IndustryFraud Detection
ServiceData & Analytics

Traffic monitoring and fraud detection enhancements

ServiceData & Analytics
IndustryFraud Detection

Traffic monitoring and fraud detection enhancements

Meet Lunio

Lunio see the Pay-per-click (PPC) advertising ecosystem as being in a broken state, where budgets are swallowed up by opaque, black-box driven platforms that no one can hold accountable.

Lunio look at things differently. Showing advertisers exactly what they’re spending their money on and arming them with all the data they need to take back control. They offer their customers total honesty, transparency and privacy, eliminating click fraud across all channels.

Upgraded traffic monitoring POC

Mitigating PPC fraudulent clicks

Business challenges

Lunio’s offering delivers an unrivalled approach to fraudulent PPC click mitigation. As their service incorporates data from over 25 Ad networks (Facebook, Google, LinkedIn, Reddit, TikTok and so on), as well as offering near real-time analytics that can scale to an enterprise level, they were faced with multiple challenges.

Their main obstacle was that they were unable to process more than 10,000 requests per second. They needed to be able to handle more requests per second, in a way that was cost effective against their fraud detection platform.

Why Firemind?

Lunio had begun their cloud journey with a separate provider. Unfortunately, they were not satisfied with the state of the project and the advances that were being made.

It was at this point that they reached out to Amazon Web Services, in search of a partner that could better deliver on their requirements, and enable and more comprehensive and personalised service.

After some introductory calls, Lunio chose Firemind as their preferred cloud consultancy, teaming up with us to work through a Proof-of-Concept that would enable higher traffic and cloud requests.

The project with Lunio began with a concentrated scoping of requirements. We knew that their previous interaction with another consultancy had remained lack lustre, so we set off to ensure we covered the development, review stage, documentation timeline and production pricing estimates, early on.

In its simplest explanation, the PoC needed to provide the Lunio team with data to analyse, then send a response back to a customer’s website. This process had to happen thousands of times a second, to ensure validity of data transferred for any active customer campaigns.

The PoC worked through the following workflow, utilising a range of core AWS and cloud services:

  • Amazon CloudFront – Served a 1×1.gif file from an Amazon S3 bucket.
  • AWS Lambda@Edge – The Lambda@Edge was written in Python, and connected to a DynamoDB that uses a single table design and contains shared application states. DynamoDB checks the request host, validates it against a set of allowed hosts, then returns a suitable CORS header for pre-flight requests. This generates a return, with a random number between one of three possible HTTP status codes: 200, 201 and 202. Lunio will utilise this response logic and expand on the code for their own purposes.
  • Amazon Kinesis – An Amazon Kinesis Data Stream takes the log requests from CloudFront, then passes it through Kinesis Firehose for record modification.
  • VPC based AWS Lambda – This allows record modification, outputting logs into Parquet format (Lunio will expand on this code further).
  • Output storing – Outputs will be stored in a standard “Year, Month, Day, Minute’ format, sharding into S3.
  • Output storing – A simple HTML page that includes a JavaScript ‘fetch’ example that calls the 1×1.gif and outputs console.log, based on the response header.

Added value

Firemind’s PoC was able to quickly replicate the current workflow of Lunio’s, at a more rapid and scalable pace. The PoC demonstrated how a refined and modernised architecture could allow for more than 10,000 requests per second, broadening data access to each of their customers.

60,000 + requests

Compared to the previous 10,000.

Time to value

Within 12 days, we had successfully onboarded Lunio, undergone a thorough discovery and created the working PoC. This streamlined timeframe allowed us to thoroughly test the new infrastructure and ensure it was built with best practice ion mind.

Scaled requests

As Lunio’s business is based on thousands of requests per second, the need to ensure accurate data delivery of more than 10,000 requests was a necessity for the business to expand its services. With this PoC being introduced to their cloud architecture, Lunio can be sure that their risk mitigation services continue to perform across thousands of campaigns, customers and platforms.

Firewall readiness

By using Firemind’s AWS experience, Lunio can benefit from multiple managed rules for Web Application Firewalls (WAF). Having WAF within the architecture ensures agile protection against web attacks, with less than a minute needed to roll out security updates and rules across environments.

Customer satisfaction

“Firemind are a professional partner and have been a great joy to work with. They have helped us solve several cloud architecture challenges, and helped train the team to conquer future obstacles. They are dedicated to deliverables and quality execution, providing a wonderful team.”

David Triger

Chief Technology Officer

How can we help you?

Looking to become Cloud Native? Or looking to modernise your cloud architecture?

Get in touch