Best IVT detection 2026
22 min read
DoubleVerify missed the same bots 21 percent of the time. DoubleVerify's stock dropped roughly 70 percent.
Simul Sarker
Founder & Product Designer of DataCops
Last Updated
May 27, 2026
In March 2025, a 240-page Adalytics report covered by the Wall Street Journal found that Integral Ad Science labeled known URLScan.io bot traffic as human 77 percent of the time across a 2019 to 2024 dataset. DoubleVerify missed the same bots 21 percent of the time. DoubleVerify's stock dropped roughly 70 percent. A securities class action followed in July 2025.
The foundational assumption of the IVT detection category collapsed. MRC accreditation does not guarantee bots get blocked.
Meanwhile the fraud problem got worse. Fraudlogix measured global IVT at 20.64 percent across 105.7 billion impressions in 2026. Pixalate's Q1 2026 benchmarks showed 20 percent IVT on web, 39 percent on mobile app, 25 percent on CTV. Juniper Research projects $100 billion in global ad-fraud losses for 2026, scaling to $133 billion by 2028, driven by AI botnets and autonomous agents. Made-for-advertising sites now represent roughly 21 percent of programmatic impressions, many buried inside Performance Max where the buyer cannot inspect placement.
Most comparison lists respond to this by ranking IVT tools by feature count. The useful question is different: which layer does this tool actually cover? Pre-bid, click-time, and conversion-time IVT are three different problems requiring three different vendors. Most buyers solve the first or second and think they are done. The damage from the third is the one that compounds inside Smart Bidding and Meta's Andromeda algorithm.
I cover 25 tools in this guide, tagged by layer, with honest takes on what each one actually catches. Including where DataCops fits and where it does not.
Quick answers
What is IVT in advertising?
Invalid Traffic. The MRC defines two types. GIVT (General Invalid Traffic) covers declared bots, datacenter IPs, and non-human signatures you can identify from lists. SIVT (Sophisticated Invalid Traffic) covers residential proxies, human-mimicking automation, hijacked devices, and in 2026, autonomous AI agents. GIVT is what most pre-bid filters catch. SIVT is what passes through them and trains your bid algorithms on bad signals.
What is the difference between GIVT and SIVT?
GIVT is easy: known datacenter ranges, declared crawlers, hosting company IPs. Desktop shows 27.03 percent IVT versus mobile's 19.30 percent per Fraudlogix 2026, largely because datacenter traffic concentrates on desktop. Internet Explorer shows 79.5 percent IVT and Edge 41.2 percent per Fraudlogix 2026 because bot operators favor legacy browsers. SIVT is the hard one: malware-based bot networks, click farms using residential proxies, autonomous AI agents submitting lead forms with stolen PII. SIVT passes through pre-bid filters and infects your conversion training data.
What is an acceptable IVT rate?
The MRC considers 10 percent de minimis for impression counting. In practice, Fraudlogix puts global web IVT at 20.64 percent. Lunio's 2026 report on 2.7 billion clicks found gaming at 18.49 percent, education at 14.41 percent, telecom at 14.26 percent, real estate at 13.61 percent. Finance and legal verticals exceed 42 percent per Fraudlogix. For conversion-layer IVT, no industry standard exists because most buyers do not measure it.
Is Google's IVT detection enough?
No. TrafficGuard's 2026 analysis estimates up to 25 percent of Performance Max budget is wasted on click fraud that Google's own systems do not intercept. Google does not publish its detection methodology. The entire reason third-party IVT vendors exist is accountability against the platform itself.
Can IVT be detected in real time?
Yes, at click-time and conversion-time. Pre-bid tools operate at auction milliseconds. Click-time tools fire on the click event. Conversion-time tools validate before the event reaches a CAPI endpoint. The 2025 controversy is that pre-bid real-time filters from accredited vendors were still missing declared bot traffic at scale.
What is the best IVT detection tool?
Depends on the layer. For pre-bid programmatic blocking, HUMAN Security is the current category leader after the Adalytics findings damaged DV and IAS credibility. For click-time paid search protection, Lunio or TrafficGuard. For conversion-time filtering that protects what Smart Bidding and Andromeda actually train on, DataCops is the only tool in this review that addresses the CAPI layer directly. No single vendor covers all three layers.
How does MRC accredit IVT vendors?
