Speech Analytics is a process of analysing the voice of the customer, consisting of the collection of numerous retrievable, identifiable and re-usable information on the interactions between this customer and a company.
Speech Analytics is traditionally used in customer relations by companies to increase their knowledge of the Voice of the Customer and is now a key lever in customer relationship management, particularly in call centres.
Speech Analytics can go further by being combined with Quality Management or Quality Monitoring, i.e. the analysis of the advisor's voice in addition to the customer's, thus validating the quality control and conformity control of the interaction between customer and advisor.
EN
BATVOICE AI: EMOTIONS, SPEECH ANALYTICS and quality monitoring
What is Speech Analytics?
The basics of Speech Analytics: definitions, KPIs, use cases for customer relations, future prospects...
Definition of Speech Analytics
The future of Speech Analytics
The field ofEmotion & Speech Analytics" now makes it possible to integrate an additional dimension into Speech Analytics: humanemotion. Compared to a traditional written transcription system combined with a keyword search, the "emotional" factor provides customer relations managers with a finer understanding of the reasons for calls or customer irritants.
While Speech Analytics analyses the words you speak, Emotion & Speech Analytics technologies analyse the non-verbal content beyond speech (or prosody): how customers and advisors interact, including physical symptoms that indicate emotions such as strong anger or delight, in order to link them to call topics and patterns so as to prioritise actions for improving business processes, products and services, or advisor arguments and speeches.
Need advice on how to implement Voice of the Customer analysis & Quality Monitoring in your business?
Or do you simply want to see what Speech Analytics
can do for your customer relationship management?
DEMONSTRATIONcan do for your customer relationship management?
THE BEST CUSTOMER RELATIONS CENTERS TRUST US
Issues addressed by Speech Analytics
Reducing costs
-300 K€
Time spent by advisors and supervisors on low value-added calls digitised as a priority.
Improving productivity
-30%
Less time spent on your Quality processes: calibration, picking, evaluation and debriefing.
Save time
2X less
Time to carry out each Quality Monitoring assessment and time to analyse the results and assist the advisors in their training.
The Speech Analytics Glossary for Customer Relations Centres
CRC: Customer Relations Centre
CRCs, or Customer Relationship Centres, are a service dedicated to handling incoming requests from customers and/or prospects, made through one or more communication channels (telephone, email, live messaging, chatbot, etc.). The term "customer relations centre" refers to a set of technical and human resources whose missions are focused on customer information, after-sales service and even sales.
In France, the quality of customer relations centres is validated by obtaining NF Service Relation Client (NF 345) certification, which is based on the international standards ISO 18295-1 & ISO 18295-2. This certification was created in 2004 by customer relations professionals and AFNOR to provide consumers with a benchmark for the quality of service provided.
FCR: FIrst Contact Resolution
FCR(First Contact Resolution ) is a rate measured by dividing the number of calls resolved on first contact by the total number of calls made in the relevant contact centre (or by the number of first calls if you wish to exclude repeat calls from your calculation).
The FCR is seen by managers as one of the key success indicators in customer relationship centres as it reflects not only the efficiency of the teams' handling of inbound enquiries but also a limitation in the number of low value calls (repeat calls), leading to increased customer satisfaction and lower service costs.
Churn rate or attrition rate
The churn rate churn (or attrition rate) is an indicator of retention that can be used to assess the extent of customer attrition, although this indicator can also be applied to employees (equivalent to employee turnover). The churn rate is mainly expressed as a percentage and is calculated as follows: by dividing the number of customers lost over a given period by the total number of customers over the same period (NB: if you want to calculate turnover, replace customers by employees).
CSAT: Customer Satisfaction
CSAT(Customer Satisfaction ) is a customer relationship centre indicator measured in percentages most of the time and obtained through dedicated questionnaires. CSAT is measured as follows: number of satisfied (score of 4/5) and very satisfied (score of 5/5) customers divided by the total number of customers interviewed in the CSAT survey.
Although CSAT is used in many situations, it is important to study it alongside other performance indicators to get a representative picture of the customer experience: the response to a CSAT questionnaire, administered by email or telephone, may be conditioned by the quality of the previous interaction between customer and advisor. For example, a dissatisfied customer may not respond to the questionnaire at all, whereas a satisfied customer may take the time to respond, thus skewing the sample.
For more representative results, it is recommended to combine CSAT with NPS or to use Voice of the Customer analysis.
ESAT : Employee satisfaction
ESAT(Employee Satisfaction ) is an indicator used in particular by customer relations centres to measure the satisfaction of their employees, expressed as a percentage and obtained through dedicated questionnaires. ESAT is particularly useful in sectors with high employee turnover/churn.
ESAT is measured as follows: number of satisfied (score of 4/5) and very satisfied (score of 5/5) employees divided by the total number of employees interviewed in the ESAT survey.
