The Conversion Data Mirage: What Your Android App Setup is Really Missing

The CAPI race ended in April 2026. Meta launched free one-click Conversions API on April 15. Google Tag Gateway went live in January. Between those two releases, the floor dropped to zero for the two biggest ad platforms on earth. Any tool in this category whose entire value proposition was "we connect your server to Meta" is now competing with a free button inside Events Manager.

That changes what this article needs to be. This is not a ranking of who built the best pipe. The pipe is free. The question that matters now is what you are sending through it.

I have tested or evaluated more than 25 CAPI and server-side tracking tools since iOS 14.5 broke attribution in 2021. The finding that nobody publishing comparison articles in 2026 is willing to say plainly: most CAPI implementations are pumping bot-contaminated events straight into Meta's training data. The platform receives them, labels them as real conversions, builds Lookalike Audiences from them, and charges you to find more people who look like your bots. Project Andromeda, fully deployed in October 2025, acts on contaminated signals within hours, not weeks. Your next campaign launches against an algorithm that has already learned your bot traffic as a conversion signal.

Fixing the pixel-to-server handoff was a solvable engineering problem and in 2026 it is essentially solved. Fixing what goes through that handoff is a different problem. Most tools have not touched it.

What "best CAPI tool" actually means in 2026

For three years the market sorted on one axis: does it send server-side events or not. That axis is dead. Every tool listed in this article sends server-side events. Meta and Google both do it free.

The axes that matter now are: what do you filter before you send, what consent infrastructure sits in front of the send, and how many platforms does a single stack handle.

A quick scan of what is actually different between tools in the category reveals four distinct architectures. First, free native tools from Meta and Google that handle single-platform CAPI with no filtering or consent layer. Second, infrastructure platforms like Stape and Addingwell that host your server-side GTM container and let you build whatever you want on top. Third, managed no-code platforms like Tracklution and SignalBridge that get you live in 15 minutes and handle multi-platform routing. Fourth, bundled stacks that add bot filtering and consent management into the pipeline itself before the event fires.

The buyer decision depends entirely on which architecture matches what they actually need.

Quick answers

Is Meta's free CAPI good enough for most stores? For a single-store Shopify merchant running only Meta ads, the free one-click CAPI launched April 15, 2026 handles the basic pixel-to-server bridge. Setup takes four minutes. It does not filter bots, does not handle Google, TikTok, or LinkedIn, and has no consent management. If those limitations apply to your situation, you need a paid tool. If they do not, start with free.

Does server-side tracking stop ad blockers? Partially. Server-side processing means events no longer depend entirely on the browser executing a third-party script. But the browser still needs to fire the initial first-party hit that triggers server processing. A fully blocked session produces no server event. First-party CNAME setup on your own subdomain recovers 30-40% of those sessions because uBlock and Brave block third-party CDNs by name, not first-party domains. Without a CNAME, server-side alone does not recover blocked traffic.

What is EMQ and why does it matter? Event Match Quality is Meta's score for how well your CAPI events match real user profiles. EMQ 8.6 to 9.3 produces 18% lower CPA and 22% ROAS lift based on Meta's own published benchmarks via AdExchanger. Most tools can get you to EMQ 7-8. Getting to 9+ requires enriched first-party data: hashed email, phone, external ID, and user agent all passing simultaneously. Bot events structurally pull EMQ down because bots do not carry accurate PII.

Do I need a CMP to run CAPI legally in the EU? Yes, and this is the part most comparison articles skip. Google Ads Consent Mode v2 became mandatory for all EEA advertisers on June 15, 2026. You cannot run Google CAPI in the EU without a compliant consent management platform signaling consent status to Google. Meta has similar requirements under GDPR. The CMP is not optional; it is the legal gate for the entire CAPI pipeline.

What is the real cost difference between DIY server-side GTM and a managed tool? A self-hosted sGTM setup typically runs $5,000-10,000 in initial developer time plus $50-300 per month in Google Cloud Run or equivalent hosting. First-year total cost of ownership sits between $11,880 and $36,600 before any add-on tools for consent, bot filtering, or attribution. Managed tools start at $17/month. The math only favors DIY if you have dedicated in-house GTM engineers and a reason to need full container control.

How much conversion data is actually bot traffic? Global invalid traffic rate hit 20.64% in 2026 per Fraudlogix data. Meta's average IVT sits at 8.20%, Instagram at 38%, Audience Network at 67%. Finance and legal verticals run 42% bot rate. For a store with 10,000 monthly CAPI events, that is roughly 820 to 4,100 bot conversions landing in Meta's training data every month, depending on your traffic mix.

