Artificial intelligence (AI) developed for customer service is not intended to replace agents, but to complement them. The emotional intelligence of agents combined with the automation provided by AI is the perfect combination to deliver a flawless customer experience (CX).

But its power does not end there. This technology also has a positive impact on the agent experience (AX), freeing them from managing simple enquiries and performing repetitive tasks where they do not add value.

Here are ten use cases that demonstrate the benefits of integrating AI to help agents be faster, more efficient and work on their continuous improvement.

1. INTENTIONAL QUALIFICATION

Bots are ready to ask an open-ended question like “How can I help you?”, interpret the customer’s response in the moment, and decidewhether to offer self-service (if within their capabilities) or transfer the conversation to the most qualified agent to handle the request.

As with IVR menus, you can use them to identify the user’s intention and intent and create transfer rules that take into account agent skills.

Unlike IVR, conversational and/or generative AI bots do not require you to configure the conversation diagram per channel and provide a better customer experience by understanding the customer’s query first.

2. USER IDENTIFICATION

Another task a bot can do before transferring the conversation to an agent is to identify the person it is talking to.

By querying the CRM, ERP or ticketing tool, it can search the database for the ID card, phone number or any other identifying information the user has provided.

This way, when the agent receives the call or chat, they can address the customer by name and quickly check their call/message history and customer record.

3. CONVERSATION SUMMARY WITH AI

When the customer indicates they want to speak to an agent, or the bot has determined it’s best to transfer the conversation to an operator, the agent receives the call or chat along with a summary of the previous conversation.

In addition to a transcript of the conversation and a record of all actions taken by the bot, the summary of the conversation identifies keywords and the customer’s sentiment.

With this information in advance, the agent can resolve the customer’s query more quickly and efficiently, because they know exactly what the customer needs, what the bot wasn’t able to provide, and what tone to use depending on the customer’s emotional state.

For example, if the bot has detected that a customer is angry because they have a problem with their phone line, the agent will know all this information before answering the call and will be much more effective in their verbal communication and problem solving.

4. RESOLVING COMMON QUERIES

According to a recent study we conducted, bots can handle up to 80% of simple queries in a contact centre.

Whether answering FAQs about a product or service, resolving a technical issue found in a user manual, resetting a password or recommending a credit line, bots free agents from these types of calls or messages, giving them more time to handle inquiries that require emotional intelligence or to assist those customers with special needs.

5. SENDING DOCUMENTS

In addition to resolving simple queries, bots can also provide self-service by sending documents to the customer, such as invoices, contracts, forms, etc.

Whether verbal or written, the bot can instantly deliver the document the customer needs through any platform.

For example, if a customer calls the call centre and asks for their latest electricity bill, the bot can send it via the customer’s preferred channel: WhatsApp, SMS or email.

6. TICKET GENERATION

As part of self-service, bots can also generate a ticket detailing the customer’s problem. By gathering all the relevant information from the conversation, they can create a new ticket, fill it in, and send the customer the incident tracking number.

Especially for queries that cannot be resolved in the first instance, bots can handle this part of the process, reducing human error and freeing agents from repetitive tasks.

7. INSTANT TRANSLATION

As well as being omnichannel, Artificial Intelligence is multilingual, meaning it speaks and understands all languages. This feature opens up a world of possibilities for contact centres, not only to better deal with requests that are not in our native language, but also to provide service in any language.

Whether it is a voice or chat conversation, we can ask the bot to transcribe the customer’s answers in the language we want, and also respond to the customer in their own language.

This functionality is ideal for when we have a spike in Spanish calls in the queue, as we can redirect these calls to agents with basic knowledge and have them assisted by the bot to help customers more smoothly and professionally.

8. KNOWLEDGE BASE

In addition to assisting during a call as a translator, the bot can also help agents when they have a question about a product/service or a clause in a contract.

If we share all our documentation with the AI: manuals, legal documents, blog posts, use cases, etc., we will end up with an internal ChatGPT that is knowledgeable about all our processes and products/services.

This will be the great ally of the agents, because if they don’t know how to answer a customer’s question, they can ask the internal AI Agent, and if it’s in the documentation we’ve provided, it can answer it.

This functionality not only avoids prolonging the conversation, but also allows any agent to provide support on a topic they are not an expert in.

9. ACW OPTIMISATION

The After Call Work (ACW), refers to all the tasks that agents must complete after each interaction before moving on to the next customer.

