LayerCall's answer

Most fraud APIs answer one question per call — is this IP a VPN, is this email disposable, is this phone real. LayerCall returns IP, email, phone, domain and device together, and scores the relationships between them. A brand-new domain paired with a datacenter IP and a throwaway mailbox is obvious in combination and unremarkable one field at a time.

Every response also carries the reasoning: a 0–100 risk score, an allow / review / block verdict, and the individual signals behind it, so a decision can be explained rather than only made.

It treats AI agents as a first-class case as well. Web Bot Auth signature verification establishes which agent is calling and whether it can prove it, and a policy engine decides what it is allowed to do — a question classical fraud signals cannot settle, because an agent arrives with a real browser, a real fingerprint and a real mailbox.

ShieldLabs's answer:

Identification that holds. ShieldLabs recognises a returning visitor with up to 99% accuracy, through cleared cookies, incognito mode, IP rotation and months between visits, because the device identifier is computed from the device rather than stored in the browser. That cuts both ways: the same recognition that exposes one person running twenty accounts also lets you greet a good returning customer without making them prove who they are again.

Anonymity detection with up to 99% accuracy, at any level rather than a binary VPN yes or no:

• Network Intelligence covers VPN, proxy, Tor, Apple Private Relay and datacenter IP.
• Device Intelligence covers anti-detect browsers, browser automation, OS mismatch and undetectable OS.

Each arrives as a separate named signal, so you know which kind of anonymity was found, not just that something was.

The risk score comes apart. Every score from 0 to 100 arrives with the named signals that produced it, so you can see why a visit scored what it did instead of trusting a number you cannot inspect.

Patterns, already correlated. The work across accounts, devices and identities is done before you open the dashboard, pointing to multi-accounting, account sharing, account takeover and account farms operating behind one connection. A starting point for investigation that is waiting when you arrive.

The full detection stack is on every tier. The API and webhooks are included from the first plan, pricing is published, and signup is self-serve all the way through.

LayerCall's answer

Developers and small product teams who need a trust decision at signup, login or checkout, and who would rather call one endpoint than integrate several vendors and reconcile their answers by hand.

In practice that means SaaS signups, marketplaces, fintech onboarding, and anyone whose free tier is being farmed by throwaway accounts.

A newer part of the audience is teams who suddenly have to decide what an AI agent may do on their site. That is a different question from classical fraud — an agent can be entirely legitimate and still need a policy — which is why agent verification sits in the same API rather than in a separate product.

ShieldLabs's answer:

Two groups, and they come in through different doors.

Developers and technical founders, most often the CTO at a team of five to fifty. The product is API-first, so whoever evaluates it is usually whoever integrates it. They want the raw signals and intend to write the decision logic themselves rather than buy a service that makes the call on their behalf.

Growth, marketing and analytics people, who may never open the API. Once the snippet is in, the analytics dashboard is theirs: which traffic sources bring anonymized and high-risk visitors, what share of sessions is clean, how it all moves week to week. For anyone buying traffic that is a direct argument about ad spend. For anyone reporting on product metrics it is the difference between counting sessions and counting people.

Installation takes one person about five minutes, and the dashboard needs no code after that, which is why the buyer and the daily user are often not the same person.

By sector it clusters where a free tier, a promotion or a signup carries real marginal cost:

• SaaS and AI products, where usage-based infrastructure bills turn each abusive signup into real money
• Ecommerce and marketplaces, where promotions and first-order discounts get farmed
• Fintech, where a signup is an account that moves money
• Gaming, where one person running many accounts distorts the economy for everyone else
• Crypto and Web3, where airdrops and referral programmes attract farms at scale

It also fits teams with no fraud problem at all. Recognising a good returning visitor is the same capability pointed the other way, whether that means personalising an experience or not asking someone to prove who they are twice.

LayerCall's answer

It started from a specific frustration: the signal that actually catches a fake signup is usually a relationship between fields, and the tools available answered one field at a time.

Blocking disposable email domains stops very little on its own. The signups that matter use real mailboxes, often on domains registered days earlier, arriving from addresses that look entirely ordinary. What gives them away is the domain's age set against the IP's provider set against whether the phone is a VoIP line — and assembling that meant several vendors, several response shapes, several bills, and writing the correlation by hand anyway.

LayerCall is that correlation as a product: one call, every signal, and the reasoning returned next to the score.

The AI-agent side came later, from the same observation in a new place. An agent has a real browser, a real fingerprint and a real mailbox, so nothing in a classical fraud stack has an opinion about it. What you need to know is which agent it is and whether it can prove it — a signature problem, not a fraud-signal problem.

ShieldLabs's answer:

ShieldLabs came out of four gaps that repeat across this category.

Who the tooling was built for. Everything that reliably identifies visitors and detects anonymity has been priced and packaged for enterprise buyers: a demo to book, a quote to wait for, a procurement cycle to survive. The teams losing the most to free-tier farming, promo abuse and fake signups are small and self-serve, and that path is not built for them.