Independent audit verifying the vendor's methodology against MRC IVT Detection and Filtration Guidelines. The process checks whether the vendor measures what it claims and applies consistent filtration. It does not test absolute detection rates against undisclosed bot populations. That is the core gap the Adalytics report exposed. HUMAN, DoubleVerify, IAS, and Pixalate carry MRC accreditation. MRC accreditation is a process audit, not a detection-outcome guarantee.
The three layers. Pick the wrong one and you are still losing.
Think of IVT detection the way you would think about water filtration. Pre-bid filters sit at the reservoir: they try to stop bots from even getting into the programmatic auction. Click-time filters sit at the pipe: they watch what gets in and try to block the bad clicks from costing you money. Conversion-time filters sit at the tap: they validate events before they reach the algorithms that decide how to spend your budget tomorrow.
Most buyers buy a reservoir filter and call themselves protected. The water still arrives. Some of it is still contaminated. The algorithm drinks it.
Here is the specific 2026 problem.
Smart Bidding and Meta's Andromeda system are trained on the CAPI signal you send. A bot that passes pre-bid filtering, clicks a search ad, fills a lead form, and generates a conversion event teaches Smart Bidding exactly what your ideal customer looks like. It learned from real data. The data happened to come from a bot. Every subsequent campaign cycle bids harder toward that pattern. The click-time blocker cannot help you here because it only stopped the clicks. The conversion already fired.
PillarlabAI documented this in a honeypot. 3,000 form submissions. 77 percent fraudulent. 650 accounts from one device fingerprint. One machine wearing 650 different identities. If those sessions came in through paid search, Smart Bidding would have logged 650 quality conversions and optimized toward that fingerprint for the next 90 days. Not one click blocker in the market would have stopped it. The clicks were valid. The attack was at the conversion layer.
That is the gap pre-bid and click-time tools cannot reach. The tools below are tagged by layer so you can see exactly which problem each one solves.
Layer 1: Pre-bid IVT detection
These tools operate inside the programmatic auction. They attempt to stop bot impressions before the ad serves. This is the layer most enterprise advertisers think of when they say "IVT vendor."
HUMAN Security
The current category leader after the Adalytics fallout. MRC-accredited for pre-bid IVT and SIVT detection. Launched AgenticTrust in 2026, specifically to detect OpenAI Atlas, Claude for Chrome, and AWS AgentCore traffic as the agentic AI bot problem became mainstream. G2 Winter 2026 Bot Detection leader. Covers the complete fraud surface from impression through website visitor, form fill, and account creation.
What does not work: enterprise pricing and implementation complexity. The AgenticTrust framework is new and the agentic AI detection methodology is still maturing. The Adalytics report implicated HUMAN alongside DV and IAS in missing declared bots in programmatic contexts.
Right for: enterprise advertisers running programmatic campaigns who need the most credible MRC-accredited SIVT detection with agentic AI coverage.
Value for money: 8/10 for enterprise programmatic.
Pricing: Enterprise custom.
DoubleVerify
Dominant pre-bid programmatic verification vendor. MRC-accredited across IVT, viewability, and brand safety. Analyzes billions of impressions daily. The Adalytics March 2025 report found DV missing declared bots 21 percent of the time. DV disputes the methodology, arguing URLScan.io was not a declared bot. Securities class action filed July 2025. Stock dropped roughly 70 percent from $39.24 to $13.90 across the preceding year.
What does not work: the Adalytics controversy has not been fully resolved. Credibility damage among sophisticated buyers who read the AdExchanger coverage. No protection at the click or conversion layer.
Right for: programmatic advertisers where DSP/SSP pre-bid integrations require DV specifically, and where impression-level waste reduction is the primary goal.
Value for money: 7/10, down from its pre-2025 position.
Pricing: Enterprise custom via programmatic platform integrations.
Integral Ad Science (IAS)
MRC-accredited pre-bid SIVT detection. The Adalytics report found IAS publisher pixel mislabeling URLScan.io bot traffic as human 77 percent of the time. A former IAS engineer alleged internally that detection code ran on only 50 percent of impressions. IAS disputes both claims.
What does not work: the credibility hit from the Adalytics findings is larger for IAS than for DV given the 77 percent miss rate in the report. No click-time or conversion-time protection.
Right for: programmatic buyers whose DSP requires IAS specifically, and enterprise brand safety programs where IAS's contextual tools are the primary value.