Artificial Intelligence can speed up this process by creating a real-time transcript of the conversation, analysing customer satisfaction by highlighting keywords that serve as typification, and linking this data to the details of the conversation.

The ability to automate AWC time as much as possible is essential to increasing agent efficiency.

10. AGENT EVALUATION

AI also has Speech Analytics capabilities, which means it can evaluate the performance of agents. This data helps contact centre managers to identify each agent’s areas for improvement and work on strengthening them.

We can train AI to generate a report after each interaction that includes: the ID number of the interaction, a title, a summary and, for example, an assessment of the agent’s presentation, friendliness and effectiveness.

This functionality contributes to the continuous improvement of agents and therefore to their speed.

FINAL THOUGHTS

As we have seen from these examples, AI provides automation across the entire customer service cycle: from the tasks that precede the interaction to the assistance it provides during and after the conversation.

Currently, there is no other technology that can optimise so many processes at once, taking into account both the customer and agent experience.

Want to meet the artificial intelligence that will revolutionise your customer service? Call us on +34 900 670 750 or fill in this form.

La entrada 10 examples of AI automation to help agents work faster se publicó primero en Enreach ES.

When artificial intelligence enters the equation – particularly with the use of chatbots for customer service – it is natural for customers to be concerned. In fact, PwC’s latest report shows that 80% of consumers are concerned about the future development of generative AI and its impact on data privacy.

Of the many myths surrounding this technology, we acknowledge that this is perhaps the most worrying of all. However, according to Termly research, 91.1% of businesses are willing to prioritise data privacy in order to increase customer trust and loyalty.

SECURITY PROTOCOLS IN CHAT CAN MATCH THOSE IN CALLS

When an IVR (Interactive Voice Response) system identifies a customer during a phone call, various security protocols are triggered to protect their data. Typically, the system encrypts the data as it is entered or dictated by the user, ensuring that it does not appear in the call recording or become visible to the agent.

Channel switching – where the customer interacts with a chatbot rather than an IVR – does not change the security measures in any way. AI-powered chatbots can apply the same privacy principles, ensuring encryption, masking and anonymisation of sensitive information.

HOW WE PROTECT NUMERIC DATA IN A CHATBOT

When interacting with a chatbot, numeric data remains protected by the same security standards. When a customer enters their insurance policy number, national ID or access code, the AI processes it securely using automatic masking and encryption.

User: I would like to check my insurance policy number 987654321.

Chatbot: Thank you, we have received your request. The policy ending in 4321 is active.

This ensures that data is never stored in clear text and can only be accessed through secure authentication and encryption processes.

HOW WE PROTECT SENSITIVE INFORMATION IN A CHATBOT

When a customer shares highly sensitive information, such as health details or religious beliefs, the AI needs to be trained to apply additional protocols that minimise risk. Chatbots can detect sensitive data and apply anonymisation, tokenisation and automatic deletion to ensure user privacy.

Here are six ways to protect sensitive data when customers interact with a chatbot.

6 METHODS TO PROTECT DATA IN AN AI-POWERED CHAT

1. DATA MINIMISATION PRINCIPLE

Customers should not be asked to provide sensitive information unless it is absolutely necessary.

For example, instead of asking “What is your disability?”, the chatbot could say “If you need specific assistance, please respond with ‘’ or ‘'”, ensuring that the user does not have to share personal information.

2. AUTOMATIC RECOGNITION AND USER WARNING

If a user types: “I have epilepsy, how do I manage my insurance?”, the chatbot can detect this and respond: “For security reasons, we don’t need you to share any medical information in this chat. Would you like to speak to a specialist agent?”

This prevents sensitive information from being stored unnecessarily in the chat.

3. REAL-TIME ANONYMISATION

When a chatbot collects sensitive data, it can be replaced with generic labels before storage.

For example, if a user types: “I’m a diabetic and want to know what coverage I have,” the database stores: “User inquiring about medical coverage”, ensuring that no specific medical details are recorded.

This method transforms the original message so that it contains no personally identifiable information. While the original structure is lost, the general intent of the query remains intact.

4. AUTOMATIC DELETION OF SENSITIVE DATA

A chatbot can be programmed not to store certain words or phrases that have been identified as sensitive data.

For example, if a user asks: “I have a chronic illness, how does this affect my life insurance?”, the chatbot will process the query without recording the specific medical condition and will only store: “User inquiring about life insurance terms”.

