Fraud detection challenged by new fraud types – get resilient with data

Digital companies face a new era of fraud. This article looks at fraud types that might silently erode digital budgets.

Digital companies face a new era of fraud. In this article, we look at fraud beyond financial transactions. “Soft fraud” is about loopholes in marketing incentives or policies, rather than the typical “hard” definitions of payment or identity fraud. The goal is to look at fraud that could silently happen to you and how to address it with data. Lastly, we check what is needed for successful fraud detection with machine learning.

Many companies transform digitally to stay ahead of the curve. At the same time they expose themselves in a digital ecosystem. As digital presence grows, so does the surface area that attracts malicious actors. “The crime of getting money by deceiving people” according to the Cambridge Dictionary takes many forms when you deceive systems instead of people. Once fraudsters identify a loophole, they scale their approach with bots leading to substantial financial loss. This likely explains why fraud and debt analytics ranks among the top ten AI use cases according to McKinsey’s state of AI report

Soft fraud

Fraud that is less clear-cut from a legal perspective involves bad actors that systematically exploit loopholes within usage policies, marketing campaigns or products. We could refer to it as soft fraud:

Bad actors systematically misuse policies, products or services to divert money or goods from the digital ecosystem to themselves.

So, what forms can soft fraud take?

Photo by Noelle Otto

Digital marketing giveaways. The digital economy offers a vast range of services, and so does it offer endless possibilities for fraud. One of the biggest areas is digital marketing. It gets attacked from two sides: Humans and algorithms that mimic human behavior, also known as bots. Both try to exploit usage policies, ad campaigns or incentive schemes. For example, a customer creates accounts to claim sign-up bonuses, also called sign-up fraud. Another one involves a customer that uses a product once and yet returns it, referred to as return fraud. Sharing accounts across friends or family is a famous example for companies like Netflix. Non-human actors, like bots, click on paid-ads or exploit affiliate schemes to claim rewards, such as a payout for each new customer registration. 

Humans reap bonuses. Most of the traffic still comes from humans, estimated around 60%. They become interested in your product and explore your digital offering. Some try to take advantage of promotional schemes such as newsletter sign-up bonuses, giveaways or related incentives. They reap bonuses multiple times, for example by using generic email addresses. Others try to push boundaries on usage policies. For example, when multiple persons use one account or share content protected by paywall. With a genuine interest in your product, they count as “friendly fraudsters”, happily using blind spots in web-tracking or marketing campaigns. Those customers invest time to access your products. So, they reveal a strong preferences for your offering. Rigorously blocking them to bring down fraud may hit innocent customers as false positives. Additionally it kills the potential to re-engage with previous fraudsters in a more secure way. That is why in the world of fraud detection, experts refer to it as the “insult rate”.

Bots dilute metrics. Up to estimated 40% of website traffic comes from bots. They click ads, fill out web forms and reap giveaways. The entire lifecycle of digital marketing gets compromised. Bots dilute key performance metrics which leave you wondering about low conversion rates, high cost-per-click or low lead quality. They negatively impact key metrics such as cost per acquisition (CPA), customer lifetime value (LTV), cost per click (CPC), marketing qualified leads (MQL), etc.

Adapt fraud detection to these types

Photo by lil artsy

Below you find a list that provides an overview about fraud types you can encounter. It divides into non-human actors like bots, human actors like users and eventually both. It includes anyone who gets incentivized by your digital presence to commit fraud.

Non-human actors like bots

  • Click fraud: Viewing ads to get paid per click.
  • Inventory fraud: Buying limited goods like sneakers or tickets and holding inventories.
  • Fake account creation: Registering as users to dilute the customer base.
  • Campaign life-cycle fraud: Competitors deploy bots which eat up marketing budgets.
  • Lead generation fraud: Filling out forms to sabotage sales efforts

Human-only actors like customers or competitors

  • Multi-account usage: Different persons use a personalized account.
  • Return fraud: Customer uses product and returns it damaged
  • Bonus fraud: Get discounts multiple times after newsletter sign-up or account registration.
  • Account takeover: Leaked login details or weak user authentication
  • Friendly fraud: Customers receive a product, dispute the purchase and chargeback the money

Either human or non-human

  • Affiliate fraud: Bots click exploit a strategy in affiliate campaigns to unlock compensation
  • Bad-reputation fraud: An attack on your product reviews from competitors

Some of these can be tackled with data analytics and possibly machine learning, while some are more about designing policies and services in a safer way, so that they cannot be easily exploited.