Value for money: 6.5/10 given reputational headwinds.
Pricing: Enterprise custom.
Pixalate
MRC-accredited for SIVT detection on desktop and mobile web. The strongest quarterly benchmarking reports in the category. Q1 2026 benchmarks: 39 percent mobile-app IVT, 25 percent CTV, 20 percent web across 82 billion impressions. Purpose-built for CTV and mobile app IVT, which DV and IAS cover less thoroughly.
What does not work: primarily a measurement and analytics tool rather than a real-time blocker for paid search. No click or conversion layer coverage.
Right for: CTV and mobile app advertisers who need rigorous IVT measurement and publisher-side supply validation. The best source for benchmarking data.
Value for money: 8/10 for CTV and mobile.
Pricing: Custom.
Fraudlogix
The source behind most industry IVT statistics including the 20.64 percent global figure. Pre-bid and post-bid GIVT and SIVT data provider. Integrates directly into DSPs, SSPs, and ad tech infrastructure.
What does not work: not a self-serve tool for individual advertisers. API integration and technical setup required. Not the right choice for a brand wanting to protect a single Google Ads account.
Right for: ad tech infrastructure teams, DSPs, SSPs, and large publishers who need IVT data at the infrastructure level.
Value for money: not ratable on the SMB scale. Custom API pricing.
GeoEdge
Ad quality and malvertising protection. Monitors for malicious ads, forced redirects, auto-downloads, and phishing units in publisher inventory. Operationally adjacent to IVT but focused on ad quality rather than audience validity.
What does not work: not an IVT detection tool in the traditional GIVT/SIVT sense. Does not protect the conversion layer. Limited to ad quality use cases.
Right for: publishers who need malvertising protection alongside ad quality enforcement.
Pricing: Custom.
Moat (Oracle)
Ad measurement and verification platform acquired by Oracle in 2017. MRC-accredited viewability and IVT measurement. Primarily serves publishers and enterprise brands for attention measurement alongside fraud filtering.
What does not work: Oracle's restructuring has created product uncertainty around Moat's roadmap. Less active development than HUMAN or DV.
Right for: enterprise measurement programs where Moat is already embedded in existing Oracle contracts.
Pricing: Enterprise custom.
Layer 2: Click-time IVT detection
These tools operate after the auction clears. They watch the click event and attempt to stop invalid traffic from consuming budget or poisoning session data. This is the right layer for paid search fraud on Google Ads, Meta Ads, and Microsoft Ads.
DataCops
DataCops operates at Layer 3, the conversion layer, not Layer 2. It does not add IPs to Google Ads exclusion lists. It does not watch the click event. For budget drain from invalid clicks, you still need a dedicated click-time blocker. But no tool in Layer 2 can reach what DataCops addresses.
Full DataCops coverage is in the Layer 3 section below.
Lunio
13+ ad platform coverage. 2,000+ behavioral tests per visit. Self-learning AI. Wasted Ad Spend reporting shows exactly which campaigns are consuming fraudulent budget. Originally PPC Protect, rebranded Lunio in 2022. Strong EU market presence.
What does not work: pricing behind sales calls, no transparent self-serve tiers. Most effective above $10,000/month ad spend. No conversion-layer protection.
Right for: larger multi-channel advertisers wanting unified invalid traffic intelligence with clear spend waste reporting.
Value for money: 7.5/10 at target spend.
Pricing: Custom.
TrafficGuard
The only click-time tool in this comparison built specifically for Performance Max placement fraud. Real-time invalid traffic detection with P-Max placement analysis that other click blockers cannot reach because they cannot access Google's internally-controlled placements. MMP integrations for mobile app install fraud. Free monitoring tier up to $2,500/month ad spend.
What does not work: 2 percent of ad spend pricing gets expensive above $50,000/month. No conversion-layer protection.
Right for: Performance Max advertisers who want placement-level fraud analysis inside Google's automated campaign type, and mobile app advertisers needing install fraud protection.
Value for money: 8/10 for P-Max heavy accounts.
Pricing: 2% of ad spend/month. Free monitoring tier available.
