Accelerating the Anti Fraud System Integration Process

by Tamas Kadar

Key obstacles for businesses in need of fraud management are the integration time and costs. Let’s demystify the process.

In early 2010, an unknown startup with a strange name began making waves in Silicon Valley. Its goal was simple: to allow anyone to accept card payments.

Here is what an early adopter had to say about it on the Y Combinator forum:

“I’ve been using it for a few months and honestly it’s the best API I’ve ever used. The documentation is clear and concise. It’s customized to your account so you can literally copy and paste and see the result. Just like it says, it gets out of your way. I was up and running and accepting recurring payments in less than an hour or so. Looking forward to them eating every other payment processor’s lunch.”

You’ve guessed it: the company in question is Stripe. Fast forward a decade, and the company has indeed become one of the most trusted, powerful, and profitable payment processors on the planet.

The early adopter’s words sound more true than ever. And as they put it, the company’s success had everything to do with its ability to simplify the integration process. As we’ll see, it’s also something that can be done with fraud detection.

Fraud Prevention: Buy or Build?

Most newer online businesses experience fraud prevention as part of their default payment system. These one-stop-shops do offer a level of protection, but often fail to really cover specific use cases. They take a one-size-fits-all approach, which cannot be tailored to your own fraud needs.
You then find yourself at a crossroad: should you build a layer of protection in-house, or outsource the tech to a third party company? Each provider has its pros and cons.

Anti Fraud System Spreadsheet detailing the pros and cons of In-House and Outsourced solutions

And to add to the list against outsourced solutions: relying on a third party provider is always risky. Even something as highly regulated as payments, where trust is primordial, can backfire if things go wrong, as exemplified by the recent Wirecard failure.

But what if we told you that all the cons of outsourced fraud detection had been anticipated by more modern vendors? Let’s break down each point and see if they still stand.

Calculated Integration Complexity

In many cases, integrating a fraud detection solution is still a huge obstacle for online businesses – and for good reasons.

This is because when people think of fraud prevention, they tend to gravitate towards legacy solutions, which are convoluted by design. Their business model is to insert themselves into your business operations so that discontinuing the service would eventually become too painful to consider. And their marketing and sales budgets ensure they will convince your business that they’re the only alternative.

Certain providers also target industries like banking with multi-billion dollar yearly deals. The integrations are measured in years, providing a plethora of professional services, and complicated project timeframes.

In fact, the incentive is to drag the integration process out for as long as possible. But what if newer, innovative vendors wanted to disrupt that model?

Changing the Integration Game

In essence, the biggest challenge of integrating fraud systems is in understanding where the system should go. Should it communicate with the frontend, backend or payment system?

To answer that question, many modern fraud prevention vendors are now offering innovative plug and play solutions. And the integrations come in many shape and forms, all purpose-built with the goal of simplifying life for developers:

Public API Documentation: the more confident fraud prevention providers will let anyone peruse their API docs, to offer transparency into the integrations before purchasing their services.
Browser extension: for fraud managers who need extra help, for instance in the form of data enrichment, you can now query certain fraud providers directly from the comfort of your web browser

And best of all, businesses needn’t be entrapped by the old “rip and replace” mentality, thanks to a new approach to multi-layered protection.

Modular Risk Tools for Multi-Layered Protection

The most innovative anti fraud software companies know and understand that risk needs to be reduced with tailored tools.

This is why you will find more and more solutions breaking down their features into modules, which you can add or remove as needed, based on your business model:

Device fingerprinting: ideal for companies who need to understand connections between users. Flag attempts at browser spoofing, proxy connections and emulators, by creating an ID for each software and hardware configuration. You only need to integrate a JavaScript snippet to your web app, or iOS / Android app.
Data enrichment tools: perform social media lookup and get a user’s digital footprint based on a single datapoint. For instance, an email address, phone number or IP address is enough to build a profile by aggregating related data from external open sources.

The goal is to let fraud managers pick and choose the best tools for their needs. No need to install on a full platform. This reduces integration cost and effort, and combined with a transparent API-call-based-fee, gives your company complete control over the fraud reduction ROI.

