FAQ
That is, questions and answers to better understand ABTShield.
2
3
What is ABTShield?
ABTShield is a service designed to protect your advertising budget from "low-value" traffic, which mainly consists of ad fraud. This fake traffic is generated by automated and malicious mechanisms such as bots that click on ads, browse pages, or mimic user actions, ultimately wasting your ad budget without delivering real business value. ABTShield identifies this low-value traffic and enables you to take action to eliminate it.
2
3
What problem does ABTShield solve?
ABTShield solves the significant problem of ad fraud, which can drain advertising budgets, distort campaign performance data, and prevent advertisers from reaching real customers. It helps ensure that your advertising spend is directed toward genuine, high-quality traffic sources that can generate actual business value and conversions.
2
3
What is ad fraud and SIVT (Sophisticated Invalid Traffic)?
Ad fraud is a general term for advertising fraud. SIVT, or "Sophisticated Invalid Traffic," is a more advanced form of bot traffic that is non-transparent in its operation and uses various techniques to evade detection. The goal of SIVT is to simulate human activity and capture revenue through illegal means. ABTShield focuses on detecting various categories of ad fraud, including SIVT (e.g., behavioral simulations initiated by bots, click farms, automated browsers, malware, botnets).
2
3
How does ABTShield work? What are its main components?
ABTShield operates based on two main pillars:
a. Identification: It uses a proprietary non-cookie identification solution (fingerprinting technology) to collect, analyze, and process high-quality data on every device interacting with your website. This enables unique device identification, even if cookies are deleted or not used.
b. Behavioral Modeling: Advanced algorithms analyze user traffic and behavior patterns in real time. This helps segment traffic sources, identify bots, eliminate ad fraud, and assess the business value of different traffic segments.
2
3
How does ABTShield detect anomalies and bot activity?
ABTShield uses a two-pronged approach:
a. Passive Fingerprinting: This mechanism identifies and tracks individual devices interacting with your website. It enriches every event (such as a page view or click) with a unique identifier representing the device, based on network (TCP/IP, TLS) and device characteristics—not just cookies.
b. Anomaly Detection Process: Once a device is identified, ABTShield analyzes its activity using over 50 consistency and outlier behavior tests to determine whether its actions are typical of a genuine user or indicate automated traffic. This includes BOT scoring, which uses statistical and heuristic tests (both behavioral and configuration-based) to flag suspicious devices.
2
3
What are some reasons why a device may be flagged as a bot by ABTShield?
ABTShield uses an extensive list of behavior- and configuration-based reasons. No single indicator determines a bot flag; it’s a combination of factors and weights. Some examples include:
a. Behavioral: Machines focused on generating connections to single pages, unnaturally high activity in a short time, suspiciously similar time intervals between page requests, repetitive traffic patterns, frequent cookie deletion.
b. Configuration-based: Modification of browser name or language, suspicious browser setup, devices impersonating GoogleBot outside the Google network, botnet risk due to hardware signature.
2
3
What is a “non-cookie identification solution”?
It refers to ABTShield’s ability to identify and track unique devices without relying on traditional browser cookies. It uses advanced fingerprinting techniques that analyze a combination of device and network characteristics (e.g., IP address, port, TCP/IP stack, HTTP headers, TLS handshake parameters, and even subtle browser configurations) to create a persistent and unique identifier for each device. This method is more robust than cookies, which can be easily deleted or blocked.
2
3
How does ABTShield handle dynamically changing IP addresses (e.g., in mobile networks or when using VPNs)?
Although the IP address is one of the analyzed signals, the ABT_ID identifier is not based solely on it. ABTShield correlates the IP address with many other, more stable features from the network, transport, and application layers. An IP change is taken into account, but if other key fingerprint elements remain consistent, the system can still attribute activity to the same unique device with high probability. In the case of VPN use, the analysis can detect characteristics typical for tunneled traffic, which may serve as an additional risk signal.
2
3
How often are ABTShield’s machine learning algorithms updated and what data are they trained on?
ABTShield’s machine learning algorithms are continuously monitored and adapted. The update process is ongoing and includes:
a. Regular effectiveness testing: Comparing results with reference data and identifying areas for improvement.
b. Training on new data: Models are retrained on a continuous stream of anonymized web traffic data, enabling them to learn new fraud patterns and adapt to changing bot behaviors.
c. Research and development of new signatures: Introducing new features into the models as they are discovered and validated.