CHEQ (formerly ClickCease)
ClickCease rebranded as CHEQ Essentials after CHEQ acquired it. 14,000+ SMB customers. 2,000+ behavioral tests per click. Google and Meta API approved. Pixel Guard Connector prevents pixels firing on invalid users, which reduces Smart Bidding contamination from click-level fraud. CHEQ at enterprise covers the full funnel from click through form fill.
What does not work: annual billing not prominently disclosed on signup. January 2026 Trustpilot reviewers describe discovering 12-month lock-in at cancellation. Microsoft Ads exclusion requires manual CSV upload. Google's 500 IP exclusion limit per campaign becomes a management burden after 6-12 months.
Right for: SMBs on Google and Meta wanting the most-tested click blocker with competitor monitoring.
Value for money: 7.5/10
Pricing: CHEQ Essentials from $99/month. Annual commitment.
ClickGUARD
Deep per-campaign rule customization, 50+ configurable protection features. Rebranded in September 2025 as ClickGUARD 2.0. Full forensic visibility into every blocking decision rather than automated defaults.
What does not work: 30-45 minutes to configure properly. Requires a PPC specialist. Narrower platform coverage than Spider AF or Lunio.
Right for: in-house PPC teams who want full control over exclusion logic and forensic transparency.
Value for money: 7.5/10
Pricing: From $74/month.
ClickPatrol
Four modules: ads, audiences, data, and forms. FormProtector addresses junk leads and CRM spam. EU-native architecture. Native automated blocking on Google, Meta, and Microsoft Ads. 800+ detection signals per visit.
What does not work: no free trial. Less established outside EU markets.
Right for: EU-based lead generation advertisers running forms-based PPC who want click fraud and lead form poisoning addressed in one tool.
Value for money: 8/10 for EU lead gen.
Pricing: From €59/month.
Fraud Blocker
Bootstrapped, transparent, no annual contract. 100+ signals per visitor. Automated IP exclusion. Month-to-month from day one.
What does not work: small team raises questions about feature velocity. Meta protection requires higher tier. No competitor monitoring. No conversion layer.
Right for: small businesses wanting click protection at transparent pricing without annual commitment.
Value for money: 8/10 for the honest pricing.
Pricing: From $69/month.
Spider AF
30+ ad networks from one dashboard. CRM-level fake lead filtering. Published white paper: valid-click CVR runs 2x invalid-click CVR. Documented case: removing fake leads and MFA placements produced 152% ROI increase and 85% CPC reduction.
What does not work: more expensive than SMB-tier click blockers. Overkill for single-channel Google Ads accounts.
Right for: multi-platform advertisers across 5+ ad networks needing single-pane fraud intelligence with CRM-level lead validation.
Value for money: 7.5/10
Pricing: From $150/month ($120/month annual).
Anura
Ad fraud detection tool focused on the intersection of click fraud and conversion fraud. Real-time session validation with a strong focus on lead generation quality. Transparent accuracy marketing.
What does not work: smaller brand than CHEQ or Lunio. Less deep P-Max integration.
Right for: lead gen advertisers who want session-level fraud validation without the enterprise complexity of CHEQ at its top tier.
Value for money: 7.5/10
Pricing: Custom.
Clixtell
Session recording plus call tracking alongside click fraud protection. The only tool in this comparison connecting phone call conversions back to the paid click that drove them. 2026 ad fraud statistics report is one of the more credible public benchmarks in the category.
What does not work: session recording tools that store interaction data require careful GDPR handling. Less comprehensive network coverage than Spider AF or Lunio.
Right for: service businesses running Google Ads where phone call attribution matters and you want session-level visibility.
Value for money: 7/10
Pricing: Contact for current pricing.
Fraud0
Munich-based, privacy-first. 4,000+ data points per visitor, no cookies or PII stored. 15,000+ customers. Dr. Augustine Fou serves as advisor.
What does not work: starts in detection-only mode, blocking requires additional configuration. Thinner review base than ClickCease.
Right for: EU advertisers who need cookie-free, PII-free fraud detection that passes GDPR legal review.
Value for money: 7.5/10
Pricing: From €50/month.
Singular
Mobile attribution platform with IVT filtering for app install campaigns. SDK-level detection. Strong on mobile fraud patterns including click injection and click flooding.
What does not work: web and search fraud is not the core use case. Limited click-time protection for desktop paid search.
Right for: mobile app advertisers running UA campaigns where install fraud is the primary exposure.
Pricing: Custom.