Which CAPI tools actually filter bots before sending to Meta? As of mid-2026: DataCops (361B+ IP database), SignalBridge (basic bot filtering), and Cometly (fraud detection on attribution layer). Most tools, including Stape, Tracklution, Elevar, Addingwell, and TAGGRS, do not filter at the CAPI level. They forward what the browser sends.

Who should use what: decision matrix

Shopify, under $50K GMV/month. Use Meta's free one-click CAPI for Meta. Use Google Tag Gateway for Google. You are not large enough for the economics of bot filtering to matter on a per-dollar basis. Upgrade when monthly ad spend exceeds $3,000-5,000.

Shopify, $50K-500K GMV/month, Meta primary channel. TrackBee ($79/month) or SignalBridge ($29/month) handle Meta and Google with minimal setup. If bot traffic is measurably hurting your Meta EMQ or you are seeing attribution anomalies, DataCops at $49/month adds filtering before the event fires.

Shopify, $500K+ GMV/month, order-level fidelity matters. Elevar ($200-950/month) has five years of Shopify-native checkout hooks that nothing else matches. Shop Pay, Apple Pay, Checkout Extensibility, App Pixel depth. If you can absorb the price, the signal quality at checkout level is the best available. If the bill is growing past $450/month and you do not need Shopify-exclusive depth, the alternatives below cover most use cases at lower cost.

Multi-platform (WooCommerce, Webflow, headless, B2B SaaS). Elevar exits this conversation. Tracklution ($31/month), SignalBridge ($29/month), DataCops ($49/month), or Stape ($17/month plus Cloud Run) depending on whether you want no-code simplicity or full container control.

Agency, 10+ client accounts. Tracklution's white-label multi-account structure is purpose-built for this. Stape if your team has GTM engineers on payroll. DataCops if bot filtering is a differentiator you want to put in client pitches.

EU-first, data residency required, regulated vertical. Datahash. Nothing else in this list carries the compliance documentation for finance, healthcare, and legal. For EU SMBs who need Consent Mode v2 and CAPI without the Datahash price tag, Addingwell (now Didomi-owned) or DataCops both handle first-party consent plus CAPI under one roof.

B2B SaaS, offline conversion tracking, lead quality scoring. Attribution platforms like Cometly or Hyros are better fits than pure CAPI tools. The problem is not just sending events; it is stitching closed-won revenue back to the original ad click. HubSpot offline conversion import through DataCops at $49/month covers a subset of this if you run Meta or LinkedIn and want lead scores passed back to the platform.

The tools

DataCops

The only tool in the category with bot filtering at the IP level before CAPI payload construction. Every event passes through a 361 billion IP database covering 146.4B datacenter IPs, 202B residential and mobile IPs, 11.9B VPN endpoints, and 620M proxy addresses before it routes to Meta, Google, TikTok, or LinkedIn. The first-party CMP loads from your own subdomain rather than a third-party CDN, which means it does not get blocked by uBlock Origin or Brave. Anonymous analytics continue after consent rejection because anonymous data is always legal. Identifiable data waits for consent.

The architecture runs on a single script tag and one CNAME record. Live in 5-30 minutes on Shopify, WooCommerce, Webflow, or custom. No developer required. First-party cookieless persistent identity means returning users are recognized without cookie expiry or ITP degradation. In the EU, identity resolution activates after the consent banner; outside the EU, it activates by default.

What does not work: SOC 2 Type II is in progress, not complete. If your procurement process requires certified compliance today, Tracklution or Datahash are the correct answer. No Pinterest CAPI, no Snapchat CAPI. If Pinterest drives meaningful acquisition volume, this is a hard stop. Shopify-native checkout hooks at the Checkout Extensibility level are less mature than Elevar's. CAPI starts at Business plan. Free and Growth plans ($0 and $7.99/month) include bot detection and first-party analytics but not CAPI routing.

Right for: Multi-platform advertisers running paid on two or more channels who want bot-filtered, consent-enforced CAPI without assembling three separate vendor relationships. Value 9/10. Business plan $49/month.

Read more about the Meta CAPI setup and fraud traffic validation on the DataCops site.