The goal is to maintain the highest possible effectiveness in a dynamically evolving threat landscape.
2
3
Does ABTShield use any active probing techniques on devices, or does it rely solely on passive traffic analysis?
The core philosophy of ABTShield is based on passive traffic analysis. This means the system analyzes connection characteristics and data naturally transmitted by the user's device during interactions with the server, without actively sending additional requests or scripts aimed at “probing” the device for specific information. This approach minimizes the risk of disrupting the user's device and is less invasive. JavaScript on the client’s site is used mainly to collect information about events and some browser characteristics that are voluntarily shared.
2
3
How do I start using ABTShield?
a. Registration: Contact the ABTShield team to create an account.
b. Integration: Integrate the ABTShield JavaScript code with your website or use tracking tags by following the provided instructions.
c. Activation and Monitoring: Once integrated and your campaign is live, activate ABTShield services to begin monitoring advertising and web traffic. You’ll receive access to a personalized report (e.g., in Looker Studio) to review your data.
d. Automatic Elimination Setup: Work with the ABTShield team to configure procedures for the automatic elimination of low-value traffic tailored to your specific campaign types.
2
3
How can ABTShield collect data? What are tracking tags and the abtshield-snippet.js library?
We offer clients two integration strategies:
a. Tracking tags: These are URLs that typically direct the browser to a 1x1 pixel image. The tag can include parameters that pass information about the context of the call. In advertising, they capture interactions with the creative (impressions, clicks). It’s also possible to prepare redirect tags to the landing page.
b. abtshield-snippet.js library: This is a JavaScript library that enables comprehensive collection of traffic information on your website. It operates in the page’s JavaScript layer, allowing for more extensive data collection compared to HTTP GET protocol tags.
Using both strategies simultaneously is the most effective approach.
2
3
Is it possible that, despite the code being implemented on a click-through, data is not being collected in the ABTShield system?
Yes, this can happen. Most often, it is caused by the fact that the ABTShield system may not want to redirect to non-secure (http, without SSL) websites.
2
3
How technically advanced is ABTShield integration? Do I need dedicated developer resources?
ABTShield integration is designed to be as simple as possible. Typically, it involves adding a short JavaScript snippet to your website (similar to other analytics or marketing tools) or configuring tracking pixels. Basic integration can often be performed by the person managing your website or advertising systems. For more advanced setups or integration with mobile applications, developer support may be helpful, but the ABTShield team provides detailed instructions and technical support to keep the process as effortless as possible.
2
3
What does the ABTShield implementation process look like? How long does it typically take from signing the contract to full system operation?
The implementation process is typically quick and efficient:
a. Consultation and contract: Discussion of needs, terms, and contract signing.
b. Technical integration: The client receives instructions for implementing the JavaScript code or pixels. The ABTShield team offers support at this stage.
c. Activation and calibration: System launch, collection of initial data, and algorithm calibration (usually takes a few days).
d. Access to dashboard and reports: The client is provided with tools for monitoring and analysis.
e. Support and optimization: Ongoing cooperation in the area of analysis and optimization.
The entire process from contract signing to full system operation can take from a few days to a few weeks, depending on the complexity of the client’s website/application and the speed of integration on their side.
2
3
What types of web resources and traffic can ABTShield monitor?
ABTShield is versatile and can monitor traffic across various digital touchpoints, including ad impressions, clicks, and behavior on campaign landing pages (LP). It can detect both organic and paid web traffic from partners, advertising networks, and social media.
2
3
Does ABTShield support UTM tags?
Yes, ABTShield supports the use of UTM tags. This enables effective campaign tracking, understanding of traffic sources, and optimization of advertising spend by identifying the sources that generate the most value.
2
3
Can ABTShield track conversion events?
Yes, ABTShield can tag conversion events. This feature helps track and measure campaign results, analyze the quality of traffic leading to conversions, and make informed decisions about advertising budgets.
2
3
How does ABTShield help optimize campaigns beyond just identifying fraud?
ABTShield provides data for campaign optimization. By identifying and enabling the blocking of low-value traffic segments and IP addresses, it helps to:
a. Focus ad spend on high-quality sources.
b. Improve overall campaign performance (conversions, user engagement).
c. Maximize return on investment (ROI).
d. Support or even replace some media planner tasks by providing clear data on the quality of traffic from individual sources.