What anonymity detection had been reduced to. In most stacks it is a side feature: a binary VPN yes or no bolted onto a product built for something else, usually resting on IP reputation lists that go stale as fast as residential proxy pools rotate through consumer addresses. Anti-detect browsers, the working tool of anyone doing multi-accounting at volume, are rarely a first-class signal at all.

Explainability. Most detection products return a number. When a paying customer trips over that number, the team has nothing to inspect and nothing to tell the user.

The work that begins after the purchase. A product that hands over raw signals has moved the problem rather than solved it, because someone still has to build the model, correlate across accounts and devices, and keep the whole thing current. Time to a first useful answer gets measured in weeks.

So the product was built around four commitments:

• Anonymity as the product, not a checkbox. Device and network signals together, up to 99% accuracy in detecting anonymity from a clean visit through VPN, proxy, Tor and Privacy Relay to anti-detect browsers and browser automation, each surfaced under its own name.

• Pricing published, stack whole. The full detection stack on every tier, self-serve from the first plan through the top one, so the path from landing page to first signal runs inside a single session.

• Every score decomposable into the named signals behind it, so the customer's code owns the decision while ShieldLabs owns the evidence.

• Usable on day one. One JavaScript snippet, about five minutes to the first signal, patterns already correlated rather than left as an exercise, and an analytics dashboard that reads without code, so the person who needs the answer is not waiting on the person who can write the query.

LayerCall's answer

Because of what comes back in the response, not what it costs.

Every result carries a 0–100 risk score, an allow / review / block verdict, and the individual signals behind it — so a decision can be explained to a customer, a colleague or an auditor rather than only made. Strictness is tunable per request without re-scoring, which means the same integration can be strict at signup and forgiving at login.

Two smaller things tend to matter more in production than they sound. When a data source is unavailable, the response says so instead of quietly scoring lower, so an incomplete answer stays distinguishable from a clean one. And test keys return fixed, fictional data that never bills and never touches live reputation data, so a test suite can assert on exact values without polluting anything.

Beyond that, it is worth comparing directly rather than taking our word for it: the live demo runs the real scoring engine with no signup, and the free tier needs no card.

ShieldLabs's answer:

Depth at a self-serve price. The level of anonymity detection that usually sits behind an enterprise conversation is available on the entry tier here, and API access is included on every plan rather than unlocked further up.

You can be running today. One JavaScript snippet, about five minutes to the first signal, and the whole path from the pricing page to a working integration is self-serve: no demo to book, no sales call, no quote to wait for. Evaluation happens on your own schedule and on your own traffic.

The score is explainable. When a legitimate customer gets caught by your rules, looking up why is a query rather than a guess.

Patterns arrive pre-computed. The correlation work across accounts and devices is done before you open the dashboard.

Traffic quality is a first-class view, not a by-product of scoring. The analytics dashboard gives you:

• an overall risk score for your traffic, and how it moves over time
• the split across Clean, Low, Medium and High risk bands
• the same breakdown per traffic source, so you can see which channels bring anonymized and high-risk visitors

Your analytics tool tells you where traffic came from. This tells you what arrived. For anyone buying traffic, that turns a monthly ad invoice into something you can argue with.

Pricing is published and flat:

• the rate per 1,000 identifications drops as you move up the tiers, so growth makes each identification cheaper
• yearly billing saves 20%
• the free tier is 5,000 identifications with no card and no expiry, enough to measure signal quality on your own traffic before committing to anything

Your code makes the decision. ShieldLabs returns the signals and the score, so the logic specific to your business stays in your codebase where you can change it.

LayerCall's answer

TypeScript on Next.js, running on Vercel's Fluid Compute, with Postgres (Supabase) behind accounts, keys and usage.

The scoring path is deliberately boring. No third-party SDK sits in the request path; every external feed is fetched under its own timeout inside a request-wide deadline, so one slow source cannot hold up a response. A feed that fails degrades the result rather than failing the call, and the response names any signal that was unavailable so the caller can tell the difference between a clean answer and an incomplete one.

On the client side: official Node/TypeScript and Python SDKs, Express and Next.js middleware, a published OpenAPI spec, and an MCP server so AI tools can call the API directly.

ShieldLabs's answer:

What you integrate is deliberately small. One ES module loaded from the CDN, and everything after that arrives through a REST API and signed webhooks. It assumes nothing about your stack.

There is an install guide for whatever you already use: JavaScript, React, Next.js, Vue, Angular, Svelte, Preact, React Native WebView, plus WordPress, Shopify and Tilda.

On WordPress, Shopify and Tilda that means no developer at all. The module goes where the platform already keeps custom code, and that is the whole integration.

Nothing to maintain afterwards. Detection keeps improving without you shipping an update, so what you install today gets better on its own.