Unlike anonymisation, this method preserves the original message structure with minor adjustments. Instead of replacing sensitive data with generic labels, it simply removes them.

5. COMPLIANCE WITH THE GDPR AND THE EUROPEAN AI LAW

If sensitive data is essential (e.g. in health insurance cases), the chatbot must ask for explicit consent before storing or sharing the information.

For example, when a user starts a chat, the chatbot could greet them by saying: “To help you, we may need medical information. Do you agree to share it according to our privacy policy?”

6. ACCESS RESTRICTION AND TOKENISATION

If sensitive data needs to be stored, the best approach is to use a token instead of the actual data.

For example, if a user discloses that they have celiac disease, the database can store “medical condition #874632″, where only an authorised system can decode the actual condition. This prevents employees or unauthorised third parties from directly accessing sensitive information.

This method is particularly useful for organisations with multiple levels of data access, ensuring that no agent can view sensitive data without proper authorisation.

FINAL THOUGHTS

The same security protocols that protect data during a phone call apply when a customer interacts with a chatbot. AI in customer service does not change the rules of the game – it automates and strengthens the protection of personal data.

A well-configured chatbot can prevent users from sharing unnecessary information and ensure that sensitive data is not stored without consent.

DISCOVER THE SECURITY MEASURES WE IMPLEMENT AT ENREACH

La entrada How to protect customer data when using a chatbot se publicó primero en Enreach ES.

In fact, research shows that 35% of businesses fear that AI could make mistakes that negatively impact the customer experience (CX).

Here’s the uncomfortable truth: AI isn’t perfect. But then again, are humans? Just as you trust your best agents, knowing that they might occasionally make a mistake, AI works on the same principle. The mistakes it makes aren’t due to its nature, but rather to unforeseen scenarios or a lack of continuous improvement.

However, AI has one undeniable advantage over humans: it’s far more consistent. It sticks to the guidelines you give it, it doesn’t get tired, it doesn’t go off script, it doesn’t improvise.

So the key question isn’t whether AI can make mistakes, but rather: “What can I do to make sure it doesn’t make mistakes that my best agents would never make?” Read on, because we’ve got the answers you need.

3 KEYS TO GETTING AI RIGHT

1. KEEP YOUR KNOWLEDGE BASE ORGANISED AND UP-TO-DATE

The challenge:

A bot that relies on disorganised or out-of-date information will not only provide incorrect answers, but may even “hallucinate” by making up details that don’t exist.

How to fix it:

• Develop a clear strategy outlining who is responsible for keeping the knowledge base up to date and how often it should be reviewed.
  Organise your information by separating similar processes, such as the steps for subscribing to and unsubscribing from a service, to avoid the bot mixing things up and confusing customers.
  Use an AI solution that flags conversations by satisfaction level, and assign someone from your team to review the bot’s responses in interactions flagged as “unsatisfactory”.

2. TRAIN YOUR BOT TO HANDLE MALICIOUS QUERIES

The challenge:

A bot that isn’t prepared to recognise requests that have nothing to do with your business, or attempts by hackers to exploit it for free services or products, puts your brand at risk.

How to fix it:

• Configure the bot to respond only within its area of knowledge. Irrelevant queries should be redirected to a human agent or met with clear responses such as: “Sorry, I can’t help with that.”
  Protect your brand by limiting the bot’s capabilities: don’t allow it to process payments, close sales or access sensitive data without supervision.
  Implement AI tools that label conversations by query type. Regularly review interactions in the “other” category to monitor how the bot handles unknown topics.

3. USE A BOT THAT UNDERSTANDS EMOTIONS

The challenge:

A bot that can’t detect a customer’s emotional tone could escalate a tense situation or mismanage a complex request, ultimately damaging the CX.

How to fix it:

• Make sure your bot uses natural language processing (NLP) algorithms to detect emotions such as frustration, urgency, or dissatisfaction in customer messages.
  Train your bot to distinguish between simple queries, such as FAQs, and complex issues that require human attention.
  Design a clear process for the bot to automatically escalate emotional or critical queries to a rep, letting the customer know that a specialist will be handling their case.

IS AI READY TO REPRESENT YOUR BRAND?

The answer is yes – but only if you take the time to implement it properly and focus on continuous improvement.