NPS: Net Promoter Score
The NPS or Net Promoter Score is a very popular indicator among customer relations professionals that measures the ability or not of a customer to recommend a solution to others. This indicator, a true barometer of the customer relationship and experience, is appreciated by professionals because it is obtained by means of a single question, the answers to which provide information on both customer satisfaction and loyalty.
The NPS is measured using a dedicated questionnaire that asks the respondent to rate on a scale of 0 to 10 their likelihood of recommending the product/service concerned to others. Respondents giving a score between 0 and 6 are considered detractors, those giving 7 or 8 are considered neutral/passive, while those giving 9 and 10 are considered promoters. The NPS is calculated by subtracting the percentage of detractors from the percentage of promoters in the sample surveyed.
CES: Customer Effort Score
Similar to the NPS, the CES is characterised by the formulation of a single question to obtain information on the state of the customer relationship, asking the customer to rate themselves and the effort required to get their case resolved. Typically, the CES will be measured by a choice of 5 to 7 degrees of agreement (from "strongly agree" to "strongly disagree") with a statement such as: "The resolution of your dispute was easy to manage."
The CES metric is used in customer relationship management, but also in improving the customer experience (CX). The CES can thus help in the prevention of customer churn.
The CES is then calculated by subtracting the percentage of people with different degrees of disagreement from the percentage of people with different degrees of agreement with the statement.
QOS: Quality Of Service
QOS(Quality Of Service ) measures the level of service quality that is provided by the customer care centre to its callers. Often, QOS is measured by a combination of several individual metrics such as the FCR rate, the volume of calls handled daily or the average call handling speed (ACL).
Thus, the QOS makes it possible to determine the level of service to be provided by an outsourcer by means of objective measurements, and represents a fundamental subject when defining the SLA (Service Level Agreement). During the service provided by the outsourcer, the implementation of quality monitoring processes (a process that can now be automated thanks to artificial intelligence) makes it possible to validate the application of the QOS conditions provided for in the contract.
SLA: Service Level Agreement
The SLA (or Service Level Agreement) is the contract that describes the level of service provided to a customer by the supplier. The purpose of this type of contract is to identify and quantify the characteristics and various metrics of the service to be provided to ensure the quality of the service to the customer. For example, for customer relations centres, a particularly common SLA clause consists of a commitment to answer 100% of incoming calls within X seconds (to be agreed between the two parties).
The SLA is particularly important for companies that outsource their customer relationship management, especially in the case of call centres. Among other things, it makes it possible to guarantee homogeneous levels of responsiveness, customer relations and customer satisfaction throughout a company's customer relations, regardless of the channel used to handle complaints.
ASA: Average Speed of Answer
Average Speed of Answer ( ASA) is a critical metric for the operational efficiency of customer contact centres and their customer experience. This indicator measures the time it takes for an advisor to answer an incoming call, from the moment the customer is placed on hold to the moment an advisor picks up the phone to handle the request.
If it is too high, the ASA may reflect a lack of efficiency on the part of the adviser teams. New methodologies (digital tools, management, internal organisation, etc.) can then be put in place to reverse the trend.
ACL : Average Call Length
ACL(Average Call Length ) is another crucial metric for customer relationship centres and their operational efficiency. This indicator measures the time it takes for an advisor to handle an incoming call, from the moment the customer is connected to the advisor until the customer hangs up after processing the request. If the LCA is too high, it may reflect a lack of efficiency in the adviser teams that is detrimental to the customer experience. It is useful to set an LFA target that allows teams to assess and manage the workload on the platform.
To calculate this, divide the sum of the durations in seconds of each incoming call by the number of incoming calls and you will obtain the LCA in seconds of your customer relations centre.
CPC: Cost Per Call
CPC(Cost per Call ) is a key indicator of the operational efficiency of your customer contact centre. It is calculated by dividing the total cost of all calls handled by the total number of calls made, and its average value varies widely across industries.
Many customer relationship centres set a certain amount as a CPC target and direct their customer relationship improvement action plans towards lowering this CPC (this is even more true for CRCs with sales targets). However, be careful not to use the CPC as the only indicator, since a very low CPC may reflect a low investment in customer relationship management and a low quality of service. It is therefore important to use this indicator in combination with other indicators, whether they relate to cost efficiency or advisor effectiveness.
Call Abandonment Rate
The Call Abandonment Rate is, similarly to the CES or FCR, one of the quality of customer experience indicators that allow us to assess the difficulty of resolution, by measuring the number of customers who have been put on hold and then hung up before being taken care of. This indicator is critical to avoid maximum frustration for the end customer, although it should be noted that this indicator varies according to usage: for example, a customer calling for technical support will on average show a higher resistance to waiting than a customer contacted for a commercial offer. Generally, it is acceptable for a customer relations centre to have a call abandonment rate of around 5%.
The Call Abandonment Rate is usually expressed as a percentage, and is calculated as follows: subtract the total number of calls answered from the total number of incoming calls, and then divide this number by the total number of incoming calls.
