. 69% of completed tracebacks led the originating provider to warn or cut off the caller, down from 84% in 2023.
Enforcement chases a moving target. The traceback group identified 275 new VoIP providers in 2024, after 270 in 2023, and notes that new providers are easy to create. In December 2024 the FCC ordered 2,411 providers to cure deficiencies in their Robocall Mitigation Database filings, the registry every provider carrying voice traffic must file in. DeepSearch counted that live registry: of 11,577 current filings, 42.1% certify complete STIR/SHAKEN implementation, the caller-ID authentication framework, while 36.8% certify none. The FCC prints no official total, one provider can hold more than one filing, and filings change daily. Yet in 2024, the FCC reports, neither it nor the Attorney General collected forfeiture penalties or criminal fines for section 227 violations: $0 collected. Unpaid forfeitures are referred to the Department of Justice.
AI-generated voices are part of this, but they are not separately counted. The best-documented case is political: ahead of the 2024 New Hampshire primary, robocalls carried a deepfake AI voice imitating President Biden and showed a spoofed caller ID. The FCC imposed a $6,000,000 penalty on Steve Kramer; the deepfake voice message had cost $150 to make. None of the sources on this page counts AI voice-cloning scam calls or their losses separately, so any precise total quoted elsewhere does not come from these regulators.
Why these sources disagree
Different definitions, reporting windows and collection methods produce different totals. For example, the FTC's $1.1 billion in losses from fraud that began with a phone call and the FBI's more than $2.9 billion in call-center fraud losses both describe 2025, but the first groups reports by contact method and the second groups them by crime category. Neither is a second estimate of the exact same measure.
Scroll horizontally to compare all columns
Do not add the FTC and FBI totals. IC3 is a contributor to Consumer Sentinel, so the datasets overlap. Adding them would double-count some reports.
Do not equate complaints with calls or victims. A person may receive many calls without filing a complaint, and an unwanted call is not necessarily a fraud attempt. A rise in complaints can reflect changes in reporting as well as changes in unwanted calls.
Match the window before comparing a trend. Calendar years, fiscal years and partial years cover different periods. A fall in estimated robocall volume and a rise in reported fraud losses can coexist: call frequency and reported financial harm measure different things.
How we compiled these statistics
Every figure was read in its source on the retrieval date printed under each chart and table, and each one links to the source list at the end of the page. We used no figure that appears only on a statistics roundup site.
As of mid-October 2026 the FTC had not published its 2025 Data Book, so the 2025 FTC figures come from its congressional testimony, its 2026 releases, and its public Consumer Sentinel dashboard, with data as of 30 June 2026. We read the dashboard through its own year filters. To check that reading, we compared the dashboard's 2024 phone-call figures with the printed 2024 Data Book: 284,659 reports, $948 million in losses and a $1,500 median in print, against 284,613 reports and the same losses and median on the dashboard today. The small difference in reports is the FTC adding late reports.
Figures marked as derived are our arithmetic on published numbers. The CSV names the inputs for each one.
What these numbers can't tell you
- Reports are self-selected. The FTC and FBI count what people choose to report, and most fraud is never reported.
- Contact method means first contact. It records how a scam first reached someone, and many reports do not name one.
- The FTC and FBI datasets overlap. Never add them together.
- Do Not Call data runs on fiscal years, October to September.
- Robocall volumes are YouMail estimates, not counts.
- AI voice cloning is not counted separately by any source on this page, so these figures cannot say how many scam calls use a cloned voice.
- Dashboard figures are revised as late reports arrive, so a figure read later may differ slightly.
Frequently asked questions
Use and cite this data
Every figure on this page in one CSV, one row per value, each with its source, locator and retrieval date. Free to reuse with a link back to this page.
Cite this page
DeepSearch (2026). US phone scam statistics 2026: Losses, robocalls and trends. Data through November 2026. https://deepsearch.app/blog/phone-scam-statistics
Update history
Reordered the page to lead with reported losses, moved the what-to-do guidance to the top, placed each FCC complaint finding beside its chart, named the United States in the title, and removed the metro robocall and state Do Not Call tables. Added DeepSearch analysis of the same FCC snapshot: the prerecorded share of typed complaints, how often the callback number differs from the displayed caller ID, toll-free caller IDs, complaint rates by state, and Colorado's within-year change. Added 2024 robocall enforcement figures and the deepfake AI-voice penalty from the FCC report already cited, re-read on this date. Refreshed the FCC complaint snapshot through October 2026 and re-read the FTC dashboard (data as of 30 June 2026) for a matched first-half comparison; added caller-ID lifespan and year-by-year reuse. Coverage now runs to October 2026 for FCC complaints.