Even generic AI solutions marketed as “easy to implement” require time and effort. It’s important to define the bot’s purpose from the outset, tailor its tone of voice to reflect your brand identity, and test it thoroughly with malicious queries before deploying it on your channels.

This initial phase is essential to correct up to 90% of the bot’s potential errors. Bear in mind, however, that the remaining 10% can’t be identified and resolved until the system is live, as it’s simply impossible to predict every future scenario.

Remember, well-configured AI can reduce errors by 40%, but poorly implemented AI can damage not only your reputation, but also consumer trust in the technology as a whole.

So if you decide to invest in a bot, make sure you do it right. The AI you choose and how you manage it will reflect your commitment to your business, your customers and the future of technology.

READY TO TALK AI?

La entrada 3 tips for implementing AI in customer service without compromising CX se publicó primero en Enreach ES.

The contact centre industry faces a significant challenge: how to train new agents quickly and effectively so that they can deliver quality service from day one. 

Typically, the training period is very short, and agents begin handling customer interactions without fully understanding all the concepts, often relying heavily on the constant supervision of a colleague or supervisor. 

The cost of a key team member taking time out of their day to help a new colleague is quite high. However, Artificial Intelligence for contact centres can take on this role and even tailor the process to each individual. 

Here are three use cases for automating the training and support of new agents: 

1. CREATE A BOT TO TRACK PROGRESS

By breaking down all the documentation that agents need to learn into different sections, we can create a bot that scores them once they’ve completed each module. By shortening the training time, agents are less likely to forget what they’ve learned. 

At this stage, the AI doesn’t just act as an examiner; it can also identify the format in which the agent learns best and personalise the learning process, for example by providing more video rather than text. 

2. CREATE A BOT FOR INTERACTIVE SIMULATIONS

Once new agents have read all the documentation, we can develop a bot with a simulator to allow them to practice real cases in a controlled environment. 

By feeding the bot historical call data from the contact centre, it can mimic a real customer and challenge the agent. 

After each simulation, the AI can also provide real-time feedback and even generate a report for the supervisor, ensuring they are kept informed of the agent’s progress.

3. CREATE A BOT FOR REAL-TIME GUIDANCE

When the agent starts handling real customer interactions, we can implement a bot to monitor their responses during calls. If they make a mistake, the bot can immediately suggest the correct response. 

In this scenario, AI acts as a support tool for all agents, but for newcomers it provides an extra layer of confidence, taking the pressure off them to deliver a great customer experience. 

AI IN THE CONTACT CENTRE IS NOT HERE TO REPLACE AGENTS

The foundation of customer service is people, and that should never be lost. In an ideal world, we would have the resources for the best agents to personally mentor new recruits, but this isn’t always possible. 

At first glance, having a bot accompany new team members may seem impersonal, but it’s actually quite the opposite. By involving agents and supervisors in the creation of the bot, we can give the AI a human touch, using the same tone that a colleague would. 

READY TO TRANSFORM YOUR AGENT TRAINING PROCESS?

La entrada 3 AI automations for agent training se publicó primero en Enreach ES.

Using a product or service less frequently, not purchasing additional services in the future, and speaking negatively about the brand are all consequences of a poor customer service experience.

A recent Gartner report on the behaviour of new generations warns the industry that now, more than ever, it needs to invest in improving the customer experience (CX).

Fortunately, contact centre professionals are taking the demands of new consumers very seriously, and 86% of customer service leaders will prioritise improving the customer experience in 2024.

“86% of customer service leaders will prioritise improving the customer experience in 2024”.

– Gartner

The question many are now asking is: how do we improve the customer experience? In this post, we’ll outline five processes that need immediate improvement, whether it’s configuring features in contact centre software or integrating artificial intelligence tools for customer service.

1. POOR SOUND QUALITY AND DROPPED CALLS

Customer service is the visible face of the brand; even when dealing with a BPO, consumers believe they are speaking directly to the company. We cannot afford poor sound quality.

Hearing background noise and not understanding what the agent is saying can be extremely frustrating, as can dropping the call halfway through due to poor call quality.

Having noise-cancelling microphones so the customer can’t hear what’s happening around the agent is essential, as is having customer service software with the right features.

Providers that develop their own contact centre software (CCaaS) and are also telecom operators offer all-in-one solutions that simplify operations and administrative management.

2. LONG WAITING TIMES

According to Spanish news channel La Sexta, insurance companies take between 6 and 36 minutes to answer a call, while energy companies take an average of 16 minutes.

