How is machine learning driven by experimentation?

By on March 6th, 2020 in Machine Learning, Product and Technology
microscopes in a laboratory

Building scalable technology requires constant evaluation and improvement. Experimenting is defined by trying new things and creating effective changes that help teams to make informed decisions around product development. Trying new things creates momentum, and organizations that are driven by experimentation turn that momentum into growth.

Machine learning and artificial intelligence support large-scale, concurrent experimentation that helps these technologies to improve upon themselves. With the right tools in place, you can test a variety of scenarios simultaneously.

For example, we use our systems to track changes in the collection process and better understand how our digital collections efforts can be improved. Since digital-first channels offer thorough tracking and analysis, including real-time tracking on our website, we can learn in short cycles and continuously improve our product. 

This kind of frequent experimentation helps to avoid making product development decisions based on untested hunches. Instead, you can test your instincts, measure them carefully, and invest energy where it matters.

Machine learning drives the experimentation engine

Aggregating historical data and processing it using machine learning algorithms and artificial intelligence helps you to understand their effectiveness. Regardless of how intelligent your learning algorithms may be, waiting to test and expand your knowledge base before marching blindly ahead can make or break the success of your product.

To launch an experiment, we follow these steps: 

  1. Start with a hypothesis that you want to test
  2. Assign a dedicated team to manage the experiment
  3. Monitor the performance of the test as it is guided by machine learning
  4. Iterate

B2B companies can benefit from partnering directly with clients to customize experiments for their unique product lines in order to make experimentation-based optimization an ongoing process for both new and existing business. Keep in mind that the goal of product optimization is not always jumping to the finish line. 

Understanding how your product works ultimately offers you and your customers more value, but it’s easy to become distracted by positive outcomes. Effective, scalable products require intentional design; if you’ve accomplished a goal, but the path there was accidental, taking a few steps back to review that progress and test it can help you to get a clearer picture and grow the way you want. 

Below are two sample experiments we conducted to optimize our machine learning algorithms. 

Experiment #1: Aligning Payments to Income

Issue

The number one reason payment plans fail is consumers don’t have enough money on their card or in their bank account. 

Hypothesis

If you align debt payments with paydays, consumers are more likely to have funds available, and payment plan breakage is reduced. 

Experiment

We tested three scenarios: a control, one where we defaulted to payments on Fridays, and one where consumers used a date-picker to align with their payments with their payday. After testing and analysis, we determined that the date-picker approach was the most effective as measured by decreased payment plan breakage without negatively impacting conversion rates.

By understanding which payment plan system was the most effective, we were able to provide our AI content that offered these plans as options to more consumers and integrate the knowledge back into our systems and track those improvements at a larger scale!

Experiment #2: Longer payment plans can re-engage consumers

Issue

Customers dropped off their payment plans and stopped replying to our communications.

Hypothesis

Customers can be enticed to sign up for a new plan if offered longer payment plan terms. 

Experiment

We identified a select group of non-responsive consumers that had broken from their payment plans and sent them additional text messages and emails. These additional messages offered longer payment plan terms than the plans they broke off from.

Ultimately, we found that offering longer payment plans, even with reference to the consumer’s specific life situations didn’t lead to an increase in sign-ups. The offers that we sent had high open and click rates but did not convert. This indicated that we were on the right track but needed to iterate and come up with another hypothesis to test.

This experiment was especially important because it illustrates that not every hypothesis is proven to be correct, and that’s okay! Experimentation processes take time, and the more information you can gather, the better your results will be in the future.

We’re able to simultaneously update our product and continue experimenting, thanks to algorithms called contextual or multi-armed bandits. Here’s what you need to know about these algorithms and how they help!

Building the newest, most innovative products feels exciting, but building without carefully determined direction can be reckless and dangerous. By regularly evaluating the effectiveness of machine learning algorithms, you can make conscious updates that lead to scalable change, and experimentation paves the way for consistent product improvement.

5 tips for building scalable email infrastructure

By on February 6th, 2020 in Product and Technology

Using email as a channel for consumer communication seems like a simple way to dive into the digital revolution, but internet service providers (ISPs) actively develop tools to combat spam and abuse.

You may have the best intentions, but these service providers want to help consumers feel like they are protected which means blacklisting and filtering out junk mail. Unfortunately, emails sent by the untrained email sender can veer dangerously close to junk. 

