The Credit and Collections industry is undergoing a revolution driven by Artificial Intelligence (AI).
At Saascoms, we’re harnessing this powerful technology to streamline our processes and empower our clients. This translates to improved customer engagement and positive outcomes for everyone involved.

We understand that some people may be concerned about the use of AI in credit and collections. However, we want to assure our clients that AI is used to improve efficiency and that human agents are still involved in the collection process.

So let’s peel back the layers of AI in the Credit and Collections sector and understand how everyone is benefiting.

AI: Tell me more!

AI refers to the development of computer systems and software that can perform tasks that would typically require human intelligence.
Tasks can include:

• Problem-solving
• Understanding natural language
• Recognising patterns
• Making Decisions

AI systems aim to simulate or mimic human cognitive functions such as reasoning, problem-solving, perception and language understanding.

Approaches to AI

1: Tag-based processing
Using keywords to be recognised within a conversation, which then assigns a preset workflow. Keywords such as “illness”, “off sick or “death” can be tagged as ‘Vulnerable’, “HMRC”, “Fraud”, and “Court Protection Order” can all be tagged as ‘Crime’.
Tag processing could be described as rudimentary, but when used in the correct context, it can be highly effective within your customer journey experience.
Example:

2: Natural Language Processing (NLP)
This is AI that understands the intent and sentiment within the customer conversation, it then automatically assigns the preset workflow or customer journey. NLP is essential for Chatbots, language translation and sentiment analysis.
Example:

3: Machine Learning (ML)
Machine Learning is a subset of AI that involves training algorithms on large datasets. ML learns patterns and makes decisions based on previous interactions and outcomes.

AI does not benefit from pure design, meaning you do not create it, and it simply works forever. AI requires awareness of improvement and feedback from agent activity.

Where does the data set come from?
We have established that to run “AI” you need data, and the right sort of data is key.

There are two types of data sets, or Large Language Models (LLMs):

Open Source
Open Source comes from a wide range of public and commercial data sets. For example, research institutes, universities and Government agencies. It can also be obtained from crowdsourcing, i.e., from companies and research projects that use general public collected information shared through online platforms. Social media is another source of data, as are data marketplaces—companies & organisations that operate within data markets where they sell datasets.

Controlled/Closed Source
This refers to datasets that are not openly or freely available to the public. Usually, the data is proprietary and has restricted access. Your customer data, used within your systems, would be an example of this. Limited sharing and data security offer important operational benefits.

Did you know Saascoms has its own closed LLM data specifically trained for Credit & Collections, having analysed over 130 million customer interactions?

How can AI help you with Credit & Collections?
By using AU technology correctly across all of your non-voice channels, you can increase productivity by freeing up valuable agent resources, as well as providing an enhanced customer experience.

• Real-world examples of AI being used in Collections are;
• Allow AI to ID&V your customers before assigning them to an agent.
• Allow AI (after ID&V) to assist with simple requests, which include balance checks, request statements, reporting a change of circumstance, and completing an I&E.
• Watchguard for customer vulnerability. Saasbot language is designed to follow TEXAS, IDEAS and CARERS. Send links to advise bureaus when a vulnerability is detected, for example, Step Change Macmillan etc.
• Provides out-of-hours overwatch for FAQ’s and simple queries.

AI Legislation and Compliance
The government is drawing up plans to regulate AI, with concerns based on data security, human rights, and safety.
We collectively need to ensure that AI does not display biases against particular groups if trained on large datasets scraped from the internet, which could include racist, sexist, or other undesirable material.
Open source data “scrapes” the internet for its information. As a business, you will be responsible for the responses it provides to your customers.
Closed source data is a much safer option.

If you use AI, you should be able to demonstrate:

• Safe, Secure and Robust: Prove that risks from AI are managed
• Transparency: Be able to explain when and how systems-making decisions are being used and risks are mitigated.
• Fairness: That you are complying with UK laws regarding quality, DPA, and discrimination.
• Accountability: Compliance oversight, accountability for outcomes.
• Contestability: A process or route to allow customers to dispute the AI outcome or decision.

What would Saascoms recommend?
We think conversational AI is a great service, and when approached with a clear plan, it can become a highly impactful tool within your customer engagement strategy.
Our top tips are:
1: Start with one channel. Whether that is live chat, two-way SMS, or email.
2: Use controlled/closed data providers.
3: Start with “Tag” and “NLP” in the first instance.
4: Work with your tech partner to move towards a semi-supervised environment. Allow the AI engine to suggest a response to your customer, or perhaps the next task, Allow a human to supervise the interaction, i.e., an agent to tag and label each interaction with a sentiment and intent score. This will allow the algorithm to move towards Machine Learning (ML) and then true AI automation.

An example of a recent conversation between a customer and a Collections company—a real-world example. This highlights how AI can help.
(Disclaimer: the personal information has been altered to protect data.)

In conclusion, AI is revolutionising the Credit and Collections industry, and Saascoms is at the forefront of this change. By harnessing the power of AI, we can streamline processes, improve customer engagement, and achieve positive outcomes for everyone involved.

Contact us to learn how our AI-powered solutions can help you improve efficiency, improve engagement, and provide a better customer experience. The benefits of implementing AI to assist your business are numerous, and the results now speak for themselves.