Waiting in a queue for an eternity is a universal frustration that inevitably makes us think “we are off to a bad start”. Being put on hold for more than ten minutes in an emergency is undoubtedly very damaging to CX.

The forthcoming Spanish customer service law, which will require customers to be dealt with within three minutes, is about to be passed. To avoid fines of up to €100,000 and not ruin our customers’ experience, there are three solutions we can apply:

1. Integrate a bot that allows users to hang up without losing their place in the queue and receive a callback when they reach the top position. This bot is quick to implement and gets to the root of the problem.
2. Automate the handling of frequently asked questions with an AI bot that can qualify the user’s intent, resolve their query, and reserve agents for more complex inquiries.
3. Enable new customer service channels to reduce call volumes. Millennials and Generation Z, the digital savvy, prefer to interact via their preferred platforms, such as web chat or WhatsApp.

3. REPEATING THE REASON FOR THE CALL

Being passed from one agent to another and having to repeat the problem is frustrating. If the customer does not have a ticket number or is not connected to an expert on the first call, they will inevitably have to repeat the reason for their call.

To avoid this, we can:

1. Implement a skills-based routing system that recognises the user’s need and directs them to the right department.
2. Use artificial intelligence to identify the customer, review their call history, and connect them with the agent they spoke to previously.

Integrating customer service software with a customer management programme (CRM or ERP) allows agents to have customer information with every call, enabling a more personalised experience.

4. DROPPED CALLS DUE TO QUEUE OVERLOAD

Another major frustration when contacting customer service is when the call is dropped because the queue has run out of time.

Therefore, in the configuration of the customer service software, we need to change the rule that causes this to happen.

Alternatively, we can place users in another queue when the current one is about to overflow or, as mentioned in the second point, use callback bots and automate responses to common call reasons.

5. AUTOMATED RESPONSES THAT END THE CALL

At the pinnacle of the worst customer experience are automated systems that do not provide personalised attention. We’re talking about IVR systems that ask users to press numbers on a keypad and play only prerecorded messages without transferring the call to an agent.

The aforementioned Spanish customer service law also prohibits the exclusive use of these systems, penalising companies that do not provide accessible and human customer service.

If we receive a high volume of calls and/or have a small team of agents, we can leverage personalised service with artificial intelligence that automates responses to simple queries.

Unlike automated responders, AI bots can provide the same level of service as an agent, but only for queries that do not require emotional intelligence, such as sending an invoice, explaining the return process or resolving an issue outlined in a manual.

DOES YOUR CONTACT CENTRE SOFTWARE ALLOW YOU TO OPTIMISE THESE PROCESSES?

Discover how Omnichannel Contact Center features are designed to improve your customers’ experience, and how it uses artificial intelligence to reduce waiting times and assist your agents.

La entrada Top customer frustrations when calling customer service se publicó primero en Enreach ES.

Speech Analytics, or assessing the customer service we are providing at any given moment, can be automatedthanks to artificial intelligence for the contact centre.

This technology is capable of analysing millions of pieces of data, classifying them and presenting an assessment in a matter of seconds, which means we can evaluate every single conversation that has taken place with just one query.

The four steps we need to follow to get the best results are:

1. DEFINING GOALS AND METRICS

First, we need to decide which elements of the conversation we want the bot to consider; for example, politeness, problem-solving skills, protocol adherence, etc.

Once we know what we want to measure, we need to establish metrics for each aspect, such as using a 0-10 rating or qualitative analysis.

2. ACQUIRING A BOT

To achieve high Speech Analytics, the bot must be able to generate summaries. This means we need generative AI with an NLP (Natural Language Processing) model to process the language and evaluate the content.

Important! If we want to monitor the calls in our contact centre, we also need to ensure that the bot is able to transcribe speech to text, which means it should include an ASR (automatic speech recognition) model.

3. ACQUIRING POWER BI

This tool, developed by Microsoft, allows us to have customised charts with the information that the AI extracts from all conversations, as well as tables that record the analysis and summary of each call.

4. DEVELOPING THE WORKFLOW

Once we’ve got the bot and Power BI in place, we need to create the workflow that the AI will follow to collect, analyse and send the information so that it’s synchronised within Power BI.

For example, if we want the bot to provide an assessment of a call that includes a summary, the agent’s courtesy rating, and the original audio of the call, we need to create the following workflow:

Step 1 – User request

The process begins when we ask: “Which agent and which call?