2
3
What are “whitebots” and how does ABTShield deal with them? What are bad bots?
a. Whitebots: These are legitimate bots and crawlers from well-known sources such as search engines (e.g., GoogleBot) and ad networks. They are usually not malicious and serve useful functions like content indexing (so-called GIVT – General Invalid Traffic). ABTShield uses a combination of static IP lists and ISP names to identify these whitebots and excludes them from the "bad bot" score calculation. This ensures that legitimate automated traffic is not incorrectly flagged. On average, they account for about 1% of traffic.
b. Bad bots: This refers to the previously mentioned SIVT (Sophisticated Invalid Traffic). This is a more sophisticated form of bot traffic that simulates human activity to capture revenue through illegal means.
2
3
What is “AD Fraud scoring” or the “low-value traffic” segment?
This scoring combines “BOT scoring” with an analysis of key business events (such as purchases or conversions). The goal is to identify a segment of traffic that, although potentially active, generates significant costs but does not deliver benefits in the form of conversions. This “low-value traffic” segment can then be safely eliminated from campaigns to improve efficiency.
2
3
Where can I view data and reports from ABTShield? Do I need a Google account?
You will receive access to the ABTShield dashboard, often provided as an interactive report on the Looker Studio analytics platform. This dashboard delivers interactive reports on your traffic quality, including details about unique users, ad clicks, conversion rates, ad fraud reports (bad domains, IPs, devices), and campaign-specific KPIs.
Yes, ABTShield reports are shared with client email addresses that function within the Google Accounts ecosystem. This means the email must be a Gmail account or another address linked to Google Accounts.
2
3
How quickly will I see data on the dashboard after integration? Is the dashboard provided immediately?
Typically, after completing integration and activating ABTShield services, initial data and reports become visible on the dashboard within about 2 full days. This allows for sufficient data collection and system calibration. The dashboard is not provided immediately after installing the pixels, as earlier access could result in an inaccurate bot traffic percentage due to low data volume.
2
3
Can I access the ABTShield dashboard if I haven’t yet measured any traffic in my campaigns?
The Looker Studio dashboard is provided to clients who have signed a contract, have their unique identifier, and have integrated their system with ABTShield. Access is granted after the data collection process is initiated. Providing the dashboard earlier is not practical, as its components only function correctly once they are supplied with data.
2
3
Does the Looker Studio dashboard allow for real-time tracking of collected data?
Data is collected in real time (each event has its own unique timestamp), but updates in the browser occur daily. This means the database is refreshed once per day, and the dashboard data covers up to the end of the previous day. A live view feature with same-day data is planned for future release.
2
3
What is EVENT COUNT in the dashboard, and is there a difference between a click and event count?
Event count refers to the number of records in the database. An event can be either an impression or a click. Therefore, if your campaign/traffic only measures clicks, the event count will correspond to the number of clicks.
2
3
I have identified bot traffic in my campaign. What percentage of bot traffic is considered high?
Benchmarks for certain campaign types (data from the past year):
a. Google Ads / Search: 16%
b. Programmatic: 21%
c. Paid Social: 7%
d. Direct / mailing: high variance, no calculated average.
2
3
How can you exclude bot traffic from future campaigns?
There are several options:
a. Creating a specific segment (low-value traffic and verified-traffic) via the ABTShield DMP account for Adform, DV360, Meta, Google Ad Manager, and others.
b. For Google Ads, a segment is created that is available in all campaign types (including as a “signal” in Performance Max). This requires access to the client’s website and communication with GAds through Google Tags.
When low-value traffic, fake segments, or specific bot IP addresses are identified, ABTShield provides the means to block them.
2
3
What are the main sections and filters in the ABTShield client report?
The standard dashboard is divided into sections:
a. GENERAL STATISTICS: Cross-sectional information, basic statistics.
b. SOURCE OVERVIEW: Traffic quality from individual advertising channels.
c. FRAUD INSPECTION PER DEVICE: Device specifications of detected traffic.
d. FRAUD INSPECTION PER IP: Detailed information about traffic and quality for specific IP addresses.
e. BOT EXPLORER: Overview tables and a special set of filters for in-depth analysis.
f. RAW DATA: Data broken down by each event, with download option.
g. RECOMMENDATION OF EXCLUSION IP: Table suggesting whether a given IP should be blocked.
The report includes dynamic filters, such as: bot_type (HUMAN, WHITE BOT, BOT), HTTP traffic source (referer), AdSource (event type), date range, and UTM filters (utm_campaign, utm_source, utm_content, utm_medium, utm_term).