Stape

The most widely deployed server-side GTM hosting platform in the category. 80+ pre-built templates covering Meta CAPI, Google Ads, TikTok, Pinterest, Snapchat, LinkedIn, and more. The template library is genuinely the best in class, and if you have a GTM engineer in house, the container customization ceiling is higher than any managed tool. Stape runs on Google Cloud infrastructure with regional container options.

What does not work: Stape is infrastructure, not implementation. Building a production sGTM setup requires real GTM knowledge. Getting the Meta CAPI template configured correctly with deduplication keys and proper event parameters is not a 30-minute task for a non-technical marketer. Stape introduced Smart Pause in April 2026: if your container exceeds the usage limit by 10%, it gets auto-paused. Only Business+ plans get a 30-day grace period. Lower tiers face a hard tracking outage during traffic spikes, which is exactly when CAPI accuracy matters most. No bot filtering. Cookie Keeper is a paid add-on. Custom Pixel injection on Shopify carries a 5-8 second lag that can affect attribution window accuracy. Reviewers on G2 and Trustpilot consistently cite the pricing structure as complex: base plan plus Cloud Run plus add-ons adds up faster than the headline number suggests.

Right for: In-house GTM engineers who want full container control and the deepest template library in the category. Value 7/10. Pro $17/month plus Cloud Run $50-300/month.

Tracklution

The cleanest no-code alternative to Stape for teams that want multi-platform CAPI without GTM expertise. No container to configure. Plug your platforms in through a UI, enter your CNAME, done. SOC 2 Type II and ISO 27001 certified, which matters for enterprise procurement and regulated-industry clients. EU data residency available. White-label multi-account structure makes it the strongest agency play in the managed tools tier.

What does not work: Bot filtering is not a core feature. Tracklution routes clean and bot-contaminated events to Meta with equal efficiency. If your traffic mix includes significant VPN, datacenter, or proxy sessions, you are building Lookalike Audiences on that mix without any upstream filter. Reviewers cite the Starter plan as thin on support. Pricing converts from euros, which creates ambiguity for non-EU buyers on invoicing. The integration catalog is narrower than Stape's template library.

Right for: EU-leaning agencies and multi-account teams who want managed no-code CAPI with compliance certification and white-label options. Value 8/10. Starter €31/month.

Elevar

The deepest Shopify tracking implementation available. Elevar works at the data layer level on Shopify, with hooks into Checkout Extensibility, Shop Pay, and Apple Pay that capture identity and order data at a fidelity no platform-agnostic tool matches. Five years of Shopify-native architecture means edge cases like subscription recharges, multi-currency orders, and App Pixel behavior are handled out of the box.

What does not work: Shopify only. If you run any non-Shopify storefronts, Elevar cannot help you. Pricing escalates steeply with order volume: $200/month at 1,000 orders, $450/month at 10,000 orders, $950/month at 50,000 orders with overage charges layered on top at $0.03-0.15 per extra order. Q4 spikes produce unpredictable overage bills. No bot filtering at the CAPI level. Reviewers on G2 flag the pricing trajectory as a recurring complaint, particularly for stores that grew into the middle tiers faster than expected.

Right for: Shopify-only stores at $500K+ GMV/month where order-level checkout fidelity is worth the premium and the platform is not changing. Value 7/10. Essentials $200/month.

SignalBridge

The best pure value play in the managed CAPI tier in 2026. $29/month covers Meta CAPI, Google Enhanced Conversions, TikTok Events API, LinkedIn CAPI, bot filtering, funnel analytics, and ad spend sync. No comparable tool bundles that feature set at that price. Setup takes 5-15 minutes with no GTM requirement.

What does not work: Newer brand than the category incumbents. Review volume is thinner, which means less community troubleshooting knowledge and fewer agency integrations. Bot filtering exists but the IP database depth is not published, which makes it hard to compare to DataCops's documented 361B IP coverage. No built-in CMP, which means EU advertisers need to layer a separate consent tool for Consent Mode v2 compliance. The funnel analytics are basic compared to dedicated attribution platforms.

Right for: SMBs and growth teams who need multi-platform CAPI with basic bot filtering and do not want to pay Elevar or DataCops prices. Value 9/10. Base $29/month.

Addingwell (now Didomi)

Addingwell was acquired by Didomi for $83M in April 2025. The combined entity is the EU's most serious CMP-plus-server-side-infrastructure vendor. Addingwell brings Tag Health monitoring with real-time alerts, EU-hosted Google Cloud infrastructure, and GDPR-compliant data residency. Didomi brings TCF 2.2 consent management with enterprise compliance documentation. The combination was the logical endpoint of where the market was heading: one vendor for consent and one for server-side was always two vendors too many for EU compliance teams.

