Phone: The Best Sales Channel for Club Med

ClubMed Slide

On the last day of VoiceTech Paris, we attended another great presentation by Amina Chaabane, Head of Digital Sales B2B and Caroline Launois-Beaurain, Head of Global Omnichannel Sales Tools at Club Med, the luxury, all-inclusive holidays firm. 

During the presentation, Amina and Caroline focused on 3 key points:

  • Their omnichannel strategy, which is driven by voice.
  • The transformation of their customer experience.
  • How media attribution is contributing to revenue growth.

Omnichannel Strategy

For many years, the luxury travel industry has operated ‘offline’. Holiday packages and services have been promoted through booklets, specialised magazines, non-digital advertising, and by the agencies themselves.

All the heavy-lifting was done by agents, who also recommended or influenced options, and completed the final booking. 

With the digital world quickly taking over, customers are also expecting a full service to be available ‘online’. Companies are being forced to integrate digital marketing and online processes with their traditional ‘offline’ customer-facing processes to secure growth and provide the high service levels customers expect.

In the breakout session, Amina Chaabane presented Club Med’s digital transformation process, which was launched in 2016. This initiative stems from Club Med’s international strategy to upscale their business.  

Analysing the Customer Journey – Insights into Customer Behaviour

After extensive analysis that combined data from customers’ online and offline journey with data from Club Med’s CRM system, results showed that:

  • Customer journeys were complex and involved multiple touch-points. 
  • Most customer journeys included at least one phone call.

As a result, Club Med launched their omnichannel strategy. The primary focus is voice, as customers chooses phone calls as the main communications channel for sales.

The Power of Contextual Information

The unique opportunity that voice creates, to engage with customers and influence purchase decisions, has inspired the creative ‘juices’ of many voice technology providers.

During VoiceTech we heard many technology providers and customer firms who have developed sophisticated solutions to dissect conversations and gain insights. They have been using voice technology to understand intent, instructions, sentiment, and outcomes to improve understanding after phone calls. 

Club Med collect all the data from customers online and offline journey to predict intent, which has enabled them to influence the customer experience at every stage of the buying process.

Remove the IVR to Increase Sales

Visibility over every customer journey uncovered the first area for Club Med to address — their interactive voice response system (IVR).

Their IVR system, used to route callers to the right agent, was losing many sales opportunities. The unpopularity of IVR was causing many callers to hang up, before talking to an agent.

After running an A/B test, Club Med decided to remove the IVR and use predictive routing. The caller’s online behaviour is used to predict the caller’s intent and automatically route the call to the right team.

Read “Predicting Caller Intent” for reasons why more companies are looking to understand caller intent before the call.

Results shared by Amina Chaabane showed that by delivering a better customer experience and connecting callers with agents faster, they reduced the number of missed calls significantly and increased sales conversion rates.

Seamless Omnichannel Bookings

Once the call is received by the agent, Club Med is able to display the caller’s contextual information to the agent in real time.

Club Med uses computer telephony integration (CTI), so when a customer in the middle of an online booking calls, the agent can take over that booking in one click. Behind the scenes, Freespee passes the booking id from the web session to the PBX.

This new way of engaging with customers is transforming the customer experience. And Agents are enjoying providing a better service that is more efficient for them and more successful for completing sales. 

Marketing Attribution 

Last but not least, Club Med explained that by gaining access to this 360° view of the customer journey, they are able to accurately attribute revenue back to leads and leads back to campaigns.

All data collected by Freespee is sent to Google Analytics in real time, where Club Med can see:

  • A granular view of the keywords and campaigns that are driving the most leads.
  • Which campaigns are converting at the highest rates and the revenue generated by each campaign.

This analysis is helping Club Med optimise their marketing efforts and invest in the campaigns that reap the highest return on investment.

You can read through the presentation Club Med gave here, in English.

Total Direct Energie Benefits from Voice Data

Total Direct Energie Workshop

At VoiceTech Paris yesterday, Total Direct Energie gave a fascinating talk explaining how they have been using voice data over the last four years.

François Rommel, Head of Analytics at Total Direct Energie explained how combining voice data and website data has given them a greater understanding of customer behaviour.

This follows on from the great presentation Amina Chaabane, Head of Digital Sales B2B and Caroline Launois-Beaurain, Head of Global Omnichannel Sales Tools at Club Med, gave “Phone: The Best Sales Channel for Club Med.”

Visibility of Customer Journeys

Total Direct Energie could use traditional web analytics tools to see how customers came to their website, the pages they visited, and online sales conversions.

However, over 50% of sales were completed during phone calls, so Total Direct Energie were only able to see part of the picture. 

