Case Studies

Riley + Publift: Maximizing ad revenue for online publishers with cloud-driven programmatic advertising

Publift delivers 55% boost in client ad revenue, increases customer base by 150% in one year, and saves 30% in manpower hours per week with Google Cloud, using CloudCDN, CloudFunctions, BigQuery and DataStudio.

Google Cloud Results

- Frees up developers time to grow business offering by saving 30% of manpower hours per week

- Enables global reach through millisecond-level latency in content delivery network

- Promotes cost-effective operations through intuitive infrastructure that eliminates the need for a dedicated DevOps team

- Enables innovation for global growth with ML toolkit to power customer performance analysis

- Delivers scalability to absorb 150% client base surge

The internet today is a sea of 1.7 billion websites and counting. At the same time, former Google advertising executives Colm Dolan and Tobin Page had a front-row seat on the struggles of online publishers (websites and bloggers) to attract quality advertising to monetize their content.

The experience led the friends to launch Publift, an ad-tech startup that applies the new field of programmatic advertising to empower digital publishers to optimize revenue while focusing on content creation. Programmatic advertising levels the online advertising playing field by creating an automated marketplace for advertisers to bid on website ad space, instead of seeking agency-mediated deals. By opening the market to greater competition, programmatic advertising enables publishers to secure higher rates and better ads.

"Our vision is to bring unique value to programmatic advertising with cloud-based innovation and scaling power that gives publishers access to hard-to-reach advertisers. We were able to create a user-friendly engine to crack open the online advertising marketplace with a combination of Cloud CDN, Cloud Functions, BigQuery, and Data Studio."

Joe Wongchanglor, Head of IT, Publift

Colm and Tobin aimed to offer unique value through its proprietary platform called Fuse that maximizes the number of advertisers competing for ad space, thus optimizing the publisher's revenue potential. To succeed, Publift needed unlimited scalability and global reach, to keep pace with demand spikes and connect buyers and sellers anywhere in the world. It also needed world-class data analytics to crunch the vast amounts of bidding data coursing through the system.

Publift turned to Google Cloud and discovered the ideal toolkit to open a new chapter in online publishing. "Our vision is to bring unique value to programmatic advertising with cloud-based innovation and scaling power that gives publishers access to hard-to-reach advertisers," says Joe Wongchanglor, Head of Technology, Publift. "We were able to create a user-friendly engine to crack open the online advertising marketplace with a combination of Cloud CDN, Cloud Functions, BigQuery and Google Data Studio."

Fostering a bootstrapped startup's growth with intuitive cloud architecture and proactive support

Colm and Tobin set out five years ago to disrupt digital advertising with a small team of developers that had little expertise in managing cloud infrastructure. They were also a bootstrapped startup that relied on platform revenue for growth rather than venture capital. On both fronts, says Joe, Google Cloud provided key support to get Publift's journey off the ground.

The team began life on a rival cloud provider and soon found it was spending more time on infrastructure tasks, and more credits on cloud services, than it could afford. The picture changed when Publift won an introduction to the Google Cloud for Startups program, which empowers qualified startups with mentorship, training and cloud credits, and decided to give it a try.

Unlike the original provider, the Google Cloud program fostered Publift's venture by supplying credits as long as they were used up within a finite period of time. On the operational side, the team found Google Cloud's infrastructure services to be more intuitive, with automated features that required no additional work from the team.

"The startup program liberated us from the anxiety of how much we were spending on developing the platform, enabling us to innovate and build more into our platform," recalls Joe. "Likewise, the intuitive interface and automation on Google Cloud freed our team to do what it does best, which is to create code."

Joe estimates that the simple switch to a more user-friendly Google Cloud infrastructure saved the team up to 30% manpower hours per week on maintenance tasks. Three years on, Publift is running an award-winning platform that enables its clients to enjoy 55% average revenue growth after six months of using the Fuse platform.

Throughout this journey of success, Publift didn't need to invest in a separate DevOps team to support the infrastructure. With partner Riley, which provides guidance on backend optimization in monthly catch-ups, Publift got to enjoy a priceless cost advantage. For example, Riley recently provided Publift with advice on tweaking BigQuery settings to bring down data warehouse spend, and those savings are expected to roll in within upcoming months.