This can make breaking into emailing consumers difficult, but it makes sending emails by the thousands (and millions) impossible without building email infrastructure that is sustainable and scalable. Establishing that infrastructure begins with recognizing the challenges you might face and then considering how to best confront them.

Why scaling email infrastructure is difficult

Email communication is heavily regulated by automated filters and systems in a way that more manual forms of communication aren’t. Cell service providers, for example, do not have nearly as much control over the volume or quality of calls that their customers receive. 

ISPs have dedicated engineers that design algorithms to keep their users happy, engaged, and protected from malicious senders, and an inbox packed with spam mail makes for a poor user experience. These algorithms are not perfect, and when they are designed, they lean on the side of being more restrictive than less which can lead to some misunderstanding. They may accidentally filter out an email from a legitimate sender that, according to their understanding of what is deemed safe, seems suspicious.

To make matters more complicated, each ISP has unique criteria that serve as the basis of their filtering rules. An email that is flagged as spam by Google could land safely in a Yahoo Mail inbox and vice versa. These rules are also constantly changing and updating to fight back against more advanced scammers making it impossible to create a one-and-done solution to properly sending emails at a massive scale.

Here are just a few things that spam filters analyze that you’ll need to consider:

  • Content: What do your emails say? Do you have any suspicious attachments or links?
  • Design: How do your emails look?
  • Sending time: Did your email arrive at 4pm or 4am?
  • Sending volume: How many of these emails did you send out at once?
  • Sending frequency: How often are you trying to email people?
  • Consumer engagement: Is anyone actually opening/clicking your emails?

Working to get all of these answers (and more) right is essential or you might find your email domain permanently blacklisted from one or all of the ISPs that you’re sending to. So what can you do to build a scalable infrastructure and work within these restraints?

How to successfully send email at scale

As we mentioned above, there isn’t necessarily a single, perfect solution for overcoming the innumerable hurdles to large-scale emailing. It takes dedicated and focused strategy to improve your long term inbox placement rates. Here are a few tips that our team keeps in mind as we continue to grow.

Create valuable content

The first step to making sure your emails are well-received by both users and ISP filters alike is creating the right content. Well-designed UX and carefully curated text are important, but it’s equally important that you steer clear of some phrases and keywords and trigger red flags.

Here’s a list of some spam trigger words that you might want to avoid!

Having a dedicated content team gives you the flexibility to create more personalized and more human messages that have a better chance at reaching your intended audience!

Talk to experts

We know we’ve been thorough, but fully understanding the challenges of sending email at scale isn’t something we can teach you in a few hundred words. TrueAccord has a full team of email deliverability experts on staff that can provide industry specific knowledge and know the ins-and-outs of different ISPs’ requirements. 

They also regularly audit our deliverability rates so that we can iterate on our processes and improve and help segment our domains and IP addresses as we grow.

Segment domains and IP addresses

Thankfully, our email experts can help explain what that last bit means. Segmenting your domains simply means building different domains that you can email consumers from. For example, some of your emails may come from emails@companyA.com and others may come from emails@help.companyA.com. The same goes for segmenting IP addresses; you may send some of your emails from your main office and others from your satellite office.

This process can help to limit the risk to your brand’s reputation with ISPs as you are less likely to take a big hit if only one of your many email addresses makes a mistake (e.g. bouncing frequently, receiving a lot of spam complaints, having many of its emails remain unopened). 

This process is intricate and methodical. Creating ten new domains can’t solve deliverability problems because brand new domains also lack authority. If an ISP’s filters see that a brand new email address is sending out 100,000 emails, it’s likely that it’ll be swept to the side. Which brings us to our next point!

Take it slow

Scaling your program too quickly is heavily penalized even among senders with high engagements. Many well-established companies that want to build a large scale email strategy with their existing customer base make this mistake, and sometimes there isn’t a way to fix it. Placing strict limits on email volume growth can help ensure that ISPs don’t flag your domain.

Track your data

Set your benchmarks, track your performance, and make changes as you go. Data is the life blood of a scalable email program. As you’ve seen, there’s a lot to keep track of, and if any segments of your strategy spring a leak, the ship might sink. 