DataDome
Bot management and fraud protection at the infrastructure layer. Covers web scraping, credential stuffing, account takeover, and ad fraud in one platform. Used primarily by e-commerce and media.
What does not work: enterprise-grade complexity and pricing. Primarily a security tool, not an ad fraud tool. Limited paid search click exclusion.
Right for: enterprise e-commerce brands where bot management, account fraud, and ad traffic quality all need to be addressed from one vendor.
Pricing: Enterprise custom.
Kasada, Imperva, Shape Security (F5)
Bot management platforms operating at the infrastructure layer, not the ad fraud layer specifically. Kasada, Imperva, and Shape Security (now part of F5) protect web applications from automated attacks including credential stuffing and scraping. They detect bots before they reach your application.
What does not work for ad fraud specifically: they do not connect to Google Ads exclusion lists. They do not file Invalid Activity reports. They do not protect the CAPI signal layer. They protect your site from bot activity, which is adjacent to but not the same as protecting your ad campaign from IVT.
Right for: enterprise engineering teams who need comprehensive bot management beyond the ad campaign scope.
Pricing: Enterprise custom.
Forensiq (Impact)
Ad fraud detection acquired by Impact. Focused on affiliate and partner marketing fraud, including cookie stuffing, click hijacking, and fake lead attribution.
What does not work: primarily for affiliate and partner channel fraud, not paid search or programmatic.
Right for: brands running large affiliate programs where commission fraud is the exposure.
Pricing: Custom via Impact platform.
Adverity
Marketing analytics and data integration platform with IVT monitoring built into reporting pipelines. Not a real-time blocker but surfaces IVT patterns in marketing data.
What does not work: an analytics tool, not a blocking tool. Cannot stop budget from being consumed by invalid clicks in real time.
Right for: data teams who want IVT visibility inside their marketing data warehouse rather than a standalone blocking product.
Pricing: Custom.
Layer 3: Conversion-time IVT detection
This is the layer almost nobody is selling and the one that matters most for algorithm protection.
When Meta's Andromeda or Google's Smart Bidding receives a conversion event via CAPI, it treats that event as a real customer signal. It adjusts audience targeting, bidding strategy, and creative delivery based on that signal. If the event came from a bot, the algorithm learns bot behavior as success behavior. Pre-bid filters did not stop this. Click-time filters did not stop this. The bot converted.
One anonymous brand executive told AdExchanger: "None of them work, so we're just buying vaporware to make ourselves feel better." That quote was about pre-bid vendors. It captures the frustration of buyers who spent heavily on Layer 1 protection and still watched their Smart Bidding models degrade.
The conversion layer is different from the click layer in one critical way: the event is a training signal. A blocked click wastes a dollar. A bot conversion that reaches CAPI teaches the algorithm to spend more dollars on traffic that resembles that bot. The compounding cost over a 90-day bidding cycle is not linear.
DataCops
DataCops is the only tool in this review that operates specifically at Layer 3, filtering IVT before events reach the conversion API.
One script tag, one CNAME record, live in 5-30 minutes. Three detection layers run before any event is counted or forwarded to Meta CAPI, Google Ads Enhanced Conversions, TikTok Events API, or LinkedIn Insight CAPI. IP intelligence against 361B+ network ranges (146.4B datacenter, 202B residential/mobile, 11.9B VPN, 620M proxy/anonymizer, 160K fraud email domains). Browser and device fingerprinting across 50+ signals including Puppeteer, Selenium, and Playwright headless detection. Email intelligence at the form layer against 160K+ fraud email domains.
SignUp Cops extends filtering to the form layer: disposable email detection, multi-account device fingerprinting, email risk scoring. An agentic bot submitting stolen PII gets caught before the conversion event fires. A TCF 2.2 first-party CMP is bundled.
What does not work: DataCops does not add IPs to Google Ads exclusion lists. No session recordings. No competitor click monitoring. No refund claim filing. No P-Max placement analysis. For budget drain from invalid clicks you still need a Layer 2 click blocker. SOC 2 Type II in progress.
Right for: performance marketers who want the conversion signal reaching Smart Bidding and Andromeda to reflect real human intent rather than what passed through the click-time blocker.
Value for money: 9/10 for conversion signal quality.