By tying every sales call into a web visitor session, they were able to understand:

  • Which digital advertising campaigns were driving sales calls.
  • The overall customer experience for online and offline.
  • Deeper insight into customer behaviour.

Between 2015 and 2019, more and more people became aware they could easily switch energy providers and the market became much more competitive. During this time the number of Total Direct Energie customers rose from under 1 million to 4 million.

Transformation Journey

François Rommel led the process through a number of stages, each time using data to guide the way forward.

To get full data visibility of online and offline behaviour, he implemented Freespee with Google Analytics integration. Since then, they have progressively improved visibility into customer journeys and enriched their audience targeting.

A/B Testing Increases Calls 10%

An early experiment challenged the effectiveness of the special 4-digit phone numbers that Total Direct Energie typically used in marketing campaigns.

Common sense would suggest that dialling 4-digits is going to be easier than dialling a regular 10-digit phone number. But after some careful A/B testing, François Rommel proved that regular phone numbers actually boosted calls by +10%.

It is thought that some people were reluctant to dial an unusually short phone number.

Identifying High-Quality Sales Calls

Identifying which web visitors were calling proved useful, as there were many cases where looking at web analytics data on its own could be misleading.

For instance, it may seem that a web visitor had lost interest and exited. But really, they had just switched channel by picking up the phone and calling Total Direct Energie.

This often happens when customers have lots of questions or feel the purchase is complex. Speaking to a human is fast, efficient, and reassuring.

Understanding there was a call is useful. However not all phone calls are equal.

Customers will call on many issues including support calls and for simple, quick questions. Careful analysis revealed the types of call that were ‘good’ calls:

  • How long was the call? If it was too short, it could not be a successful call.
  • What drove the call and which web pages did they visit before calling?

Knowing which calls were high-quality sales calls meant Total Direct Energie could more accurately:

  • Predict people’s intent before they even called.
  • Identify and optimise successful marketing campaigns.
  • Build more effective retargeting audience.
Read “Predicting Caller Intent” for reasons why more companies are looking to understand caller intent before the call.

Using Offline Revenue for Online Targeting

Total Direct Energie use sophisticated digital marketing for customer segmentation, including segmenting customers before they visit the Total Direct Energie website.

They develop lookalike audiences to identify people with similar characteristics to previous customers, so they know which advertising campaigns are most likely to be effective for them.

Where the customer journey is purely digital, they can use cookies and digital ids end-to-end to tie visitors to advertising platforms, web visitor sessions, and conversions on Total Direct Energie’s internal systems.

To make this work for all customer journeys, Total Direct Energie needed to integrate offline phone call data into their online data, by:

  1. Tying customer transactions in Total Direct Energie’s CRM system to the specific phone call with that customer. 
  2. Tying that specific phone call to the digital ids used by the sophisticated range of marketing and advertising technologies they use including Google Ads, Google Analytics, and Content Square.

By ‘closing the loop’ Total Direct Energie can optimise their marketing to focus on the campaigns and audience that drive the most revenue.

Using Content Square enables a much richer understanding of the customer’s journey and how to improve the usability experience. This helps improve the effectiveness of the website for the customer types likely to purchase, encouraging them to sign up online or over the phone.

You can read through the presentation Total Direct Energie gave here, in French.

Predicting Caller Intent

Predicting Caller Intent

Written By Tim Wirth

Today and tomorrow, Paris is hosting VoiceTech — an annual event discussing ideas and innovations around voice technology. 

A key theme is brands can really gain insights from phone conversations to improve the customer experience and increase sales revenue.

This year, VoiceTech includes some great talks covering a wide range of topics covering all aspects of voice technology, including: AI, IVR routing, biometrics, sentiment analysis and voice bots. 

These are technologically sophisticated solutions that help companies identify people’s intent after they have spoken, and can be used during or after phone calls.

Understanding Phone Calls, Before They Even Start 

The words people speak, the way they say those words, the timing, the tone of voice… can all produce a mass of data to be digitised and analysed.

To gain a fuller, more complete picture of the caller’s intent, you can look at what each caller was doing beforehand, and what drove them to call you.

  1. If they have been on your international luxury holiday website searching for holiday locations and have spent the last few minutes looking at pictures for one specific location — they probably want to book a holiday there.
  2. If they have been looking at the FAQ pages on your website to see how they can switch from another energy supplier — they probably want you to become their energy supplier. 

Understanding the caller’s online behaviour before they call you, gives you a more complete picture of their intent and gives you the opportunity to improve the customer experience.

For instance, in the first example you could send that caller to an agent who speaks their language. And if they have started an online booking process you can pass that booking to the agent.