By frequently and carefully monitoring performance—from open and click rates to inboxing rates to bounce rates—you can maintain a full view of your email strategy and make improvements as you build. 

No one has the power to flip a switch and send millions of emails per month without risk, but if you build slowly, you can lay the foundation for a successful email strategy. If you have any questions, let us know in the comments below! 

TrueAccord sends 40x more emails and has up to 70% higher inboxing rates than other collection agencies. Chat with our team today to learn more about what that means for you!

Tracking Performance Data With Digital Debt Collection

By on October 21st, 2019 in Product and Technology

Call centers are notorious for reaching hundreds, if not thousands, of consumers several times per week (and even several times per day!). The debt collection industry is plagued by the perception that collectors are relentless and uncaring, which makes resolving debts even more challenging. Digital debt collection strategies aim to alleviate the stress of incessant calling for consumers, and also provide unique, powerful solutions for creditors.

Collection metrics

Digital-first debt collection strategies provide creditors the ability to track and aggregate more objective performance metrics that help strengthen their collections strategy. Qualitative metrics from traditional call centers are still subject to the endlessly variable human element of a phone call. 

When outreach is entirely automated, it becomes easy to A/B test simple changes (new subject lines, different greetings, etc.) and determine which are the most effective. But how do we define effectiveness? At the end of the process, an effective collections strategy is one that leads customers to make a payment. 

There are a few key metrics that call centers use to drive customers to this end goal that can be easily supplemented or overtaken by digital collection strategies.

Calls per account and calls per agent

Traditional collection agencies, like any other sales call center system, track the total amount of calls made to each customer and by each agent on the team. When individual agents are responsible for contacting customers, they have to hit an outreach quota. This quota reflects directly back on the calls per account, or how many times an individual customer has been contacted. 

As agents are required to call customers and collect on accounts, the calls per account may increase to a point where customers feel overwhelmed and over-contacted (which can even lead to symptoms of anxiety and depression). At the same time, if countless calls are being made, and an account is not paying, there is a clear gap in effectiveness. 

One of the advantages of a digital debt collection strategy is that agencies can reach customers with relevant messaging at times that work for them. This can include hours in which call centers are no longer legally allowed to reach a customer—before 8am or after 9pm. With these legal limitations in place and the need for agents to meet quotes, traditional collections strategies encourage an artificial inflation of outreach numbers that may not be positive.

Hit rates, percentage of outbound calls resulting in promise to pay (PTP), and call quality 

Call volume is not the end-all-be-all of call center metrics though. Simply tracking output numbers isn’t enough when engagement is the key metric. Hit rate is defined as the total number of calls divided by number of those calls that are answered by customers. While this number can be helpful in narrowing which calls were more successful than others, it cannot reach the same level of detail as a full digital strategy.

In the case of a phone call, there are limited options once the phone has been dialed:

  • The customer does not answer
  • The customer answers but ends the call before promising payment
  • The customer promises to pay

Trying to understand what leads to a successful payment on a call is then dependent on the agent’s perspective. Digital debt collection conducted through machine learning is able to communicate using personalized and consistent content. Hit rate, PTP, and call quality analysis can then be expanded on, and performance can be measured by:

  • Email Deliverability
  • Email open rates
  • Link click rates
  • Website engagement (Including clicking on further links, filling out forms, viewing specific webpages, and more)
  • Online payments

These data points can help pinpoint where in the process a customer was lost, improve the next attempt at outreach with that data in mind, and eventually guide the account to a payment. With more data and longer periods of time, machine learning processes only continue to improve.

Updating your collections strategy 

TrueAccord takes our digital strategy a step further by looking beyond simply using digital channels and focuses on the power of machine learning to continuously improve our collections performance. We’ve come to understand that creating an effective, empathetic collections experience actually comes from creating a more analytical and AI-driven process.

With better visibility into performance, more granular data points, and more accurate reporting available than ever before, digital debt collection strategies strengthen the power of any collections team.

TrueAccord and the Future of Digital Debt Collection

By on August 6th, 2019 in Industry Insights, Product and Technology

In January 2019, AccountsRecovery.net launched a survey of more than 100 companies in the credit and collections industry to “assess the penetration of digital communication tools and how much they are being used in the industry.” “Digital communication” includes channels such as email, text messaging, and web portals that work to reach to consumers. 