What does not work: Pricing has moved upmarket post-acquisition. The free tier covers 100,000 requests per month, which is usable at SMB scale, but paid tiers run higher than pre-acquisition Addingwell pricing. No bot filtering. The platform is fundamentally sGTM hosting with premium monitoring, which means configuration complexity is similar to Stape. Non-EU buyers typically find the feature set overbuilt for their needs relative to the price. The combined brand identity is still consolidating, which creates some support routing friction.

Right for: EU enterprises that need consent management and server-side CAPI infrastructure from a single vendor with documented data residency and real-time tag monitoring. Value 7/10. Free to 100K requests, paid tiers EUR-based on volume.

Meta 1-Click CAPI (free)

Meta launched this in Events Manager on April 15, 2026. It requires no developer, no sGTM, no external tool. For a Shopify merchant who has not yet bridged their Meta Pixel to server-side, this is the correct first action before evaluating any paid tool. EMQ improvement is real and immediate for stores that have been running pixel-only.

What does not work: Meta only. No Google, TikTok, or LinkedIn routing. No bot filtering. No consent management. No first-party identity resolution. The EMQ ceiling is limited compared to enriched server-side implementations. Events Manager bot detection is basic and does not filter at the IP level before the event is recorded as a conversion. This is the floor, not the ceiling.

Right for: Single-store Shopify merchants running only Meta who need the simplest possible CAPI upgrade at zero cost. Value 10/10 for that specific use case. Free.

Google Tag Gateway (free)

Google's answer to the same problem. Launched January 2026. One-click deployment via Google Cloud Platform, Cloudflare, or Akamai. Routes Google Ads Enhanced Conversions server-side with no sGTM configuration required. Google only. Same structural limitations as Meta's free tool: no cross-platform routing, no bot filter, no consent layer.

Right for: Advertisers running only Google Ads who want a free, zero-configuration Google Enhanced Conversions upgrade. Value 10/10 for that use case. Free.

Cometly

Attribution platform with server-side CAPI as a supporting capability rather than the core product. The value proposition is multi-touch attribution and customer journey visibility, not raw event relay. Cometly captures the complete funnel from first click to conversion and feeds enriched data back to Meta, Google, and TikTok through server-side connections. For B2B SaaS teams where understanding which campaigns drive pipeline matters as much as the accuracy of the CAPI connection, Cometly is one of the few tools built explicitly for that use case.

What does not work: Custom pricing on a sales-led model, which published sources indicate runs $199-499/month. That puts it in a different price tier from the managed CAPI tools above and requires a demo to get actual numbers. Not purpose-built for ecommerce. No bot filtering at the IP level. The attribution intelligence is the product; pure CAPI accuracy is secondary.

Right for: Growth and performance marketing teams at B2B SaaS companies who need attribution visibility across the full funnel alongside server-side conversion feeds. Value 7/10. $199-499/month.

Triple Whale

Attribution dashboard for Shopify ecommerce, not a CAPI tool in the traditional sense. Triple Whale ingests your conversion data and builds multi-touch attribution, cohort analysis, and creative performance reporting. CAPI is a data feed, not the product. The distinction matters because Triple Whale improves what you see in your dashboard; it does not improve what Meta's algorithm receives as a training signal. If your upstream CAPI feed contains bot conversions, Triple Whale charts them beautifully alongside your real ones.

What does not work: Shopify-only. Expensive entry point at $179/month on annual billing. The attribution intelligence is a dashboard layer on top of data it receives; it cannot fix bad data at the source. GMV-based pricing escalates sharply above $5M. No bot filtering. Using Triple Whale for CAPI and attribution while running bot-contaminated events is the exact scenario the Layer 5 problem describes: numbers that look right in the dashboard while the algorithm underneath is training wrong.

Right for: High-volume Shopify brands who want best-in-class creative and cohort analytics alongside attribution, and have already solved upstream data quality through another tool. Value 7/10. Starts $179/month annual.

Northbeam

Enterprise attribution platform with CAPI and MMM integration. Purpose-built for brands spending $500K+ per month on paid media who need modeled attribution across channels that pixel tracking cannot close. The platform ingests server-side conversion data alongside media mix modeling to produce blended attribution that accounts for incrementality.