Effective fraud detection builds on data

Now that we have seen different types of fraud, what can we do about it? Do we want to detect them when they happen, or do we want to prevent them from happening at all? Let us see how data & analytics can help us.

Leverage machine learning. Fraud tends to happen systematically. Systematic actors need a systematic response. If your data captures these patterns and lets you identify fraud, you have everything to build effective solutions with rules, heuristics or eventually machine learning. Machine learning is an approach to learn complex patterns from existing data and use these patterns to make predictions on unseen data (Huyen, C., 2022. Designing Machine Learning Systems). 

Rephrasing this from a business perspective would lead to  the starting question for machine learning:

Do you face a (1) business-relevant and (2) complex problem which can be (3) represented by data?

  1. Business-relevance: Can you become more profitable
  2. Complexity: Is data available in volume or complexity that heuristics likely fail?
  3. Data representation: Is data extensive and consistent enough for a model to identify patterns?

Machine learning requires detailed and consistent data to make it work. There is no silver bullet.

Identify fraud in data. Preventing fraud comes down to data. How well you can track web sessions, impressions and click paths becomes central in dealing with fraud. Without tracking data, chances are low to do anything about it. Even third-party anti-fraud software might be ineffective since it solves generic use cases by design. Different firms attract different fraud types. Third party solutions cannot possibly know the specifics on a complex range of products or services and their vulnerabilities. Therefore, a tailored approach built together with internal domain experts such as product or marketing could effectively prevent fraud.

Machines or humans. One major challenge is to differentiate between bots and humans. Nowadays, bots have become better at mimicking human behavior. At worst they come in thousands to interact with whatever incentive you expose to the outside world. Due to the sheer traffic volume it is infeasible to manually analyze patterns. You have to fend off algorithms with algorithms. The depth of data you have, directly determines whether you have any chance to deploy machine learning.  

Honeypots for bots. One way to label bots is to use so-called honeypots to lure bots. Honeypots are elements on your website invisible to humans, like hidden buttons or input forms. Bots scrape the website source-code to discover elements they can interact with. If your website tracking logs an interaction with these hidden elements, you clearly identify bots. You can see a summary of the honeypot method in this article by PerimeterX: How to use honeypots to lure and trap bots.

As bots act more like humans, their digital footprint blends in with anyone else’s. This poses a major challenge to any data-driven solution and there is no magic solution to that. Creating honeypots that lure bots could be one way forward. Along the lines of Gartner’s Market Guide for Online Fraud Detection, a vendor on bot detection would be the safest bet, such as Arkose Labs, Imperva, GeeTest or Human to name a few.


This article talks about the rise of novel fraud types that modern fraud detection faces. Firms increasingly expose their offerings in the digital ecosystem which leads to losses due to fraud. Policy loopholes and marketing giveaways erode their digital budgets. For example, customers reaping signup bonuses multiple times with generic emails on the one hand, and sophisticated bots creating fake accounts that dilute your customer base on the other hand. Both forms lead to losses along the digital supply chain. 

I personally find the world of fraud detection fascinating. It constantly changes where preventive technology and creative fraudsters move in tandem. With the rise of bots, fraud detection becomes more complex and difficult to do with conventional approaches. If you start on your fraud detection journey, I recommend you start thinking about how your company’s digital presence is reflected by the data you have. Web tracking needs to be deep enough to enable analytics or even machine learning. 

At Solita we have the skillset to both build strategic roadmaps and create data solutions with our team of data experts. Feel free to reach out how we can help you on the data groundwork towards effective fraud detection. 

How to choose your next machine learning project

Three steps to be intentionally agnostic about tools. Reduce technical debt, increase stakeholder trust and make the objective clear. Build a machine learning system because it adds value, not because it is a hammer to problems.