Similarly, if you are the energy supplier, you may decide that a new customer switching to you takes priority over a support call, so their call is answered next.

Club Med and Total Direct Energie

Two leading companies, Club Med and Total Direct Energie, are innovative customers proving how valuable it is to understand caller intent.

By combining data from website visits, CRM systems, marketing technologies, and phone calls they have both delivered superior customer experience and commercial results.

Speakers from Club Med and Total Direct Energie will be sharing their experiences at VoiceTech.

Reasons to Understand Intent Before the Call

  1. Training Machine Learning
  2. Call Routing
  3. Prepare Your Agent
  4. Marketing Attribution

1. Training Machine Learning

The technology for understanding speech is progressing rapidly, especially around machine learning. 

One of the key practical issues we are seeing with machine learning implementations is how to build high-quality training data that is specific to a client’s business context, terminology, and languages.

As our Head of Product, Jeremy Hamill-Keays explains in “Learn to Make Stew Before You Implement Machine Learning” machine learning for voice and speech, requires high quality training data that is:

  • Very specific for each client.
  • Collected easily.

The more holistic the data, the quicker the machine will learn. Companies are looking to implement new strategies quickly to improve customer experience.

Long waiting times, busy agents, and repeated questions all add effort and friction, driving customers to look elsewhere.

Capturing the caller’s intent and recording the calls is a practical, effective way to build a high-quality training dataset. There are multiple, scalable ways to digitally capture intent data for instance:

  • Predicting intent from the caller’s online journey.
  • Asking a sample of callers to choose their reason for calling from a call to action widget.
  • Asking agents to manually categorise calls.
  • Sending callers a post-call survey widget.

2. Call Routing

The traditional method for routing inbound calls involves an IVR (interactive voice response). This legacy technology is very unpopular with:

  • 98% of callers wanting to avoid an IVR.
  • 40% of callers abandoning a call within the first 60 seconds of an IVR call.

By predicting caller intent and using a modern cloud-based call routing, each call can be routed directly to the most appropriate agent or queue.

This gives you the flexibility to route calls:

  • To give highest priority to the calls that are highest value or most urgent.
  • To direct calls to people with the right skills to handle that call.
  • To give new agents calls that are easiest to answer.

3. Prepare Your Agent

Understanding why someone is calling means you can prepare your call center or sales agent before they even answer the phone.

Presenting agents with contextual information helps them to have better, more successful sales conversations. For instance, you can show your agents:

  • Details of the products they have been looking at.
  • Information from their previous calls or interaction with your company.
  • Where they are calling from and if they are using a mobile phone.

Providing agents with insights into intent can help close sales more quickly, and it can help provide a superior customer experience.

4. Marketing Attribution

Full visibility over the entire customer journey, from responding to a digital ad or billboard, to browsing your website, through to making a phone call and purchasing in a retail outlet allows you to attribute and optimise your marketing.

Marketing teams need to know:

  • Which campaigns, ads and keywords are driving calls?
  • Which calls result in a positive sales outcome?
  • The actual revenue generated.

By tying that data together, you can optimise your marketing campaigns by investing in the most effective channels, campaigns, and keywords.

Coming to VoiceTech 2019?

If you are coming to VoiceTech 2019, see how Total Direct Energie and Club Med are leveraging pre-call caller intent data to improve customer experience and commercial results.

“Looking for Growth? See How Total Direct Energie Benefits from Voice Data”François RommelHead of AnalyticsTotal Direct Energie
26 November 2019Salle Ateliers B15:30 – 16:00
“How Club Med’s Omnichannel Strategy is Driven by Voice”Amina ChaabaneHead of Digital Sales B2BClub Med
27 November 2019Salle Ateliers B14:00 – 14:30

Please stop by stand D03, and find out how you can understand more about the customers calling your business, and how you can improve customer experience and commercial results.

Real Estate Trends 2020

Property Portal Watch Team

Written by Martin Aarestad and Jakob Kolsmyr

Last week we attended the Property Portal Watch conference in Madrid. 400 real estate professionals from around the world came together to share ideas, trends, and best practices around some of the burning issues facing real estate today and in the future.

We also had the opportunity to have insightful conversations with some of our customers, like Paul Wicks (shown), EMEA CTO at JLL, and with a wide range of other real estate experts.

Key themes people were discussing for 2020 included:

  • The shift towards a hybrid model
  • The focus on iBuyers
  • Getting closer to the transaction

Digital Transformation Has Started. A Bumpy Ride for Some

For the last 5-10 years the real estate industry has predicted a digital transformation to cater for the ever-increasing needs of tech-savvy and informed consumers who want to take control over the buying/selling process, instead of relying solely on estate agents.