Highlighted three existing FCC complaint analyses with reproducible counts and copyable citations, added a source-comparison table, and clarified the limits of timing and caller-ID findings. No source refresh; coverage and retrieval dates are unchanged.
Clarified the periods and measures in the opening, made headings question-led, and removed an unsupported inference about number-blocking effectiveness. Dataset values are unchanged.
Published. Re-checked every figure against its source. The median phone-call loss for people 60 and over is now the FTC's 2025 figure ($3,000), and the underreporting-adjusted fraud estimate is now 2025's $261.5 billion, both from Protecting Older Consumers 2025-2026 (19 October 2026).
First edition: FTC figures through 2025 (dashboard data as of 30 June 2026), FBI IC3 2025, Do Not Call fiscal 2025, and YouMail estimates through September 2026.
Sources and review
Evidence
Reviewed by DeepSearch Research and Safety Team on .
01
Prepared Statement of the FTC on "The Rising Scam Economy"
Federal Trade Commission · Total 2025 fraud reports and losses, the rise since 2020 and the 4.8 percent reporting rate
02
Consumer Sentinel Network: Fraud Reports (interactive dashboard)
Federal Trade Commission · Fraud reports, losses, median losses and loss rates by contact method, by phone-call scam type, by age, and government-imposter trends
Caveat: Sentinel reports are self-reported and unverified, not a survey. The FTC took no reports from 1 October to 13 November 2025, so 2025 is understated. Dashboard figures are revised as reports arrive.
03
Data Spotlight: Reported losses to scams on social media eight times higher than in 2020
Federal Trade Commission · Social-media losses in 2025, contact methods ranked by loss reports, and phone calls as the costliest contact method for people 80 and over
04
FTC Data Show People Reported Losing $3.5 Billion to Imposter Scams in 2025
Federal Trade Commission · Imposter-scam losses and their share of reports, and government-impersonator losses in 2025
05
Protecting Older Consumers 2025-2026
Federal Trade Commission · The median loss for people 60 and over when fraud began with a phone call, and the 2025 estimate of fraud losses adjusted for underreporting
06
Data Spotlight: False alarm, real scam
Federal Trade Commission · How often a phone call was the first contact in large imposter losses among older adults
07
Consumer Sentinel Network Data Book 2024
Federal Trade Commission · The printed 2024 phone-call figures used to check the dashboard reading
08
National Do Not Call Registry Data Book, Fiscal Year 2025: data files
Federal Trade Commission · Do Not Call complaints by call type, year and topic, and active registrations
Caveat: Complaints are unverified, and not every complainant reports a call type or topic.
09
FTC Releases Annual Do Not Call Registry Data Book
Federal Trade Commission · The comparison of Do Not Call complaints with fiscal 2021
10
FBI Internet Crime Complaint Center · Total complaints and losses, call-center fraud, elder fraud, and losses by crime type
Caveat: IC3 figures are complaints with one crime type each. IC3 also feeds the FTC's Sentinel network, so FBI and FTC totals overlap and must not be added.
11
YouMail Robocall Index: historical volume
YouMail · Nationwide robocall volume by month and year, the daily rate and calls per person
Caveat: A YouMail estimate, extrapolated from calls to its own app's users. It is a vendor model, not a count.
12
Forfeiture Order FCC 23-64 (Sumco Panama et al.)
Federal Communications Commission · The record robocall penalty, the operation's call volume and its use of neighbor spoofing
13
Federal Communications Commission · Complaints alleging illegal caller ID spoofing, industry traceback results, Robocall Mitigation Database enforcement and fines collected, new VoIP providers identified in tracebacks, and the deepfake AI-voice robocall penalty
Caveat: One FCC complaint can count under more than one section of the law.
14
CGB Consumer Complaints Data (dataset 3xyp-aqkj), issue "Unwanted Calls"
Federal Communications Commission · Unwanted-call complaints filed with the FCC each year, and DeepSearch's analysis of this year's complaints: how often a caller ID recurs, when calls come, whether calls are prerecorded, how often a caller ID is in-state or toll-free, and how often the callback number differs from the displayed number
Caveat: A live dataset counted by the date each complaint was filed; counts can change.
15
Robocall Mitigation Database listings
Federal Communications Commission · DeepSearch's count of current Robocall Mitigation Database filings by certified STIR/SHAKEN implementation status
Caveat: A live registry; filings are added, recertified and removed. One provider can hold more than one filing, and the FCC prints no official total.
Re-check trigger: The FTC's 2025 Data Book or a new quarter of Sentinel dashboard data, the next FBI IC3 or Do Not Call data book, or a quarter of new robocall index data.