What does not work: Entry at $1,500/month climbing to $5,000-10,000+ for volume accounts. Built for enterprises with dedicated media teams. No small or mid-market play here. Like Triple Whale, Northbeam improves what you measure, not what you send. Bot events in the CAPI feed produce polluted training signals regardless of how sophisticated the attribution model sitting downstream is.

Right for: Brands spending $500K+/month on paid who need MMM-level attribution and have the budget and team to use it. Value 7/10. $1,500/month entry.

Littledata

Shopify-focused tool built around GA4 accuracy and server-side tracking for ecommerce. Strongest use case is fixing the GA4 data layer on Shopify, where session stitching, UTM persistence, and checkout event accuracy are chronic problems. CAPI is part of the offering but the platform's identity is GA4 enrichment first, CAPI second.

What does not work: Shopify-centric with limited multi-platform support. Pricing at $199/month Standard scales per order. No bot filtering. Users on G2 note that setup complexity is higher than expected for a supposedly managed tool, particularly around custom event configurations. If your primary problem is Meta CAPI rather than GA4 session accuracy, this is not the best-fit tool.

Right for: Shopify stores that have chronic GA4 session and UTM attribution problems alongside CAPI needs. Value 6/10. $199/month Standard.

TrackBee

Meta and Google CAPI tool built specifically for Shopify. Zero-configuration setup, 30-day free trial, clean onboarding. Strongest positioning for mid-market Shopify brands spending consistently on Meta who want reliable CAPI without GTM expertise.

What does not work: TrackBee repriced in early 2025 to a more expensive subscription model. Multiple Trustpilot reviewers flag the new pricing structure as a pain point for entry-level stores. At €79/month entry (€25K tracked revenue, two stores), it is expensive relative to SignalBridge or Tracklution for stores still testing CAPI impact. Coverage outside Meta and Google is thin. No bot filtering. Lighter review base than Elevar or Littledata for validating edge case behavior.

Right for: Mid-sized Shopify brands running consistent Meta and Google spend who want zero-configuration CAPI and can absorb the pricing. Value 7/10. Start €79/month.

Datahash

Enterprise compliance-first CAPI platform. The only tool in this list that is purpose-built for regulated verticals: finance, legal, healthcare. Documentation depth for SOC 2, GDPR, and custom DPA requirements is the product, not a checkbox. Custom pricing starting at $500-2,000/month depending on volume and configuration.

What does not work: Not a fit for any buyer who does not have compliance-driven procurement requirements. Custom pricing means no self-serve evaluation. No bot filtering at the IP level. Setup requires enterprise onboarding.

Right for: Finance, legal, and healthcare brands for whom data residency, audit trails, and regulatory documentation are non-negotiable and no other tool in the category will pass procurement. Value 8/10 for that use case. Custom $500-2,000+/month.

TAGGRS

Budget sGTM hosting platform competing at the low end of the Stape price point. EU data residency option. Simpler pricing structure than Stape.

What does not work: Rough UX compared to the polished interfaces of Stape or Addingwell. Limited template library. Community and documentation are thin. Same structural limitation as all sGTM tools: requires GTM configuration knowledge. No bot filtering. Reviewers on G2 describe a steeper-than-expected learning curve relative to the tool's positioning.

Right for: Budget-conscious EU teams who want sGTM hosting with EU data residency and can live with a rougher onboarding experience. Value 6/10. Starts $20/month.

Converge

Multi-platform server-side CAPI tool with a clear no-code positioning. Covers Meta, Google, TikTok, Pinterest, Klaviyo, and Snapchat from a single dashboard. Setup through a UI with no GTM requirement. Platform-agnostic: Shopify, WooCommerce, BigCommerce, and custom stores.

What does not work: No bot filtering. Pricing is not clearly published and requires direct contact for most configurations beyond entry level. Review volume is thinner than the established players. The consent management situation is not native; EU advertisers need to layer a separate CMP.

Right for: Multi-platform ecommerce brands that need Pinterest and Snapchat CAPI alongside Meta and Google, and for whom those specific channels are active acquisition investments. Value 7/10. Contact for pricing.

Hyros

Sales-led attribution platform in the $1,000-5,000/month range. The core product is call tracking, funnel attribution, and offline conversion data stitching for high-ticket offers and info product businesses. CAPI is one output from the platform, not the primary purpose. Reviewers consistently cite accurate attribution for complex multi-step funnels as the genuine differentiator.