Startups are disrupting markets globally and are proving that a fully offline experience is no longer viable. These startups don’t have the legacy technology and process issues that established players have to contend with. This means they can deliver a digital customer experience to attract buyers and sellers to their websites and apps, provide all the property information required, and manage viewings — completely online.

But, a fully online experience is not an option either. The woes of purely digital real estate agencies here in the UK, such as Hatched and Emoov are well documented, and even PurpleBricks has seen their share price tumble.    

Hybrid Required — Digital and Human

As we heard during the event, the real estate industry is shifting towards a hybrid model. This model uses digital tools combined with human interaction.

Buying or selling a property is a high-value, complex purchase and people want to engage with a human. They need to talk to an agent face-to-face or over the phone, to proceed.

Both buyers and sellers value the trust and safety offered by a human real estate agent. The agencies having the most success are training their agents to be more like consultants. A good example is Swedish startup Norban, as founder Adrian Miller explained during his excellent presentation “How Hybrid Agencies are Disrupting the Traditional Real Estate Industry”.

By working in a closer relationship with agencies, portals will get closer to the transaction. This symbiotic relationship will open the door to a change in their business model and the possibility for portals to offer new additional services to customers.


One way for property portals to get closer to the transaction is to actually purchase properties as an ‘iBuyer’. In many markets around the world, iBuyers are challenging the traditional ways residential properties are bought and sold.

Using extensive data and technology iBuyers value your property and offer to buy it at a date you choose. They offer simpler, more convenient alternatives to traditional home selling.

  • Saving time and hassle — no viewings, decorating, repairs at the last minute.
  • More control over moving dates — reducing other costs like bridging loans.

The main challenge with this model is that is very cash demanding, since you are buying the properties yourself and turning them over.

iBuyers are taking risk and providing a valuable experience for sellers in return for a greater margin. This market is becoming increasingly competitive and uncertain, with more pressure on margins.

To be fully successful, iBuyers will need access to extensive data on properties, sales, liquidity, demand, and real insights into buyers’ profile, their behaviour and purchase journey. Success requires turning human and digital customer conversations into actionable data to sell properties quickly.

Closer to the Transaction

The common driver of the models listed above is that all real estate players want to get closer to the transaction. They want to offer more value-added services to provide a superior experience and improve their margins. Many real estate portals (marketplaces) may move from a subscription-based model to lead-based, moving closer and closer to managing a part of the transaction.

As new models emerge, they disrupt traditional, established business practices across different markets. Real estate market leaders are being challenged by fast moving startups that are getting closer to the transaction and generating more value for the agencies. The market players offering the clearest value will ultimately have the most control.

Improving the Customer Experience

The digital transformation taking place in the real estate industry is showing parallel trends we are seeing with marketplaces and portals in other industries, such as luxury and automotive.  Even though it will be difficult to move to a 100% online transaction, there is plenty of room to improve the overall experience for buyers and sellers.

Over the last 5 years, Freespee customers including some of the most successful real estate services firms have been leading this digital transformation. They have been supporting users throughout the entire buying journey, and providing agencies with full visibility from lead generation and media attribution, through to lead management and actual revenue.

Interested in digital transformation for your real estate services? Get in touch with us.

Learn To Make Stew Before You Implement Machine Learning


Written by our Head of Product, Jeremy Hamill-Keays

In the Rhondda Valley, in darkest deepest Wales, there are certain questions which have only one answer.  One of them is ‘Who makes the best cawl?’  The answer, of course, is “my Mam”.  As you’re probably wondering, cawl is a hearty stew beloved by the Welsh.

With winter drawing near here in Sweden, I started to long for a bowl, so I phoned her up and asked for the recipe.  It’s not a secret so here it is; I highly recommend.

  • Pot of water
  • Chopped up neck of lamb
  • Whatever root vegetables are in the cupboard
  • Parsley
  • Salt/pepper/herbs to taste

I was slightly confused; how could it be so vague?  Shouldn’t it be exact weights and measures?  My mother scoffed, and told me then it wouldn’t be cawl.

Making cawl before Wales’ next big game (World Cup 2019)

The Need for Training

At Freespee, we work heavily with providing a frictionless user experience including routing calls directly to a destination such as a high value sales queue. Freespee contains many services to optimise sales leads and one of these is predictive routing.  

Developments in machine learning are making large impacts and have huge potential to drive customer experience improvements; that gold member needs to be looked after.