Step 2 – Call identification

The bot searches the database for the specific call using the information we have provided (agent name, date and time of the call, or any other unique identifier).

Step 3 – Audio transcription

Once it has found the audio file, it will transcribe the file and convert it to text.

Step 4 – Conversation analysis and summary

The bot will analyse the conversation to assess the aspects we defined in the first step.

If one of the aspects is ‘agent courtesy’, the model will look for indicators of courtesy in the conversation, such as the use of polite language, respectful tone, etc.

It will then assign ratings or tags to the specific aspects according to the metrics we have defined (for example, a score from 1 to 10).

To create a summary of the conversation, it will look for key points such as the reason for the call, the solution provided by the agent and any relevant issues.

Step 5 – Generating the full report IN POWER BI

The bot will compile the transcript, summary and scores and send them to Power BI, and the tool will sync the information every hour.

4. CONFIGURING THE APIs

Once the process is defined, there are two things to consider. Firstly, if we are using a third party tool for transcription and analysis, we need to ensure that the API works with the bot.

We also need to configure the API to ensure that the bot and Power BI are integrated and can automate report generation.

FINAL THOUGHTS

After following these steps, our bot will be ready to provide us with scores for every agent and every call in our contact centre.

All we need to do now is test and tune the model, deploy the system and, as always, monitor its performance.

Of course, we’ll also need to ensure internally that staff know how to use the bot, and technically that the AI meets privacy and security regulations to comply with GDPR.

If this process is too complex for your team or you don’t have someone with the skills to set up a bot, don’t hesitate to contact our technical team to help.

WANT TO OPTIMISE YOUR CONTACT CENTRE MONITORING?

La entrada How to use AI to monitor conversations in the contact centre se publicó primero en Enreach ES.

AI in customer service has many facets, and one of them is using AI Agents to help team supervisors or managers train and support agents on a daily basis, whether they are new to the team or have been with it for years.

Keeping the department well trained will always help to improve their performance, as it is impossible to provide the best customer experience without the relevant knowledge, which can result in the user being put on hold to seek information.

Perhaps the most important aspect for the industry is improving agent engagement. It’s no secret that the contact centre faces a significant talent retention challenge, as the constant interaction with customers takes a lot of energy.

Therefore, providing agents with a tool that can instantly resolve all their queries is essential to avoid the stress and tension of not being able to address a customer’s concern, thereby contributing to their well-being.

4 USE CASES OF AI FOR AGENT TRAINING

1. AI AGENT AS AN INTERNAL KNOWLEDGE BASE

Imagine a ChatGPT for internal use, integrated into the call centre programme and ready to handle queries.

Both new agents and those handling queries outside their area of expertise can quickly access procedures,policies, product information or solutions to recurring problems during calls or messages.

This helps to broaden perspectives and share knowledge across the team, eliminating information silos.

2. AI AGENT AS A CUSTOMER INTERACTION SIMULATOR

In a completely private environment, a simulator can be programmed with a AI Agent to emulate customers, allowing agents to practice their responses.

The AI Agent can also evaluate agents, highlighting areas for improvement and acknowledging successes, thereby supporting their training process.

3. PROACTIVE INTERVENTION AI AGENT

Similar to copilots in search engines or CRMs, contact centre AI vendors are developing proactive AI Agents that suggest responses or provide tips during conversations.

These AI Agents can suggest next steps or offer solutions to problems.

By constantly analysing interactions and evaluating them against pre-defined parameters, they proactively offer suggestions that agents can choose to apply.

4. SPEECH ANALYTICS AS AI AGENTS

Supervisors or managers can now evaluate all agent conversations.

The bot can be programmed to send reports on conversations categorised as negative experiences, summarise interactions, rate aspects such as empathy and problem-solving skills, and provide full transcripts if required.

Reviews can generate objective ratings from the AI, helping to provide evidence and quickly identify training needs for each agent.

INTERESTED IN INTEGRATING A AI AGENT INTO YOUR CUSTOMER SERVICE TEAM?

La entrada How to use an AI Agent to train call centre agents se publicó primero en Enreach ES.

We had the pleasure of interviewing Jorge García, our Product Director, for several articles on artificial intelligence. During one of those informal meetings where we were discussing potential use cases in the contact centre sector, he told us:

“The philosophy behind self-service should be nothing more than automating customer service without losing the quality of experience you would get when talking to an agent.”

