Tag Archives: AI

what is agentic ai

What is Agentic AI? And what does it mean for customer experience

Introduction

You might just be getting your head around AI and the world of ChatGPT, Android Gemini and Apple Intelligence. However, the game has already moved on from Generative AI to Agentic AI. So what is Agentic AI? And what does it mean for customer experience?

Agentic versus Generative – what’s the difference?

Generative AI uses LLM’s (Large Language Models) to generate a response based on the prompt it is given. The AI is trained on information and builds appropriate responses which are predetermined by the question asked. Think about your typical customer services Chatbot and invariably it will be powered by Generative AI. As a result it is only as good as the information it learns from and the instructions it has been given. Saascoms Omnireach uses Generative AI and it can resolve up to 93% of customer queries.

For instance a customer may ask a Chatbot, ‘I think my account has been hacked or compromised, there are card payments that are not mine.’ The Chatbot will analyse the wording and may respond with, ‘Would you like me to freeze your account and pass you to an agent?’

As the name suggests, it ‘generates’ an answer but waits for a further prompt to take action.

Agentic AI on the other hand still uses LLM’s. However, the main difference is it aims to solve the problem entirely. So what is Agentic AI? In the example above the Agentic AI may respond to the prompt with, ‘Your account has been frozen and the investigations team has been informed.  A new card has been issued which you will receive within 3 days and we will call shortly to discuss the queried payments.’ Agentic AI doesn’t wait for further prompts, it takes multiple steps to solve the task.

This short VIDEO is a good explainer into the new world of Agentic AI.

Use Cases for Agentic AI

So here’s where fact meets fiction, we’ve used AI to generate the list of use cases for Agentic AI. Here’s a selection:

  • Cyber Security – real time threat detection, stress testing (and repair of weaknesses) and active threat hunting. AI can become the security guard for your systems.
  • Finance – from virtual Financial Advisor to Fund Manager, making investment choices through live data analysis and personal recommendations and action.
  • Supply Chain – advanced stock replenishment and supplier performance analysis, workflow efficiencies and distribution management.
  • Customer Services – end to end problem solving (as per our example) and streamlined experiences.
  • Healthcare – active diagnosis based on symptoms, drug development (AI has recently developed new antibiotics) and patient monitoring (actively linked to a smartphone or watch).

Summary

What is Agentic AI? An Artificial Intelligence that is developing rapidly and will impact on nearly all our everyday lives. It is what most people already thought AI was about; decision making, analysis, end-to-end workflows and more. It has the potential to change all our lives, how we work, what jobs we do and how we interact with the world.

global language

Customer Service Language Translation – the Omnireach solution

Introduction

In a modern world, a multi-lingual society can pose either a problem or an opportunity for business. You don’t bring your credit rating with you when you migrate, or your no-claims bonus. (There are specialist financial services for both). However, you do bring your native tongue. And we’re not just talking about migration to Britain, think of your average ex-pat in Spain – language can be a barrier. So customer service language translation needs a solution.

The Language Barrier

As we’ve often discussed, the purchase of a product or service tends to be the easy bit. It’s when it goes wrong, there is a dispute or an issue that challenges occur. Now add a non-native dialect, and customer service language translation becomes an important element of the journey.

Contact Centres cannot be expected to have language experts covering every customer preference. In the UK 10% of the population list English as not their main language, and within that 10% there are at least 15 languages spoken by over 100,000 people.

As a result, a translation solution is an important tool in the pursuit of customer service excellence.

AI translation

The Omnireach Solution

The award-winning digital chat platform utilised by 700 brands globally, Omnireach has the functionality to translate 92 languages across email, chat and SMS. Omnireach is powered by AI which resolves up to 95% of customer queries and handles administrative tasks such as ID&V. Within the platform Omnireach can detect and respond to different languages.

Should the customer be referred to a live agent, Omnireach offers a one click translation service. The agent clicks on the incoming message for instant translation, composes a response and once again a single click translates the message back for the customer. As a result the customer receives a seamless two-way conversation across email, text or webchat.

AI translate

Conclusion

Omnireach removes the need for expensive translators and empowers every agent to communicate across 92 languages, providing a simple and effective solution. For customers the initial digital journey is handled by AI communicating in their preferred language. Customer service language translation is a built in function of Omnireach, a standard feature of Saascoms award winning software.

Contact Saascoms
Man v Robot

Call Centre Agent v AI Chatbot – who wins the customer service battle?

Introduction

While call centre agents have long been the backbone of customer support, AI chatbots are becoming increasingly sophisticated. But which one is better? The truth is both have their strengths. Let’s compare five tasks where each excels and see why with Agent v AI Chatbot the best approach might be a hybrid one.

Call Centre Agents Are Best

1. Handling Complex or Emotional Issues

Customers need empathy and understanding, especially in stressful situations. Human agents can read tone, adjust their responses, and provide comfort in ways AI can’t. Plus sometimes it’s just nice to talk to someone.

2. Negotiating & Problem-Solving

Disputes, refunds, or special cases often require creative solutions. A call centre agent can adapt to unusual requests and negotiate resolutions AI wouldn’t and couldn’t think of.