How Long Should It Take To Integrate Fraud Prevention Software?

It depends a lot on the kind of solution you choose. As previously mentioned, integration of a legacy solution with a financial institution’s internal systems can take years, and cost hundreds of thousands.

At the other end of the spectrum, using a plugin like SEON’s Chrome Extension only takes minutes.

Download the SEON extension
Create an account with SEON
Enter an email address, IP address or phone number
Get data enrichment results in seconds

For more complex platform integrations, things can be just as seamless. Here is what our integration timeline says about getting started with the platform.

Kick off meeting: your team and ours will create a SEON demo account, and identify your use case. Do you need data enrichment or not? Which stage should we monitor to catch the most possible fraud (e.g. signup or withdrawal)?
Providing historical data: We can begin collecting historical data such as transactions, and use positive or negative labels to sort through fraudulent cases.
Data science analysis: the SEON data science team will analyse the historical data and create a report.
Technical integration: this is when we begin looking at the SEON Fraud API. We select the data fields that make sense for your industry, or use payload templates that match your specific needs to save time. You can also implement device fingerprinting, using a simple JS or SDK code.
Feedback loop: Time to adjust the scoring thresholds to fine-tune false positives and negatives. This stage also helps ensure that the API and manual labels are working properly.
Passive listening: where the teams work on fine-tuning the right rules to increase fraud detection rates. The rules are suggested by Machine Learning based on historical data, or manually created using custom parameters.
Live Mode: the anti fraud system is up and running. SEON still provides continuous support if you need help with custom rules or to make the most from the machine learning-suggested rules.

The entire process can take 1-2 days.

Of course, it’s important to remember that the longer the system runs, the more precise the results will be. But in terms of integration time and effort, a modern fraud management system won’t stop your operations, neither will it eat into your budget by creating too much business downtime.

How Does the SEON FRAUD API Look for Developers

At its core, the SEON fraud prevention platform operates in three simple steps.

You send user / transaction / device data
We enrich the data and deliver a risk score based on rules
You give feedback on the results

Regarding step 1: all the user, transaction and device data is sent via the Fraud API. Your first step is to define payloads for the API, populating it with as many relevant data points as possible. All the fields are optional, but the more you fill, the more precise our results will be.

For custom business-specific data points, use thecustom_fields object.
The config object helps you to fine-tune settings such as versions, response and aggregating data enrichment APIs, when required.
You must define the authentication points aka. action_type-s (account_register, account_login, purchase etc.) where risk assessment data can be collected or fraud should be prevented.
For device fingerprinting, you can use our JavaScript snippet for web apps, and the SDKs for iOS and Android mobile apps. Use session_id as your key for device data collection.

For more information, please see our developer page here.

The Stripe of Fraud Prevention

Circling back to our initial story about the success of Stripe, we can now see many parallels between the innovative payment provider and modern fraud prevention vendors.

At SEON, for instance, our developer-focused approach has earned us the nickname of “the Stripe of fraud prevention”, due in part thanks to:

Our Public API documentation: Lists 70+ predefined data points, and allows for unlimited custom fields. Select your point of authentication (login, withdrawal, deposit or signup…)
Modular tools: Integrate the modules you need to build your dream risk tech stack.
Whitebox decision-making: Get the reasons and rules that led to the risk score and state.
Customer success team: Free of charge, and ready to help you both with the integration process and further questions.
Fastest to go live: Our integration timeline is one of the fastest available on the market.
True proof of technology: A traditional proof of concept in fraud prevention rarely makes sense, due to the prohibitive initial investment. SEON is the only provider where the integration is so flexible that you can witness results using your data in a testing environment.
Free trial and transparent pricing: Last but not least, you can see for yourself how SEON works in practice, and calculate your ROI in advance with our clear, transparent pricing model.

Still unsure about integrating an anti fraud system into your business? You can speak to our team today, and see how SEON could benefit your business in days, not months.