It’s true. The word “self-service” can have a certain negative connotation because it evokes a scenario where the customer is alone in the face of danger.

For example, we call the checkout machines in supermarkets self-service, but the machine does not help the consumer to scan the products, the consumer scans them alone.

So when we talk about an AI-powered chatbot that serves users who come to the website and provides answersto their questions or redirects them to the best available agent, we are talking about a system that automates responses.

We have been stressing this a lot lately, but we think it is very important to remind those companies that are thinking of investing in this technology that if it is not applied with certain criteria, the bot can end up truncating the customer experience.

WHAT ARE THE KEYS TO MAINTAINING QUALITY WHEN AUTOMATING RESPONSES?

1. PROCESS SIMPLE AND REPETITIVE ENQUIRIES

The first thing we can start automating are FAQs. Frequently Asked Questions are those that come up all the time, either before buying a product/service or after.

One of the questions a telecommunications company could automate would be “Why has my phone bill gone up this month?”.

If we connect the AI to our customer service agent, the bot can access the customer’s bill, analyse the document and inform the customer of any additional charges.

2. TAKING THE CUSTOMER’S EMOTIONAL STATE INTO ACCOUNT

The AI that powers the chatbot not only understands the user’s request, but also has the ability to extract their emotions. As a result, it knows from the outset the severity of the incident and the emotional state of the customer.

The bot is not a human, so no matter how much we train it to deal with serious problems, it will always lack emotional intelligence.

So if the bot is talking to a customer who has a power cut at home, for example, we need to train the AI to transfer the conversation to an agent.

3. TRANSFER THE CONVERSATION TO THE AGENT WHEN THE USER REQUESTS IT

There is nothing more frustrating for a user than talking to a bot when they want to talk to an agent. So it’s always a good idea to give the user instructions on how to get in touch with an agent.

In addition, it will soon be illegal to offer customer service based on voicemail or bots that do not allow you to speak to an operator. When the new Spanish customer service law comes into force, companies that do not comply will be fined up to €100,000.

4. AVOID BOT HALLUCINATIONS AT ALL COSTS

We don’t want our bot to give a customer the wrong answer. So we need to learn how to maintain the knowledge bases properly.

When we create a knowledge article, we must avoid explaining multiple processes in the same document.

WANT TO ADD A CHATBOT TO YOUR BUSINESS?

La entrada How not to lose quality when using a chatbot to automate responses se publicó primero en Enreach ES.

Yes, artificial intelligence enhances our customers’ experience by providing immediate responses. In fact, 68% of consumers cite this as their favourite feature, according to Userlike.

“68% of consumers cite this as their favourite feature”.

Userlike

However, a good experience can turn into a nightmare if the chatbot isn’t programmed correctly.

Here are four mistakes we see on some websites and how to fix them.

1. POORLY STRUCTURED FAQS

With a conversational AI chatbot, we’re obliged to create a menu of options (similar to a written IVR menu) for users to select the most relevant question.

But what if there isn’t a suitable option, or worse still, there’s no place to ask an open-ended question? This hampers the customer experience.

The philosophy behind self-service is to automate customer service without sacrificing the quality of the agent-assisted experience.

If we don’t embrace this premise, we’ll create a bot that won’t benefit the user and will result in a flood of phone complaints.

SOLUTION 1: OFFER A VARIETY OF FAQS

Especially at the beginning, when you are unsure of which FAQs to offer, we recommend offering a range of options, from pricing and product information to job applications or demos. Once the bot is up and running, it is important to continually improve these FAQs.

Using the bot’s configuration panel, you can track how many queries each topic receives and replace those that have no queries within a month.

SOLUTION 2: IMPLEMENT A HYBRID CHATBOT

Another approach is to enable a hybrid chatbot that combines a conversational AI tree menu with open-ended, generative AI queries. This reduces user frustration when they can’t ask for what they want.

While generative AI can be creative and susceptible to trolling, it is important to limit it to specific knowledge bases and test it before activating it on our website.

This option is particularly attractive for contact centres, which can use open-ended questions to incorporate the most common ones into FAQs.

2. INABILITY TO SPEAK TO AN AGENT

Customers who contact a chatbot are generally looking for a quick, out-of-hours response. However, some may have more complex queries and prefer not to call.

