Using Ad Fraud API

Overview

The Ad Fraud API returns the real-time probability of invalid traffic (IVT) and the types of IVT associated with an IP address, device ID or user agent, so you can decide whether it's compromised before you bid on, serve or pay for an impression.

Each lookup draws on Pixalate's continuously updated fraud intelligence across mobile, CTV and web, so you can block bad actors with thresholds that fit your own supply and risk tolerance.

Usage guidelines

Make a fraud query

Send an IP address, device ID or user agent to get the probability of fraud and the associated risk types. You can send any combination of ip, deviceId and userAgent. If you send more than one, probability reflects the combined risk of all the values you sent. riskTypes lists the risk types associated with all of the values you sent during the look-back period, so it can contain more than one code.

Input:

Parameters	Accepted values
IP address	IPv4 and IPv6 addresses
Device ID	Mobile and CTV advertising IDs: ADID (Android), IDFA (iOS), RIDA (Roku OS), MSAI (Xbox), WAID (Windows), and MD5 or SHA-1 hashed device identifiers
User agent	Browser and device user agent strings

Response:

Field	What it tells you
`probability`	How likely the value is to be associated with fraud, up to 1.0. `0` means it isn't flagged. `-1` means it's whitelisted by Pixalate.
`riskTypes`	The four-digit risk type codes that explain why the value is flagged. It reports all applicable risk types so there can be multiple values.

Block with probability thresholds

Pixalate's proprietary machine-learning algorithm calculates the probability, so you can set your own blocking threshold based on the quality and scale of your supply inventory. As a guide:

Probability equal to 1.0: you filter out only the worst offenders (deterministic).
Probability greater than or equal to 0.90: you filter out users who are fraudulent beyond a reasonable doubt.
Probability between 0.75 (inclusive) and 0.90 (exclusive): you filter out users with clear and convincing evidence of fraud.
Probability between 0.5 (inclusive) and 0.75 (exclusive): you filter out users who are more likely than not fraudulent.
Probability 0: the value isn't flagged as fraudulent by Pixalate's fraud intelligence system.
Probability -1: the value you sent is whitelisted by Pixalate. Only a small number of IP addresses and device IDs are whitelisted, so you'll rarely see this value.

We don't recommend blocking any probability below 0.5. If you adjust your threshold, check impression delivery regularly, because lowering the threshold can reduce impression counts.

Explain, report and block with risk types

Risk types tell you the reason an IP, device ID or user agent is flagged for fraud. You can also use them in blocking rules, on their own or together with a probability threshold.

Each risk type has a designated four-digit code. Codes follow the format T C NN: T is the IVT class (1 = GIVT, 2 = SIVT), C is the category within that class, and NN is the type within that category. For example, 2201 = Invalid Proxy Traffic (IVT type), in the SIVT class and False Representation category. You can block at any level: all SIVT (2xxx), one category (25xx) or a single type (2502).

For the full list of risk type codes, names and definitions, see Pixalate's IVT Taxonomy 2.0.

See the Ad Fraud API developer documentation for full API details.