Pricing: Free Basic (2,000 sessions/month, unlimited bot detection, first-party analytics, 500 signup verifications, free CMP, no CAPI). Growth $7.99/month. Business $49/month: CAPI starts here, 50,000 sessions, all four platforms. Organization $299/month. Enterprise custom.
Feature comparison by IVT layer
| Tool | Layer 1 pre-bid | Layer 2 click-time | Layer 3 conversion | MRC accredited | Agentic AI | Entry price |
|---|---|---|---|---|---|---|
| DataCops | No | No | Yes | No | Yes | $49/mo |
| HUMAN Security | Yes | No | No | Yes | Yes | Enterprise |
| DoubleVerify | Yes | No | No | Yes | No | Enterprise |
| IAS | Yes | No | No | Yes | No | Enterprise |
| Pixalate | Yes | No | No | Yes | No | Custom |
| Fraudlogix | Yes (data only) | No | No | No | No | Custom API |
| Lunio | No | Yes | No | No | No | Custom |
| TrafficGuard | No | Yes | No | No | No | 2% spend |
| CHEQ/ClickCease | No | Yes | No | No | No | $99/mo |
| ClickGUARD | No | Yes | No | No | No | $74/mo |
| ClickPatrol | No | Yes | No | No | No | €59/mo |
| Fraud Blocker | No | Yes | No | No | No | $69/mo |
| Spider AF | No | Yes | Partial (CRM) | No | No | $150/mo |
| Anura | No | Yes | No | No | No | Custom |
| Fraud0 | No | Yes | No | No | No | €50/mo |
| Clixtell | No | Yes | No | No | No | Contact |
| DataDome | No | Partial | No | No | No | Enterprise |
| Singular | No | Mobile only | No | No | No | Custom |
| GeoEdge | No | No | No | No | No | Custom |
| Kasada/Imperva/Shape | No | No | No | No | No | Enterprise |
No tool carries coverage across all three layers. Layer 3 has one serious entrant for performance marketers.
Decision matrix by buyer type
Programmatic DSP or SSP operator, impression-level IVT accountability: HUMAN Security. MRC-accredited, agentic AI coverage, strongest credibility after Adalytics fallout.
Enterprise programmatic advertiser, existing DV or IAS contract: Evaluate the Adalytics findings against your own impression data before renewal. Neither vendor's credibility is what it was in 2024.
CTV or mobile app advertiser needing IVT benchmarks: Pixalate. Purpose-built for these surfaces, MRC-accredited, strongest quarterly benchmark reports.
Google Ads or Meta paid search, click budget drain: ClickPatrol at €59/month for EU lead gen. TrafficGuard for P-Max heavy accounts. Fraud Blocker at $69/month for SMB without annual commitment.
B2B lead gen, want Smart Bidding protected from agentic AI form fills: DataCops Business at $49/month alongside any click-time blocker. Covers the layer no click blocker reaches.
Multi-channel, 5+ ad networks, CRM-level lead validation: Spider AF at $150/month.
Enterprise full-funnel fraud, one contract: CHEQ at enterprise pricing.
When DataCops is not the answer here
For Layer 1 pre-bid programmatic fraud, DataCops does not operate in the programmatic auction. HUMAN, DV, IAS, and Pixalate are the right vendors.
For click-time budget drain, DataCops does not add IPs to Google Ads exclusion lists. ClickCease, Fraud Blocker, ClickGUARD, and TrafficGuard do.
For session recordings to evidence fraud to finance or dispute refunds with Google, DataCops does not produce those. ClickCease, ClickGUARD, CHEQ, and Clixtell do.
For Performance Max placement-level analysis inside Google's automated campaign type, TrafficGuard is the only tool built for that layer. DataCops is not.
If SOC 2 Type II certification is required today from every vendor, DataCops is still completing it.
The Adalytics report shook the pre-bid category. The securities lawsuit against DoubleVerify confirmed the damage. Buyers spent millions on MRC-accredited pre-bid vendors and watched declared bots get through anyway.
The pre-bid layer got the scrutiny. The conversion layer has not gotten any scrutiny at all, because almost nobody is measuring it.
Your Smart Bidding model trained for 90 days on every CAPI event your stack sent. How many of those events came from real humans with genuine intent to buy, and how many came from traffic that passed every pre-bid and click-time filter on its way to teaching your algorithm the wrong lesson?
That number exists in your Events Manager and your conversion logs. Most buyers have never pulled it.
