TrafficGuard: Standing in the way of ad fraud to help businesses reach more customers

About TrafficGuard

Headquartered in Perth, TrafficGuard provides sophisticated fraud protection to build more trust and transparency in the digital ad ecosystem, helping customers achieve a healthy return on ad spend with real-time data.

Industries: Technology
Location: Australia

Tell us your challenge. We're here to help.

Contact us

About Kasna

Kasna "lives and breathes" Google Cloud technology, helping Australian businesses create exceptional customer experiences by guiding them on their cloud journey.

TrafficGuard helps customers detect ad fraud and provides insights on ad spend with real-time data with the support of Google Cloud.

Google Cloud results

  • Helps customers optimize digital advertising spend by making data-backed decisions
  • Builds batch data pipeline in 15 minutes instead of seven hours
  • Enables TrafficGuard to take action and mitigate ad fraud with real-time data
  • Empowers data scientists to experiment and launch new rollouts with BigQuery
  • Removes the burden of infrastructure management and maintenance, allowing TrafficGuard to focus on innovating their core offering

Gains insight from billions of ad transactions daily

Advertising is an expense that usually comes with an expected return on investment (ROI). When businesses allocate ad spend, calculated risks are a necessary part of the equation, but did you know that the cost of invalid traffic is estimated to reach $100 billion by 2023? Whether you run ad campaigns for global brands or as an independent small business owner, chances are pretty high that you're part of this statistic.

Addressing the growing ad fraud problem, TrafficGuard helps businesses to verify advertising engagement, protecting both their ad spend and the integrity of their advertising performance data. Equipped with years of ad tech experience, the TrafficGuard team created the company to fill a void in the industry. "Prior to our venture into ad verification, we were an ad network, running campaigns on behalf of businesses. We started to notice the increase in ad fraud and began building our internal capabilities to protect our clients. As there wasn't an existing solution, we developed our own technology to prevent ad fraud in real time," explains Luke Taylor, founder and Chief Operating Officer at TrafficGuard.

In 2018, Luke and his team decided to pivot from the ad network business to focus on advertising measurement, verification, and fraud prevention. Their mission? To help businesses get the clarity they need to get the best advertising performance. To address the lack of available solutions for brands to verify traffic quality, they founded TrafficGuard from ground zero.

Working with Kasna has enabled a smooth migration to Google Cloud and helped TrafficGuard benefit from the expert cloud product knowledge of specialists. "Kasna has been a great partner. They took the time to really understand our business, are process-driven, and very transparent in what they do so it has been easy to work with them," says Luke.

"As we started to look into more elaborate uses of the cloud like running analytics and machine learning (ML), we found that the alignment with Google Cloud was far better than our previous cloud provider, with Google being predominantly an ad business."

Luke Taylor, founder and Chief Operating Officer, TrafficGuard

Relying on a stable cloud infrastructure to support an agile team

TrafficGuard initially used a different cloud provider, but as their business needs evolved and their infrastructure requirements became more elaborate to accommodate ML, they found more alignment with Google Cloud. "As we started to look into more elaborate uses of the cloud like running analytics and machine learning (ML), we found that the alignment with Google Cloud was far better than our previous cloud provider, with Google being predominantly an ad business," says Luke.

By taking away the need for server infrastructure maintenance, Google Cloud enables the business to focus on last-mile ad tech solutions and verification. It also means that TrafficGuard can keep a lean team to manage its proprietary services. "We're all engineers and data scientists, so we're familiar with these platforms and anyone at the company can quickly spin up environments to get access to the data they need for testing," says Luke. "Without having to worry about infrastructure, we can spend our time building proofs of concepts to continuously iterate, innovate, and find better solutions for our customers."

"Even with information that involves a longer period of time or billions of rows of transactional data, BigQuery can process that information quickly with value-added enriched data, so it can be easily accessed by our customer."

Luke Taylor, founder and Chief Operating Officer, TrafficGuard

Managing large volumes of data seamlessly

To measure, verify, and prevent ad fraud, timeliness is key because money is lost every second an ad doesn't reach its intended audience. "It's one thing to be able to scale and connect all the different signals and datasets, but if we can't make sense of it within a reasonable time, then it loses its value," says Luke. TrafficGuard uses Dataflow and Pub/Sub to stream large amounts of data, before processing it in Cloud Bigtable or BigQuery. Now, TrafficGuard can build out a batch data pipeline in 15 minutes, whereas the prior solution took over seven hours.