However, these channels are secondary to outbound calls and paper mail, practices that have remained unchanged for decades, even though 70% of companies believe that digital communications have had a moderate to significant impact on their collection rates! Updating these channels for the modern age can improve the collections experience for both the customer and collector. Let’s find out how!

Communication Channels

Email

According to the AccountsRecovery survey, more than half of the companies that took part in the survey are using email communication. A majority of respondents also said that they are sending emails to or receiving emails from fewer than 20% of their users. This means that 80% or more of their customers are regularly receiving calls from collectors to discuss resolving their debts rather than receiving digital communications. 

According to TrueAccord’s 2018 consumer survey, the majority of consumers using our site would rather resolve their debts online than through other channels. With such a large number of consumers interested in online engagement, it’s easy to see why we’ve leveraged digital channels to modernize the collections industry.

We use email communication as our primary form of contact at every stage of the customer lifecycle, and each message is customized for the individual.

Mobile and Text Messages

The prevalence of smartphones has made reaching out to users on their mobile devices an effective and essential channel for communicating with customers. Unfortunately, only 21.6% of collection companies are actively using text messaging as part of their outreach strategy! Even some of the largest agencies in the industry are only texting about ¼ of their customers. 

More than 65% of companies in the collections space that are not currently using text messaging as a channel are concerned about two things: a fear of being sued or not fully understanding what is and is not allowed of them. 

TrueAccord has taken steps to directly address these issues by hard-coding compliance parameters directly into our system, so we are able to securely reach our users where they are: on their phones. In fact, more than 85% of our web traffic comes from smartphones and tablets, and we are able to drive traffic to the right pages through push notifications on those devices. These notifications serve the same purpose as text messaging but are uniquely catered to that specific customer’s needs.

Web Portals

Portals and landing pages created for consumers should be exactly that: designed with them in mind. The vast majority of companies in the collections space have portals specifically designed for customers to manage their accounts, but 75% of those companies report remarkably low engagement through those pages. 

Creating an engaging portal means answering the question: “How can we make the experience personalized for the customer?” TrueAccord embraces this in its design methodology; Shannon Brown, TrueAccord’s Product Design Manager, says that “we’re not pushing offers to them, we’re looking for information [about the nature of their debt] to customize for their needs.” Our design embraces our mission of giving consumer’s control of their financial health. 

You can learn all about TrueAccord’s design philosophy by listening to our full interview with Shannon here!

By focusing on developing interconnected, customized content that reaches users through multiple channels, we can reach consumers via email and mobile push notifications with the goal of bringing them back to our website. 

The debt collection industry at large has a long way to go to meet consumer expectations about financial services. Our machine learning algorithm optimizes which message to a customer to send on what channel, addressing those expectations and letting users manage their debt at their own pace. This is also why we work to provide our users with as much visibility into their debt as possible through easily accessible digital channels.

How TrueAccord Creates High Performing Compliant Content

By on July 31st, 2018 in Compliance, Product and Technology, User Experience
TrueAccord Blog

In debt collection, the language one uses in customer communications makes a big difference on liquidation rates. At TrueAccord, compliant content is the lifeline of our system. We continuously create, test and revise our content to engage consumers more personably—which drives better results for our clients.

Our Goal: Create a Better Customer Experience

Communication styles in the debt collection industry are typically stiff and unapproachable. Most of the time, it sounds like “legalese,” which can be off-putting, if not intimidating, to many customers. TrueAccord has a digital-first strategy to debt collection, primarily with emails, supplemented by SMS and phone calls to effectively engage with our customers. We strive to make our content informative, actionable, and compassionate.

Our mission is to transform the debt collection industry by helping people regain their financial health. Thus, our content is written to reflect that. It’s not accusatory or condescending, but respectful and empowering. We focus on finding solutions and helping people by presenting options on how to resolve their debt.

How We Experiment with Content —and Continually Improve It

Our proprietary content management system (CMS) was designed to help us craft and edit content based on massive amounts of dynamic data. We track everything from the customer’s balance, creditor, where they are in the debt lifecycle, if they’re in a payment plan, and how long we’ve been communicating with them to craft customized emails.