“The customer is always right” is a mantra when dealing with the public. Even if phone channels are reserved for incidents, if a user contacts the chatbot with a serious problem, it is necessary to allow the chatbot to transfer the request to an agent.

TRAINING THE BOT PROPERLY

We train our generative AI to automatically transfer the chat to an agent if it doesn’t understand the nature of the query, without the customer explicitly requesting it.

Firstly, because with the new Spanish Customer Service Act, we must always offer human support as an alternative to chatbots, or face fines of up to £100,000.

And secondly, it significantly improves consumer readiness: according to Simplr, 80% of consumers are more likely to engage with a chatbot if they know the conversation will be transferred to an agent.

“80% of consumers are more likely to engage with a chatbot if they know the conversation will be transferred to an agent”.

Simplr

3. NOT ASKING FOR FEEDBACK

If we’ve recently activated the chatbot on our website, we’re missing a significant opportunity to measure customer satisfaction by not requesting feedback after each interaction.

The AI can analyse and identify user sentiment during conversations, and send us a rating for each one. However, it’s also beneficial to contrast the bot’s automated assessment with direct user feedback.

Consumers like to be heard, so while we can measure their sentiments independently, it is always valuable to involve them in the bot’s improvement process.

4. 4. ARTIFICIAL INTELLIGENCE HALLUCINATIONS

AI hallucinations occur when the bot confidently gives a wrong answer because it thinks it is the right answer.The purpose of AI in customer service is to free agents from routine queries, not to give us headaches.

To avoid this, we recommend that knowledge bases are well structured. This means avoiding mixing concepts in the same article.

For example, if a document explains how to activate a service, don’t include the steps to deactivate it.

READY TO IMPLEMENT A CHATBOT FOR YOUR BUSINESS?

La entrada 4 common chatbot implementation mistakes and how to avoid them se publicó primero en Enreach ES.

After evaluating all the available entries, we decided to submit our applications for “Best Automation and Robotisation Project” with EVA and “Best Cloud Computing Project”, showcasing our Omnichannel Contact Centre.

The good news filled us with joy! So we travelled to Madrid, our second home, to enjoy an evening surrounded by colleagues, friends and outstanding professionals in the industry.

After the presentation by Lourdes Álvarez, CEO of Peldaño, Alfred Nesweda, our CEO, and Carlos Jiménez, our CCO, took to the stage to receive the two awards, which undoubtedly recognise our hard work and dedication to the customer experience.

We recently launched a portfolio that covers all the solutions businesses need to communicate internally and externally. These awards confirm our leading position as a provider of all-in-one solutions that combine contact centre software, collaboration platform, fixed and mobile telephony, business numbering and artificial intelligence.

EVA: THE AI FOR CUSTOMER SERVICE

Our artificial intelligence speaks all languages and handles calls, chats, WhatsApps, SMS, emails and social media.

It excels at increasing service levels, reducing waiting times, improving the customer and agent experience, better allocating resources and increasing efficiency.

To date, EVA has successfully resolved more than 50% of requests for one of Europe’s leading airlines, or 150,000 requests out of a total of 300,000, without human intervention.

In addition to doubling agent productivity, the company has reduced costs by 50% and increased customer satisfaction by 60%.

And this is just one of many examples. Explore other automations designed for all customer service teams.

OMNICHANNEL CONTACT CENTRE

Our advanced customer service software allows agents to manage all communication channels from a single platform: calls, chat, WhatsApp, email, Click&Talk and SMS.

Supervisors, meanwhile, have full control over lines, queues and agents, and can monitor service levels from a wallboard and real-time statistics panel.

Its intuitive interface, integration with CRMs, ERPs and ticketing systems, automated call distribution and artificial intelligence capabilities take customer service to new heights.

In addition, the use of agile methodology allows us to deliver bi-monthly updates based on customer feedback, with our innovation and development team customising the tool to meet customer requirements.

THANKS AND ACKNOWLEDGEMENTS

First and foremost, we would like to congratulate our team of 1,000 people who work tirelessly and passionately, and thank our customers and partners for the trust they have placed in us, some for over twenty-five years.

We would also like to thank the members of the Platinum Customer Experience Awards Jury and the entire Peldaño team, especially María Muñoz, Contact Centre Hub Director, and Mario Martelo, Communications Consultant .

WANT TO BE OUR NEXT SUCCESS STORY?

La entrada We won two awards at the Platinum Customer Experience Awards se publicó primero en Enreach ES.