Fast processing speeds with Google Cloud enable TrafficGuard to immediately collect information on each click on an ad and verify its legitimacy. "Within milliseconds of invalid-traffic detection, whether it's bots or by rival companies, TrafficGuard takes action immediately by mitigating that threat so that it's not draining our customer's budget," says Luke. This is a far cry from having to batch process data at the end of the day, or end of the week, which is what legacy anti-fraud vendors traditionally do. TrafficGuard receives feedback when anomalies are detected, mitigates invalid traffic in real-time, and gives the customer real-time visibility of both invalid traffic and the performance of their advertising without the influence of invalid traffic.

The company is equipped to collect information at scale, of up to billions of transactions daily. By doing so, customers expect TrafficGuard to also turn the insights into reports quickly, so they can make more informed decisions around ad spend. "Unlike legacy players that had to build solutions on top of older systems, we focus heavily on the real-time aspect," says Luke.

By using BigQuery, TrafficGuard can not only provide transactional data, but enriched data that includes information such as the site in which the ad was shown, or even specific keywords that were used to trigger the ad. "Even with information that involves a longer period of time or billions of rows of transactional data, BigQuery can process that information quickly with value-added enriched data, so it can be easily accessed by our customer."

With so many moving pieces, Google Data Studio gives the TrafficGuard team a way to easily access analytics. Raigon Jolly, Head of Analytics AI and Data Engineering at TrafficGuard says, "Data Studio is a nice integration because we don't need to create a separate cluster for it to work. It just sits on top of what we are already running and is ready to use." The free data visualization tool is beneficial for the entire team, fostering a data-driven culture across the business.

Standing man points out something on laptop screen to two sitting women

Enabling a more efficient workflow with a scalable infrastructure

With their previous cloud vendor, compute, streaming services, and storage all required manual scaling. This added complexity to workloads, slowing innovation.

Being able to process streaming data and then use BigQuery for compute, storage and ML simplifies a lot of TrafficGuard’s processes. “We spend less time managing infrastructure and more time innovating the ways we verify advertising, prevent invalid traffic, and bring more sophisticated ML to streaming and compute," says Luke.

With large volumes of data flowing through the pipelines 24/7, it's important for TrafficGuard to always be able to process data and provide real-time verification and reporting, even in the event of advertising traffic spikes. One of the benefits of Google Cloud is the flexibility of only paying for what you use. Previously, compute was inefficient because it was often overprovisioned. Because the storage component of Google Cloud is separate from the compute side, TrafficGuard now only pays for the events it runs, keeping costs low.

"We're big on experimentation and are always innovating to improve our products. By the time the migration is complete, we should be 90% on Google Cloud, but with so many innovative products, I believe it's only a matter of time before this becomes a 100% migration."

Luke Taylor, founder and Chief Operating Officer, TrafficGuard

Building smarter features with ML

As a fraud detection platform, TrafficGuard relies on a set of signals to flag anomalies. Raigon explains, "These signals can change, and we need our models to adapt and to be able to constantly tune certain parameters and monitor for concept and data drift." For example, people's online behaviors changed during COVID-19 when they worked from home. Their online habits during this period may have been considered an anomaly, but over time, this anomaly is becoming a norm. As such, the systems have to recognize that.

TrafficGuard plans to rely on the ML capabilities of BigQuery and AI Platform to help solve some of these problems and improve its product offering. "BigQuery is not just a database. It has capabilities for ML, streaming, and reporting as well as being easy to run Spark analytics with Dataproc and connect to AI Platform for advanced ML. We want to leverage that to help us do more," says Raigon. He shares that the team can now experiment inside BigQuery and carry out new rollouts within the day. They can also schedule jobs, eliminating the need of a separate component.

Even though it is still in the midst of migration to Google Cloud, the company is already experimenting with Compute Engine and Memorystore to improve the ad monitoring journey to provide more enrichment to the data it generates.

"We're big on experimentation and are always innovating to improve our products,” says Luke. “By the time the migration is complete, we should be 90% on Google Cloud, but with so many innovative products, I believe it's only a matter of time before this becomes a 100% migration."

Person types on laptop on desk

Tell us your challenge. We're here to help.

Contact us

About TrafficGuard

Headquartered in Perth, TrafficGuard provides sophisticated fraud protection to build more trust and transparency in the digital ad ecosystem, helping customers achieve a healthy return on ad spend with real-time data.

Industries: Technology
Location: Australia

About Kasna

Kasna "lives and breathes" Google Cloud technology, helping Australian businesses create exceptional customer experiences by guiding them on their cloud journey.