2
3
Does ABTShield affect the performance (loading speed) of my website?
ABTShield is optimized for minimal impact on website performance. The JavaScript code is typically loaded asynchronously, meaning it does not block the loading of the rest of the page. Data analysis is performed server-side, further minimizing the load on the user’s browser. The goal is to provide protection without negatively affecting user experience (UX).
2
3
What happens if ABTShield mistakenly flags my legitimate traffic as fraud? Is there a risk of false positives?
ABTShield uses multi-layered algorithms and heuristic analysis to minimize the risk of false positives. The system is continuously calibrated and trained on large datasets. However, in very rare and unusual cases, there is a theoretical possibility of incorrect classification. In such situations, the ABTShield team works with the client to investigate the issue, understand the specifics of the traffic, and make any necessary configuration adjustments. Data transparency and the ability to analyze individual cases are key in this process.
2
3
Is ABTShield capable of handling very high traffic volumes?
Yes, ABTShield is designed for scalability and can efficiently process and analyze very high traffic volumes, typical for large e-commerce portals, publishers, or intensive advertising campaigns. The server infrastructure, including any-cast technology, ensures reliability and performance even under significant load.
2
3
Does ABTShield protect against specific, advanced types of fraud such as click injection or ad stacking?
Yes, ABTShield is designed to detect a wide range of advertising fraud, including more advanced techniques. New bot metrics and classification rules are continuously developed in response to evolving fraud methods.
2
3
How is ABTShield updated to cope with new fraud techniques?
Fighting ad fraud is an ongoing process, as fraudsters constantly develop new methods. ABTShield is regularly updated—this includes developing new signatures, adapting algorithms, monitoring trends, and collaborating with clients and partners.
2
3
How does the non-cookie ABT_ID differ from other cookieless identification solutions on the market?
ABT_ID is based on deep analysis of passive network signatures (TCP/IP, TLS) and device characteristics, making it difficult to manipulate. Many other solutions may rely more heavily on active client-side scripts or less stable parameters. The key advantage of ABT_ID is its resilience to standard user actions (such as clearing cookies) and its ability to identify devices across different contexts (cross-domain). Its effectiveness is regularly tested and benchmarked, and the methodology is transparent.
2
3
How does ABTShield enable the creation of segments for campaign optimization by filtering malicious traffic?
ABTShield, based on unique identifiers, allows for the creation and export of user segments (e.g., low-value traffic, traffic verified as valuable) to ad platforms such as Google Ads, DV360, Google Ad Manager, Meta, and AdForm. These segments can be used for ad targeting or to exclude specific groups of machines, directly improving campaign effectiveness. The segments are updated regularly (e.g., every 24 hours).
2
3
Is ABTShield able to detect fraud related to traffic coming from mobile applications?<br><br>
Yes, ABTShield is capable of analyzing and detecting fraudulent traffic in mobile application environments as well. The identification and behavioral modeling mechanisms are adaptable and can be used to monitor in-app interactions, provided an appropriate data collection method is implemented (e.g., via SDK or integration with the app's servers).
2
3
How does ABTShield handle advanced bots that attempt to mimic human behaviors (e.g., mouse movements, filling out simple forms)?
ABTShield uses advanced behavioral modeling—it analyzes hundreds of parameters and behavior patterns, including subtle anomalies that can distinguish a genuine user from even the most sophisticated bots. The system doesn’t rely solely on simple metrics but looks for inconsistencies across the entire activity profile of the device, enabling it to detect attempts to mimic human behavior.
2
3
Does ABTShield help identify issues related to traffic quality from specific publishers or affiliate networks?
Yes, this is one of the key applications. Through detailed traffic analysis by source (including UTM parameters), ABTShield enables precise assessment of the quality of traffic delivered by individual publishers, ad networks, or affiliate partners. You can identify which sources generate disproportionately high amounts of bot or low-value traffic, allowing for optimization of partnerships or renegotiation of terms.
2
3
How does ABTShield help combat "click farms"?
Click farms generate traffic often characterized by unnatural behavioral patterns, specific device configurations, and network characteristics. ABTShield can detect such anomalies through behavioral analysis, device fingerprinting, and network analysis.
2
3
Can I customize the ABTShield dashboard to fit my specific reporting needs?