3. Building Customer Relationships

A great agent doesn’t just solve problems, they connect with customers, strengthening brand loyalty and trust through meaningful interactions, humour and rapport.

4. Handling Multi-Step Queries

When a customer’s request involves multiple departments or processes, human agents can navigate complexities much better than AI.

5. Understanding Accents & Dialects

AI speech recognition has come a long way, but humans still outperform chatbots in understanding regional accents, slang, and nuances. How often do you shout at your smart speaker!

Man v Robot

AI v Human

AI Chatbots Are Best

1. Instant Responses & 24/7 Availability

AI chatbots never sleep! They handle thousands of queries simultaneously without customers waiting on hold.

2. Processing Routine Queries

Need to reset a password? Track an order? Get store hours? ID&V? AI chatbots are perfect for repetitive, straightforward tasks.

3. Speed & Efficiency

AI instantly retrieves data and provides answers faster than any human could. No small talk, just solutions.

4. Data Handling & Integration

Chatbots integrate with CRMs, databases, and customer histories to instantly pull up relevant information.

5. Multilingual Support

AI can communicate in multiple languages instantly, making it a great option for businesses with global customers.

The Best Solution? A Hybrid Approach!

Rather than deciding Agent v AI Chatbot, businesses should combine AI chatbots with human agents to create a seamless customer experience.

  • AI Chatbots handle fast, simple queries and direct customers to the right resources.
  • Call Centre Agents step in when empathy, problem-solving, or human insight is needed.

Using an AI driven omnichannel software platform such as Saascoms’ award winning Omnireach, combined with trained agents, provides the backbone to many global brands customer services.

While call centre agents have long been the backbone of customer support, AI chatbots are becoming increasingly sophisticated. But which one is better? The truth is both have their strengths. Let’s compare five tasks where each excels to see why the best approach might be a hybrid one.

Contact Saascoms
Saascoms AI

Artificial Intelligence in Credit and Collections

Conversational commerce has grown significantly, especially during COVID restrictions. Consumers now expect to engage at their convenience, using multiple technologies.

It is predicted that by 2025, 85% of all customer service interactions will exist without person-to-person involvement.

Saascoms Omnireach is a conversational agent platform which centralises all forms of chat (webchat, SMS, e-mail, Whatsapp and other social media) onto a single engagement platform.
The customer can communicate via their preferred channel and switch between channels during their interaction without needing to re-validate their ID.

At the heart of the solution is a powerful and trainable “frontline” AI chatbot that communicates in 92 different languages and analyses conversations using sentiment, key phrases and natural language processing (NLP). The AI chatbot is contextually aware and, by machine learning, becomes progressively smarter.

This ‘virtual assistant’ has a pre-programmed ‘Arrears and Debt Resolution’ knowledge base taken from tens of thousands of interactions in a DCA environment and is used to train machine-learning algorithms that power our AI.

The AI engine has monitored and analysed over 100million conversations and agent responses within the Collections & Arrears environment over the last four years, ensuring our NLP engine has a 99.7% intent and sentiment match success rate. This makes it possible for the AI to identify up to 75% of the customer intents at the adoption stage.

By providing ‘frontline’ support to the agent function, conversational AI analyses chat quickly and in real-time to automate conversations, drive efficiency, and respond to customer needs. In the first instance, the AI will identify and verify (ID & V) the client via an API or data feed into the client CRM database. Once security has been passed, the AI Chatbot will accurately identify the customer’s intent and respond accordingly.

The Chatbot detects nuances in a customer’s questions and responses and gives relevant answers the way a human would. The conversation continues until a question has been answered, their problem has been solved, or until the request is transferred to a live agent.

Many tasks can be dealt with without the need for agent engagement, as the Chatbot takes the customer through a workflow process The Omnireach AI typically resolved 27% of all interactions with a single line response and 42% without live-agent involvement. The latest statistics from Q1 2022 suggest 72.4% of conversations engaging with the Saascoms Collection & Arrears Chatbot are resolved on the first contact.

Accurate identification of a customer’s requirements is key to providing a high-quality customer journey, raising engagement, providing better CSAT scores (typical 12% uplift) and improving outcomes. By accurately identifying intent, the AI can route more complex enquiries to the best-equipped agent team to deal with them. Conversation CSATS suggest 89.3% of customers were completely resolved with the Saascoms AI Chatbot.

Using AI and routing to a specialist team makes customer conversations more meaningful, and agents have more time for complex interactions. Collections have been proven to increase by over 10%. The AI is trained to quickly identify a ‘vulnerable’ customer and immediately divert to a specifically trained live agent where the query can be handled appropriately.

Agents have the ability to help train the Collections & Arrears AI engine. By copying content from conversations and adding them directly into the AI engine, the Saascoms platform will evolve and ensure AI can identify future conversational meanings.

The agent can then free type a reply or select from a branded template, which then adds personal customer information into the response via a link to the organisation’s CRM. When a response is registered within the Saascoms engine, it is added to the data pool for analysing and adding to the ML algorithm.

Omnireach AI conversational technology is utilised by nine collections and arrears organisations in the UK, who, on average, process 2.8million interactions per month. Sectors include energy, telecoms, vehicle finance, home shopping (catalogues), personal finance/loans, and student loans.

Learn More about Omnireach