Channel Mix or Channel Mix
Channel Mix is an indicator that measures the contact channel used by the customer to reach an advisor. Modern customer relationship centres do much more than just handle inbound calls, they also have to deal with enquiries made via SMS, website chats and social networks. Channel Mix allows you to assess the source of inbound requests and then adapt your contact centre to the main types of inbound traffic.
The study of the Channel Mix over the last few years has established the general trend that voice channels such as call centres are declining in importance (still heavily used by large groups but much less present in small companies) while self-service support channels such as chatbots are expanding rapidly. But the decline in voice channels does not necessarily translate into a decline in the use of Voice Of Customer(VoC), whose semantic analysis principles can also be applied to self-service channels.
Existing use cases for Speech Analytics
Ensuring quality of speech in contact centres
Speech Analytics enables mass validation of the conformity of interactions with the quality grid previously established in the contact centre. It also enables the identification of "best practices", contributing to the training of advisors.
Reduce customer attrition and team turnover
On the client side, Speech Analytics prevents client departure by identifying risky behaviour. On the counsellor side, it facilitates the development of skills and guarantees an impartial assessment.
Sales conversion and upselling
Speech analytics can be used to identify the keys to success and the obstacles to sales in order to subsequently identify opportunities for cross-selling or adapting pre-existing offers.
Improving the customer experience
Customer irritants are more easily identified and escalated to dedicated teams, making user feedback an ideal part of the customer relationship improvement process.
Need advice on how to implement Voice of the Customer analysis & Quality Monitoring in your business?
Or do you simply want to see what Speech Analytics
can do for your customer relationship management?
DEMONSTRATIONcan do for your customer relationship management?
Our customers say it best
"1 point of conversion, when you estimate that we're at 30% conversion on average, that we do 200 million [euros of telephone sales per year], if you add 1 point, you do the math, we're talking about many millions of euros and it's very concrete."
How to test Batvoice AI?
The team responds to your account creation requests live from the chat, or by your preferred means of contact.
What technical efforts are required?
Batvoice AI is compatible with all telephony solutions used in call centres. We will need access credentials to your recording platform. The solution is compatible with all protocols (API, SFTP, etc.).
What results can we expect?
The thousands of hours of feedback enabled by Speech Analytics, multiplied by each customer, opens up a wide range of high value benefits to our customers: increased sales, conversion rates,upsell and cross-sell, prediction of customer ch urn and reduction of churn, improved customer experience and customer satisfaction, quality assurance and call compliance Depending on your role and objectives, you can help your supervisors and evaluators improve your quality processes with automated Quality Monitoring or improve your customer knowledge with Voice of the Customer.
What efforts by business teams are required?
Batvoice first gets in touch with your account administrator and helps you set up your users' accounts, then offers you a roadmap with two workshops, before giving you access to your new knowledge base of best practices, methods and tools to succeed in your Speech Analytics project. Your teams master the tools and are operational within a few weeks thanks to the dedicated training content of our Batvoice Academy.
Auto QM © Quality Monitoring users notice the time and efficiency gains from the first minutes of use. Batvoice adapts to your quality processes and allows you to easily migrate your internal grids, evaluations and user accounts.
Voice of the Customer users have key information on customer expectations to launch actions and monitor their impact. A strong involvement of the teams around the continuous improvement of the brand and image, the organisation and the processes is necessary to get the full benefits.
Auto QM © Quality Monitoring users notice the time and efficiency gains from the first minutes of use. Batvoice adapts to your quality processes and allows you to easily migrate your internal grids, evaluations and user accounts.
Voice of the Customer users have key information on customer expectations to launch actions and monitor their impact. A strong involvement of the teams around the continuous improvement of the brand and image, the organisation and the processes is necessary to get the full benefits.
What is this technology?
Batvoice AI is based on technologies developed 100% in-house by our R&D department: speech-to-text recognition, natural language processing, affective & social signal processing, predictive analysis. This in-house development is in line with our vision of responsible and ethical AI and the growing importance of data sovereignty, values shared with the VoiceLab, of which Batvoice is a co-founding member.
Is it GDPR compliant?
Yes:
- Your information message at the beginning of the call to your customers remains unchanged: "this call may be recorded for quality and training purposes."
- Batvoice automatically anonymises the personal data contained in the audio of the calls and in their transcription.
- The solution is available either in the "Cloud" or "On premise".
- Cloud data is processed in the European Union in accordance with GDPR policy.
- Your information message at the beginning of the call to your customers remains unchanged: "this call may be recorded for quality and training purposes."
- Batvoice automatically anonymises the personal data contained in the audio of the calls and in their transcription.
- The solution is available either in the "Cloud" or "On premise".
- Cloud data is processed in the European Union in accordance with GDPR policy.
ACCESS THE DEMO
Ready to try Batvoice AI?
Write us a few lines! We'll get back to you within 24 hours to answer your questions and offer you a demonstration.
Thank you for your demo request! We'll get back to you as soon as possible.