While machine learning has many advantages over rules-based systems, it suffers from a couple of  major drawbacks. Before you can use a machine learning technique, the system needs to be taught, and obtaining training data can be difficult. It also requires that multiple sources of data are used to reach a level of sophistication required in the modern digital world.

There are generic machine learning engines which claim to do things like call categorisation “out-of-the-box”, however real-world testing often proves that to be inaccurate. They are not adapted to any one customers data set or business.

Good, accurate training set data is essential for commercial use and this is what holds back many machine learning projects. It is important to have multiple sources of data.

Your Recipe is Unique

So, what has this to do with Mam’s cawl?

Well, a little like her recipe, machine learning requires a mixture of ingredients all working in harmony; with the exact ratios changing from company to company, and over time. 

Get it right and it works really well, which can make a significant impact on your business.

At Freespee, we have concentrated on ensuring that the data needed for machine learning engine implementation can be easily collected in many ways.  

As an example, a speech analytics call categorization service can be trained with data derived from the pre-call Call To Action (CTA) widget, which asks a percentage of visitors what’s the reason for their call. 

The CTA improves the visitor experience and is a low friction method of data collection. It can then be combined with data from CRM, IVR, or from “outcome” fields, completed by agents using Freespee softphones such as Freespee Talk and mobile apps. 

Each company has unique needs, so we feel this holistic approach is better; technology adapts to people, not the other way around. 

So if you are offered a speech analytics solution, ask – how difficult will it be for a training set to be built and does it match your process? How many ingredients or data sets are used to optimise the engine and provide an ability to adapt to an ever changing environment? Or better yet, get in touch with us to find out more.

How We Can All Learn From Lost Luggage

Talk Web Version

Written by our Head of Product, Jeremy Hamill-Keays

A friend of mine posted a photo from a recent trip to our “lads” Facebook group. His airline took his, and just about everyone else’s, cabin bags and put them in the hold. The reason, they announced, was the lack of room in the cabin overhead bins. As you can see from the picture below, the flight took off with the cabin storage nearly empty. Needless to say, my friend made it to his destination, his bags didn’t, and the particular airline didn’t exactly get a high NPS from him.

But, is the poor customer experience in the example above confined to flights? If you look closely, you will see this situation replicated throughout your everyday life. 

Each day, millions of people visit websites looking for information or to make a purchase. Still, the majority of people buying something who have a question, prefer to call the relevant company. Data shows that 54% of visitors prefer to speak with an agent than type in a web chat, a figure that will keep on growing, especially as the mobile phone has become the main device used for web browsing.

The web visitor, becomes a caller, and is then routed through an IVR (Interactive Voice Response) menu system, an automated process that provides a never-ending list of destination options. IVR menus are unpopular with callers, causing many of them to abandon, and many reporting the dial-in options are unclear. After pressing the required digits, the caller is then connected to an agent, who starts the conversation with a long list of questions to understand the query and provide the required help. 

If this customer journey was a flight somewhere, it would be equivalent to having to take multiple flights, instead of a direct flight. We would be required to check-in our hand-luggage only to have it get lost at every flight transfer. Maybe this is the way forward to combat climate change, but as an experience is it not one any of us would aspire to. 

At Freespee, we like direct flights and we like to ensure the luggage arrives to the destination with you. Great service relies on quickly helping a customer finish a purchase or get the answer to a query. Many web chat systems now send the web visitors’ browsing session information, also referred to as online context, to the agent in the chat. However, when it comes to the phone channel, many people assume there is little that can be done. However, there is absolutely something that can be done and by taking some simple steps you can vastly improve the customer experience.

As an example, a person who has completed an order form but is stuck on the payment page is most likely calling to complete the order. Freespee uses the information from the web journey, the caller’s online context, to predict their intent and automatically route this particular call straight to the high-value sales queue, completely removing the need for an IVR. This seamless call experience removes friction for customers, while it helps agents exceed their sales targets and glow with success.

Here at Freespee, we have been helping our customers accelerate sales opportunities in this way for many years.

The functionality doesn’t just stop at removing the IVR. Agents can be shown, in real time, the caller’s online context, which helps turn painful interrogations into pleasant conversations. This personalisation promotes great customer service and delivers a highly individualized buying experience. Agents are also happier and far more engaged with the caller. Customers remain loyal and NPS scores benefit.  

The final upside is shorter average call handling times (AHT). Some of our customers have reduced AHT by almost 50% and have cut costs massively by no longer having to ask questions.

Freespee has several implementations to allow predictive routing and context to be shown to agents, so regardless if your current solution in old or new, or if you want a new softphone solution, our team will be happy to help you.

Want to learn more about our sales acceleration features? Take a look here. Happy landings all.