Invalid traffic (IVT) is any ad impression or click that did not come from a real, interested human: crawlers, bots, hijacked phones, spoofed apps, fake TVs. The IVT Desk follows the fight against it every day and explains how it works, how it is caught, and how to keep it out of your budget.
Every impression starts as a visit from a device. Some are people; some are crawlers, scripts or hijacked phones pretending to be people.
General invalid traffic is traffic that routine filters identify from lists and simple rules: declared search-engine crawlers, known data-center IP ranges, non-browser user agents, pre-fetch and monitoring tools. The Media Rating Council expects every measurement vendor to filter it.
Sophisticated invalid traffic is designed to look human and needs advanced analytics to find: bots that mimic browsing, hijacked devices, residential proxies, domain and app spoofing, SSAI spoofing, ad stacking and incentivised manipulation.
Advertisers in South Africa are reportedly losing as much as R12 billion each year due to ad fraud. This highlights the significant financial impact of fraudulent activity in the country's digital advertising market.
TikTok has initiated measures to remove and restrict scam ads that promise users payment for simple tasks or participation. The platform is targeting deceptive advertising practices to protect its users from fraudulent schemes.
DataBeat analysis shows that resellers added 44,600 new ads.txt lines, compared to 10,300 for direct sellers, across the top 50,000 US publishers. Low-traffic sites contributed most of the growth, and nearly half of Tier 1 SSP domains…
Instances of invalid traffic and click fraud have been detected affecting ads related to ChatGPT. This development highlights ongoing challenges in ensuring ad quality and combating fraudulent activity in digital advertising.
Artificial intelligence is increasingly used to improve digital advertising efficiency, but fraudsters are also leveraging AI to create more sophisticated ad fraud schemes. This dual use of AI presents new challenges for the ad tech…
The article discusses the prevalence of ad fraud in the Middle East and North Africa (MENA) region, focusing on both web and mobile app environments. It highlights how fraudulent activities impact businesses by wasting advertising budgets…
Meta is facing a lawsuit related to scam ads appearing on its platforms. The legal action alleges that Meta did not do enough to prevent fraudulent advertisements from reaching users.
Reports indicate a notable rise in ad click fraud, where fake clicks are generated to drain advertiser budgets or manipulate campaign results. This trend is impacting the effectiveness and trustworthiness of digital advertising campaigns.
Cloudflare's CEO stated that automated bot traffic surpassed human web traffic in May 2026, earlier than anticipated. He projects that within five years, bot traffic could be 1,000 times greater than human traffic, raising concerns about…
Real fraud operations, what they did and what they taught the industry.
Residential proxy network built from consumer devices
MIXEDGoogle Threat Intelligence, with Lumen Black Lotus LabsFirmware-level backdoor on Android devices used heavily for ad fraud
MOBILEKasperskyUtility apps that pushed secondary apps running hidden ad traffic
MOBILEHUMAN Satori Threat IntelligenceAndroid apps that turned malicious after approval and flooded phones with ads
MOBILEIntegral Ad Science Threat Lab and BitdefenderBackdoored uncertified Android devices (TV boxes, projectors, car systems) used for ad fraud and proxies
MIXEDHUMAN Satori with Google, Trend Micro, Shadowserver and others"Evil twin" apps on third-party stores mimicking Google Play apps
MOBILEIntegral Ad Science Threat LabAn ad fraud ring earns money every time an ad is “shown”. It doesn’t need people, just traffic that looks like people.
Invalid traffic (IVT) is any ad impression, click or visit that was not generated by a real person with a genuine chance of seeing the ad. It includes harmless automation such as search-engine crawlers as well as deliberate fraud like bots, hidden ads and spoofed websites. Advertisers should not pay for it, so the industry filters and measures it. More
GIVT (general invalid traffic) is non-human traffic that can be caught with simple lists and rules, such as known crawlers, data-center IP addresses and self-declared bots. SIVT (sophisticated invalid traffic) is traffic built to look human, such as hijacked devices, residential-proxy bots, spoofed apps and hidden ads, and it needs advanced analytics to detect. More
Nobody knows precisely; published estimates range from a few billion to more than $80 billion a year depending on methodology. Juniper Research estimated $84 billion (22% of online ad spend) was lost in 2023, rising to $172 billion by 2028. Earlier ANA and White Ops studies put losses nearer $6 billion. All of these are modelled estimates, not audited totals. More
Look for numbers that do not behave like people: very high click-through rates with near-zero conversions, sessions lasting a second or two, traffic spikes at odd hours, clusters of identical devices or browsers, and visits from data centers or unexpected countries. The reliable answer comes from an independent IVT measurement or verification tool rather than from the seller's own reports. More
Click fraud is the generation of fake or worthless clicks on pay-per-click ads, by bots, click farms, malware or competitors, so that someone profits or an advertiser's budget is drained. You reduce it by monitoring click patterns, excluding suspicious IP ranges, placements and apps, tightening targeting, using click-fraud detection tools and claiming refunds for invalid clicks. More
Ad fraud is the deliberate faking of advertising activity, such as impressions, clicks, installs, leads or video views, so that someone gets paid for ads no real person saw or acted on. It is carried out with bots, malware, spoofed websites and apps, hidden ads and manipulated attribution, and it drains budgets while corrupting campaign data. More
Roughly half, by some measures. Imperva's 2025 Bad Bot Report found automated traffic made up 51% of all web traffic in 2024, overtaking human traffic for the first time, with "bad bots" alone at 37%. That is all web requests, not ads specifically; the invalid share of measured ad impressions is usually far lower because much of it is filtered. More
Yes, deliberate ad fraud can be prosecuted as a crime. In the United States, operators have been convicted of wire fraud, computer intrusion and money laundering; the Methbot and 3ve ringleader Aleksandr Zhukov was sentenced to ten years in prison in 2021. Many cases are also pursued through civil lawsuits, platform bans and contract clawbacks rather than criminal charges. More
ClearTrust, the company behind this desk, scores traffic quality with the TQI Score™: 150+ filters for bots, fake clicks and fake impressions, rolled into one number your team can act on.