We constantly run experiments to generate the right content for each person. We try new subject lines to see if we can get more people to open emails. We write different calls to action on our buttons to see what drives better engagement. We also consider how far a consumer has to scroll down in an email or a landing page to get to the call-to-action button. If something’s not working well, we try something else. And our machine-learning engine—which continuously learns from our experiences—helps us customize specific and customer follow-ups that resonate. All of these small experiments add up to get us very high open and click rates from customers.

How We Keep Content Compliant

The debt collection industry is heavily regulated and is inherently protective of consumers, as it should be. We always look at communications content through a customer-focused and thorough compliance lens.

Our system provides code-driven compliance, appending the appropriate disclosures and text to automatically comply with whatever is necessary for each user, such as debts unreported out of statute or specific state disclosures. Our compliance rules dictate the content parameters for each customer, making it easier for our content writers to focus on writing compelling content. And yet, because there is wide variation in our writing styles, syntax and payment options, our content is still engaging.

Our legal team gives our content a final review, and we get very granular to ensure the message is clear for every type of customer. We look at the actual message, the email layout and design (including button placement) and even the size of the font for our disclosures. We write content that engages customers but also clearly lays out the customer’s rights and responsibilities.

This process is highly collaborative. Our content and legal teams work in concert to continuously adapt new scenarios to see how different options might come across. Our communications library constantly evolves as we keep on improving our customer engagement.

Think About What You Can Say

Most of the industry is focused on what you can’t say, but they’re not thinking about what you can say. That’s why we spend so much time perfecting our content and why we end up with such great response rates and overall results.

Scaling TrueAccord’s Infrastructure

By on April 12th, 2018 in Industry Insights, Machine Learning, Product and Technology
TrueAccord Blog

TrueAccord’s machine learning based system handles millions of consumer interactions a month and is growing fast. In this podcast, hear our Head of Engineering Mike Higuera talk about scaling challenges, prioritizing work on bugs vs. features, and other pressing topics he’s had to deal with while building our system.

Conversion At TrueAccord: Tuning A Machine Learning Engine

By on April 3rd, 2018 in Industry Insights, Machine Learning, Product and Technology
TrueAccord Blog

TrueAccord’s system is machine learning based, but every new product type requires a little bit of tuning to beat the competition. Hear our CSO and VP of Finance in this short podcast about the Conversion Team and what it does to make sure TrueAccord stays ahead of competition.

 

Building An Experimentation Engine

By on March 20th, 2018 in Product and Technology
TrueAccord Blog

TrueAccord beats the competition on many levels, and does that through rigorous testing and improvement. Hear a talk from our CTO Paul Lucas and Director of Product Roger Lai on our approach to experimentation.

 

To download a transcript of this post, click here.

TrueAccord’s 2018 Customer Survey: Net Promoter Score and Digital Trends

By on February 27th, 2018 in Company News, Product and Technology, User Experience
TrueAccord Blog

 

We just posted our 2018 Customer Survey and the results are incredibly interesting.

Consumers in debt are definitely feeling more like TrueAccord customers, giving us a Net Promoter Score of 40, a new record for us and for the industry. We have also uncovered several interesting trends in customer preferences – not new, but definitely eye opening.

Click here to download the infographic summarizing our findings.

Podcast: Creating a Positive Impact in Debt Collection Using Technology and Building Consumer-Centric Experiences

By on January 29th, 2018 in Compliance, Industry Insights, Machine Learning, Product and Technology

Our CEO, Ohad Samet, recorded a podcast with Lend Academy discussing the positive impact technology is creating in the collections space and the need for more innovation. Will discuss TrueAccord’s unique approach to debt collection using data-driven, digital communications to create deeply personalized consumer experiences.

The podcast also covers the current state of the collections industry and where it’s likely headed as regulatory pressure, consumer preferences and compliance requirements converge.  Will cover how TrueAccord is using machine learning to deliver deeply personalized and engaging experiences for consumers while achieving higher recovery rates across various debt types.

Tune in and learn:

  • The state of the debt collection industry today and where it’s headed
  • How the use of machine learning is personalizing the debt collections experience for greater conversions
  • Why code-driven compliance outperforms traditional collections practices by reducing risk to organizations
  • How understanding consumers’ preferences for easy, self-service options with flexibility empowers  more consumers to pay off their debt and get on a path to financial health 

If you’re rather read the transcript, download it here.