As data enthusiasts we love to talk, read and hear about machine learning. It certainly delivers value to some businesses. However, it is worth taking a step back. Do we treat machine learning as a hammer to problems? Maybe a simple heuristic does the job with substantially lower technical debt than a machine learning system.

Do machine learning like the great engineer you are, not like the great machine learning expert you aren’t.

Google developers. Rules of ML.

In this article, I look at a structured approach to choose the next data science project that aligns to business goals. It combines objective key results (OKR), value-feasibility and other suggestions to stay focused. It is especially useful for data science leads, business intelligence leads or data consultants.

Why data science projects require a structured approach

ML solves complex problems with data that has a predictive signal for the problem at hand. It does not create value by itself.

So, we love to talk about Machine learning (ML) and artificial intelligence (AI). On the one hand, decision makers get excited and make it a goal: “We need to have AI & ML”. On the other hand, the same goes for data scientists who claim: “We need to use a state-of-the-art method”. Being excited about technology has its upsides, but it is worth taking a step back for two reasons.

  1. Choosing a complex solution without defining a goal creates more issues than it solves. Keep it simple, minimize technical debt. Make it easy for a future person to maintain it, because that person might be you.
  2. A method without a clear goal fails to create business value and erodes trust. Beyond the hype around machine learning, we do data science to create business value. Ignoring this lets executives reduce funding for the next data project.

This is nothing new. But, it does not hurt to be reminded of it. If I read about an exciting method, I want to learn and apply it right away. What is great for personal development, might not be great for the business. Instead, start with what before thinking about how.

In the next section, I give some practical advice on how to structure the journey towards your next data project. The approach helps me to focus on what is next up for the business to solve instead of what ML method is in the news.

How to choose the next data science project

“Rule #1: Don’t be afraid to launch a product without machine learning.”

Google developers. Rules of ML.

Imagine you draft the next data science cases at your company. What project to choose next? Here are three steps to structure the journey.

Photo by Leah Kelley from Pexels

Step 1: Write data science project cards

The data science project card helps to focus on business value and lets you be intentionally agnostic about methodologies in the early stage

Summarize each idea in a data science project card which includes some kind of OKR, data requirements, value-feasibility and possible extensions. It covers five parts which contain all you need to structure project ideas, namely an objective (what), its key results (how), ideal and available data (needs), the value-feasibility diagram (impact) and possible extension. What works for me is to imagine the end-product/solution to a business need/problem before I put it into a project card.

Find the project card templates as markdown or powerpoint slides.

I summarize the data science project in five parts.

  1. An objective addresses a specific problem that links to a strategic goal/mission/vision, for example: “Enable data-driven marketing to get ahead of competitors”, “Automate fraud detection for affiliate programs to make marketing focusing on core tasks” or “Build automated monthly demand forecast to safeguard company expansion”.
  2. Key results list measurable outcomes that mark progress towards achieving the objective, for example: “80% of marketing team use a dashboard daily”, “Cover 75% of affiliate fraud compared to previous 3 month average” or “Cut ‘out-of-stock’ warnings by 50%, compared to previous year average”.
  3. Data describes properties of the ideal or available dataset, for example: “Transaction-level data of the last 2 years with details, such as timestamp, ip and user agent” or “Product-level sales including metadata, such as location, store details, receipt id or customer id”.
  4. Extensions explores follow-up projects, for example: “Apply demand forecast to other product categories” or “Take insights from basket analysis to inform procurement.”
  5. The value-feasibility diagram puts the project into a business perspective by visualizing value, feasibility and uncertainties around it. The smaller the area, the more certain is the project’s value or feasibility.

To provide details, I describe a practical example how I use these parts for exploring data science projects.  The journey starts by meeting the marketing team to hear about their work, needs and challenges. If a need can be addressed with data, they become the end-users and project target group. Already here, I try to sketch the outcome and ask the team about how valuable it is which estimates the value.

Next, I take the company’s strategic goals and formulate an objective that links to them following OKR principles. This aligns the project with mid-term business goals, makes it part of the strategy and increases buy-in from top-level managers. Then I get back to the marketing team to define key results that let us reach the objective.