"We get together with Riley on a monthly basis to go through our system and talk through pain points, optimization, and cost control," says Joe. "Along with the user-friendly architecture of Google Cloud, Riley is another reason we don't need to build any dedicated infrastructure team."

"Without a massive, consistent content delivery network, we wouldn't be able to deliver our business at all. On Cloud CDN, our solution can instantly access the global advertising marketplace and deliver our code with minimum latency on planetary scale."

Joe Wongchanglor, Head of IT, Publift

Reaching premium advertisers around the globe with a world-class content delivery network

Publift's success depends on reliable delivery of its vast repository of JavaScript files anywhere in the world, with millisecond-level latency. That means a robust content delivery network is critical to ensure that real-time bidding proceeds without glitches that could break a deal.

Publift chose Cloud CDN after testing it against five rival offerings. Google Cloud came out on top both for speed and reliability, due to Google's global network of data centers.

"Without a massive, consistent content delivery network, we wouldn't be able to deliver our business at all," says Joe. "On Cloud CDN, our solution can instantly access the global advertising marketplace and deliver our code with minimum latency on planetary scale."

Seamless serverless architecture to enable nimble and flexible growth

Since beginning its Google Cloud journey on virtual machines (VMs), Publift has switched to an entirely serverless architecture that lives on Cloud Functions and Cloud Run. These two solutions enable Publift to deploy highly scalable applications affordably and flexibly.

"If some business requirement changes, we can quickly run new code on Cloud Functions, or modify existing services on Cloud Run, without having to dig through deep layers of code or complex dependencies," says Joe. These two solutions allow Publift to rapidly respond to sudden changes in commercial requirements.

Real-time scaling enabled by Google Cloud's serverless infrastructure has also enabled Publift to absorb a more than 150% surge in client base, from about 100 to over 250 in the past year.

Unlimited data analytics power to provide instant readouts of buyer-seller requirements

Data analytics are a critical part of the Fuse solution, integrating data points of marketplace players to automatically find the best matches. Publift deploys BigQuery as its core data analytics solution because of its ability to process datasets of unlimited size and provide instant data-driven readouts of buyer-seller needs.

It can then plug Data Studio directly into BigQuery to obtain easy-to-read dashboarding of business performance, both for Publift's team and its clients.

"BigQuery makes life a lot easier for us," says Joe. "It not only allows us to work with massive amounts of data. It's also so intuitive to use that we can continue working with a bare-bones team without having to build a dedicated data engineering team."

"The next chapter of our journey will be developing innovations that bring us more insight into how we can best optimize client revenue. So we're excited to be in a Google Cloud environment that puts cutting-edge AI tools at our fingertips to build these solutions for our customers."

Joe Wongchanglor, Head of IT, Publift

A future of global growth building powerful advertising CRM with machine learning tools

The onset of COVID-19 hit the online advertising industry hard. It also served as a catalyst for industry players to leverage the automated matchmaking offered by programmatic advertising, as a means to weather the storm.

Publift soon found its services in greater demand, as publishers sought more efficient ways to earn ad revenue. That has led it to accelerate plans for expansion beyond Australia and New Zealand, into Asia and eventually the US and Europe.

To support global growth, Publift is planning an overhaul of its cloud infrastructure to enable unlimited, low-latency scalability at planetary level. That means adopting Google Kubernetes Engine to manage infrastructure and Cloud Load Balancing to optimize seamless autoscaling in content delivery, integrated with Cloud CDN.

Cloud Bigtable, meanwhile, will bring Publift a fully managed, scalable database that can provide latency below 10 milliseconds on millions of requests per second.

Alongside fostering new markets, Publift plans to enhance its value by developing a powerful data analytics solution to optimize client performance. The tool will provide intuitive, data-driven analysis of revenue fluctuations, deploying Google Cloud machine learning solutions such as BigQuery ML.

"The next chapter of our journey will be developing innovations that bring us more insight into how we can best optimize client revenue," says Joe. "So we're excited to be in a Google Cloud environment that puts cutting-edge AI tools at our fingertips to build these solutions for our customers."