Yes, ABTShield dashboards, often built on BI platforms like Looker Studio, are designed with flexibility in mind. In collaboration with the ABTShield team, it is possible to customize views, metrics, and filters to best suit your individual analytical needs and key performance indicators (KPIs).
2
3
What is the typical return on investment (ROI) after implementing ABTShield? Do you have any case studies?
Return on investment (ROI) after implementing ABTShield can be significant and depends on various factors. Clients often observe direct budget savings, increased campaign efficiency, and improved quality of analytical data. ABTShield has case studies from various industries, such as e-commerce, where a 78% reduction in ad fraud and a 43% increase in conversions were recorded.
2
3
How do changes in browser privacy policies (e.g., blocking third-party cookies, Safari’s Intelligent Tracking Prevention) affect ABTShield’s operation?
ABTShield is designed for the "post-cookie" era. Its core identification technology relies on device fingerprinting at the network level (TCP/IP, TLS) and configuration analysis, rather than solely on cookies. This makes it much more resilient to changes in browser privacy policies compared to solutions that primarily depend on cookies. Continuous development and adaptation of signatures are key to maintaining effectiveness in a rapidly evolving environment.
2
3
Is ABTShield suitable for smaller companies with limited budgets, or is it mainly a solution for large corporations?
Although ABTShield is a scalable solution capable of meeting the needs of large corporations, its value is also evident for small and medium-sized businesses. Protecting even a smaller budget from fraud-related waste can significantly improve marketing effectiveness. The pricing model is often flexible and can be tailored to the scale of operations, so it’s worth consulting the ABTShield team to discuss your individual needs.
2
3
Does ABTShield integrate only with advertising platforms, or also with other marketing or analytics tools?
The core integration of ABTShield focuses on ad platforms (AdForm, Google Ads, Meta, DV360, etc.) to enable fraud blocking and campaign optimization. However, due to its flexibility (e.g., data export, integration with BI systems like Looker Studio), ABTShield data can also be used in conjunction with other internal client analytics systems. It’s advisable to discuss specific integration needs with the ABTShield team.
2
3
Does ABTShield offer support in data interpretation and optimization recommendations?
Yes, the ABTShield team not only provides the tool but also offers support in analyzing the collected data. Clients can expect assistance in understanding reports, identifying key issues, and implementing recommendations for campaign optimization and eliminating low-quality traffic sources. The goal is a collaborative partnership aimed at maximizing advertising effectiveness.
2
3
What are the key metrics I should pay attention to in the ABTShield dashboard?
The most important metrics depend on your goals, but typical key indicators include: Ad Fraud level / % bot share, Traffic sources with the highest fraud levels, Impact of fraud on conversions/KPIs, Reports on bot classification reasons, and Unique device IDs and IP addresses associated with fraud.
2
3
Can ABTShield help renegotiate terms with traffic providers if it turns out they deliver a lot of fraud?
Yes, detailed and reliable reports from ABTShield documenting the level and nature of fraudulent traffic from specific sources can provide a strong basis for discussions and renegotiations with traffic providers. Presenting evidence of low-quality traffic can help secure refunds, better rates, or decisions to terminate partnerships with unreliable partners.
2
3
What data does ABTShield process? Is it GDPR-compliant? Can I identify individual persons?<br><br>
ABTShield processes data related to device characteristics and online interactions to identify potentially fraudulent activity. This includes data from TCP/IP, TLS, HTTP protocols, device/application characteristics (e.g., browser version, language, time zone), and navigation events on websites (e.g., URL, event time).
Importantly, the processed data does not include information that allows identification of individual persons, and the service is not intended to identify individuals.
Data processing is based on legitimate interest (Article 6(1)(f) GDPR) to prevent fraud and ensure network security. Edge NPD (ABTShield’s provider) is a certified TCF 2.0 vendor under IAB Europe.
2
3
Can I use ABTShield to verify leads, in addition to analyzing advertising traffic?
Yes, ABTShield also offers solutions for lead verification. Its mechanisms for assessing device configuration legitimacy, detecting automation, analyzing traffic source history, and verifying contact data accuracy (e.g., email validation via SMTP testing and blacklist checks) can be used to evaluate the quality of leads obtained from contact forms or similar sources.
2
3
What support does ABTShield provide after integration is complete?
ABTShield provides ongoing support to its clients. This includes technical assistance, regular reporting, help with data analysis and result interpretation, as well as advice on campaign optimization and fraud elimination. The ABTShield team is available to answer questions and assist in maximizing the system’s capabilities.