A draft of an ideal dataset gets compared to what is available with data owners or the marketing team itself. That helps to get a sense for feasibility. If I am uncertain about value and feasibility, I increase the area in the diagram. It is less about being precise, but about being able to compare projects with each other.

Step 2: Sort projects along value and feasibility

Value-feasibility helps to prioritize projects, takes a business perspective and increases stakeholder buy-in.

Ranking each project along value and feasibility makes it easier to see which one to prioritize. The areas visualize uncertainties on value and feasibility. The larger they stretch along an axis, the less certain I am about either value or feasibility. If they are more dot-shaped, I am confident about a project’s value and its feasibility.

Projects with their estimated value and feasibility

Note that some frameworks evaluate adaptation and desirability separately to value and feasibility. But you get low value when you score low on either adaptation or desirability. So, I estimate the value with business value, adaptation and desirability in my mind without explicitly mentioning it.

Data science projects tend to be long-term with low feasibility today and uncertain, but potentially high future value. Breaking down visionary, less feasible projects into parts that add value in themselves could produce a data science roadmap. For example, project C which has uncertain value and not feasible as of today, requires project B to be completed. Still, the valuable and feasible project A should be prioritized now. Thereafter, aim for B on your way to C. Overall, this overview helps to link projects and build a mid-term data science roadmap.

Related data science projects combined to a roadmap

Here is an example of a roadmap that starts with descriptive data science cases and progresses towards more advanced analytics such as forecasting. That gives a prioritization and helps to draft a budget.

Step 3: Iterate around the objective, method, data and value-feasibility

Be intentionally agnostic about the method first, then opt for the simplest one, check the data and implement. Fail fast, log rigorously and aim for the key results.

Implementing data science projects has so many degrees of freedom that it is beyond the scope of this article to provide an exhaustive review. Nevertheless, I collected some statements that can help through the project.

  1. Don’t be afraid to launch a product without machine learning. And do machine learning like the great engineer you are, not like the great machine learning expert you aren’t. (Google developers. Rules of ML.)
  2. Focus on few customers with general properties instead of specific use cases (Zhenzhong Xu, 2022. The four innovation phases of Netflix’ trillions scale real-time data infrastructure.)
  3. Keep the first model simple and get the infrastructure right. Any heuristic or model that gives quick feedback suits at early project stages. For example, start with linear regression or a heuristic that predicts the majority class for imbalanced datasets. Build and test the infrastructure around those components and replace them when the surrounding pipelines work (Google developers. Rules of ML. Mark Tenenholtz, 2022. 6 steps to train a model.)
  4. Hold the model fixed and iteratively improve the data. Embrace a data-centric view where data consistency is paramount. This means, reduce the noise in your labels and features such that an existing predictive signal gets carved out for any model (Andrew Ng, 2021. MLOps: From model-centric to data-centric AI).
  5. Each added component also adds a potential for failure. Therefore, expect failures and log any moving part in your system.
  6. Test your evaluation metric and ensure you understand what “good” looks like (Raschka, 2020. Model Evaluation, Model Selection, and Algorithm Selection in Machine Learning.)

There are many more best practices to follow and they might work differently for each of us. I am curious to hear yours!


In this article, I outlined a structured approach for data science projects. It helps me to channel efforts into projects that fit business goals and choose appropriate methods. Applying complex methods like machine learning independent of business goals risks accruing technical debt and at worst jeopardizes investments.

I propose three steps to take action:

  1. Write a project card that summarizes the objective of a data science case and employs goal-setting tools like OKR to engage business-oriented stakeholders.
  2. Sort projects along value and feasibility to reasonably prioritize.
  3. Iterate around the objective, method, data and value-feasibility and follow some guiding industry principles that emerged over the last years.

The goal is to translate data science use cases into something more tangible, bridging the gap between business and tech. I hope that these techniques empower you for your next journey in data science.

Happy to hear your thoughts!

Materials for download

Download the data science project template, structure and generic roadmap as Power Point slides here. You can also find a markdown of a project template here.