What does not work: Entry price is prohibitive for any buyer under $100K/month in ad spend. Sales-led process means no self-serve trial. Not a fit for standard ecommerce attribution needs. No bot filtering at the CAPI level. The platform is built for a specific customer profile: high-ticket offers, long sales cycles, heavy phone-call touch points.

Right for: High-ticket businesses with phone-call sales funnels that need offline conversion stitching and can justify the price on attribution improvement alone. Value 7/10. $1,000-5,000/month.

RudderStack

Open-source customer data platform with server-side event routing as one capability among many. RudderStack is the self-hosted option for engineering teams that want to own the full data pipeline rather than pay for managed infrastructure. Meta CAPI, Google, and 200+ destinations are available through the open-source SDK.

What does not work: This is a developer tool. Marketing teams without engineering support cannot operate it. The open-source version requires infrastructure management. Cloud pricing escalates with event volume. Not a fit for any non-technical buyer. No native bot filtering. No CMP included.

Right for: Engineering-led companies that want open-source control over customer data routing and have developers who can maintain the stack. Value 8/10 for that profile. Open-source self-hosted, cloud pricing per event volume.

Reaktion

Shopify-native tracking and analytics platform with real-time profit and LTV dashboards alongside server-side CAPI. 1-click Shopify App Store install. Connects Meta and Google CAPI while showing ROAS at a per-product and per-collection level.

What does not work: Shopify only. The profit dashboard is the primary product; CAPI is a secondary feature. No bot filtering. No CMP. Not a fit for multi-platform stores. Newer brand with a thinner review base than Elevar or Littledata.

Right for: Shopify merchants who want profitability visibility alongside server-side tracking and find Elevar's pricing too steep for what they need. Value 7/10. Pricing on request.

Feature comparison

When NOT to use DataCops

This section exists because trust requires honesty and there are specific situations where DataCops is not the right answer.

Single-platform Meta-only Shopify store under $30K/month GMV. Meta's free one-click CAPI does what you need. The EMQ gap between free and $49/month does not produce enough CPA lift to justify the spend at that scale. Start with free, revisit when monthly Meta spend clears $3,000.

SOC 2 Type II certification required today in vendor approval. DataCops's certification is in progress, not complete. If your procurement team has a hard requirement for current certification, Tracklution or Datahash are the correct answers and DataCops is not.

Pinterest is a primary acquisition channel. DataCops does not support Pinterest CAPI. No workaround exists. If Pinterest drives meaningful purchase attribution, you need Converge, Stape with the Pinterest template, or another tool with native Pinterest support.

In-house GTM engineers who want full container ownership and want to build custom transformations. DataCops is an outcome-oriented tool: you point it at your store, it handles filtering and routing. Engineers who need to intercept events at the container level, run custom JavaScript transforms, or build complex multi-tag logic inside sGTM should use Stape. The ceiling for customization is higher there, at the cost of assembly time and expertise.

EU enterprise procurement with existing Didomi relationship. If Didomi is already in your vendor stack for consent management, the Addingwell acquisition makes the combined Didomi platform the natural path to add server-side CAPI. You avoid a second vendor relationship and the consent-to-CAPI integration is handled natively within one contract.

The real question for 2026

The market has converged on a solved problem. Every serious CAPI tool delivers server-side events. Meta and Google both do it free. The data pipe is not broken anymore.

The question nobody is asking when they compare tool feature grids: are the events arriving at Meta actually real humans? Of the CAPI events you sent last month, how many came from datacenter IPs, VPN exits, headless browsers running Playwright or Selenium, or residential proxy networks that ad fraud operations rent by the hour? Fraudlogix puts global IVT at 20.64%. Instagram Audience Network runs 67% invalid.

Meta's algorithm does not distinguish between a real purchase and a bot-generated purchase event. It receives both as training signal. Project Andromeda acts on that signal within hours. Your next Lookalike Audience reflects whatever mix you sent.

The advanced conversion tracking guide on this site goes deeper on what fixes the foundation, and the first-party analytics page covers identity resolution in more detail. For the consent layer specifically, the first-party CMP architecture explains why third-party CDN-loaded consent tools fail 30-40% of sessions before the banner ever loads. The B2B conversion tracking best practices guide covers the lead quality and offline stitching angle.

The pipe is fixed. The water question is still open.

What percentage of the conversion events you sent to Meta in the last 30 days came from a real human being at a real device, making a real decision?