Chatbot: your digital colleague for over 50 years
Watermelon has grown enormously in a short period of time; we’ve experienced different phases both with regards to the company and to the market. A market where the chatbot was still ‘scary’ and relatively unknown, and a market full of enthusiasm about chatbots wherein companies wanted to try out new technology. This is the market that we’re dealing with right now. We are clearly one of The Netherland’s key players in the chatbot market. Our client base has grown enormously and we are active in almost every field in more than 30 countries. This gives us the chance to share some good examples of how our chatbots can be used, so we have compiled a number of interesting client case studies.
Watermelon is continuously evolving. We pay very close attention to developments in the market and try our best to tap into them. What we’ve noticed is that many companies are interested in the possibilities that chatbots offer but want to keep the risk low. What’s more, many new providers have emerged in a relatively short period that haven’t developed their software enough and are often a bastardisation of technologies that have been available since 2005.
Watermelon remains the most user-friendly platform, but was not accessible to smaller companies until now due to our pricing and intensive onboarding. After being on the market for a year and a half it was time to reflect on both the market and our own business model. We concluded that we wanted to offer a product to a wider market. Every company should be able to make use of a chatbot on their website. Company size shouldn’t matter. That’s why it’s time for us to open a new chapter: Hyper Growth. We’ve revamped our product and our team for the Hyper Growth that we are going to experience (more on that later), but first: I want to share my reflection on the market with you. What exactly does that chatbot look like?
Developing new technology
In order to describe the chatbot market, it’s important to take a step back and look at how this technology came into being and how it has developed. Gartner developed a Hype Cycle for new technologies. When new technology comes onto the market, the question is always whether people will need it and whether it’s tenable.
Before we look at where Chatbots are on the Hype Cycle, it’s nice to get an idea about which five stages a technology goes through:
Chatbots on Gartner’s Hype Cycle
The Hype Cycle starts with an Innovation Trigger. This trigger is a technology or another innovation that arouses interest. The commercial viability of the technology hasn’t been proven yet, therefore not everybody will pick up on it straight away. To get an idea of what this means we will look at a current Innovation Trigger: new developments with regards to a self-flying car. Is this a development which the world is ready for? No, absolutely not. The airways are not ready for it, people aren’t ready for it, factories need new machines… The list goes on. This doesn’t mean that it doesn’t have the potential to be an enormously impactful technology in the future; it may change the way we transport ourselves and our goods.
After this trigger, more parties will gradually gain interest. The popularity grows and the commercial viability is confirmed. Success stories start emerging. These successes will dominate the media until the dust settles. It’s comparable to how you feel when you get a new mobile phone. Initially it’s new and amazing and you’re happy with it. The more you use it, the more disappointing things you encounter. A camera that isn’t as good as the ads claimed and a memory that fills up super quickly after all. This is also how it goes on the Hype Cycle. At the peak of the enthusiasm, critics start to appear. This goes together with negative media coverage and stories of failure. The new technology appears to not live up to the high expectations and the popularity will decline after this peak. This peak is called the Peak of Inflated Expectations. The quicker the rise in popularity, the quicker the decline. High peaks lead to low valleys. A number of companies keep working on their technology and take action, whilst others choose to avoid the risk.
After a lot of negativity, the technology will reach its lowest point, a valley known as the Trough of Disillusionment. The experiments done with the technology do not lead to the desired results. Producers of the technology fail or decide to stop. Investments keep on coming, but only for the producers that keep improving their products for the Early Adopters. This is a group of people who are the first to adopt a technology or trend. They always want the newest of the new, regardless of whether the quality has been proven. Despite the negativity, there’s always a light at the end of the tunnel, namely the thought that although the technology cannot live up to what we initially thought, it is still an interesting development. A technology currently at this stage is Augmented Reality. Augmented Reality is extremely popular, but it could take another 5 to 10 years before it really takes over a spot in the market.
Even in the trough, there will be investors that believe in the technology and producers who are committed to improving the products. This will eventually be fruitful. Ways that the technology can help companies will come to light and there will be a wider understanding of the technology. More and more companies will see this and you will notice that more producers start appearing. This growth is slower than at the beginning of the hype, but it is stable. This is known as the Slope of Enlightenment. This is an incline that the innovation scales due to ‘real’ success stories.
From the Slope of Enlightenment you will reach the Plateau of Productivity: the technology will be embraced by the mainstream. Moreover, the viability will be clearly defined. The technology is stable and multiple companies discover the advantages. 20 to 30 percent of the potential users are already using the technology. It’s also now clear whether the technology can be widely used.
For many, the word hype conjures images up of something short-lived. The fidget spinners that are all in our garbage bins now, the weirdest planking photos and the Fortnite dances are all rages and not hypes. It differs per technology how long it will take to complete this Hype Cycle, if it is ever even completed.
Chatbots on the Hype Cycle
An innovation starts with a Technology Trigger. For chatbots, this trigger wasn’t the first chatbot- to find the real trigger, we have to go further back. In 1956, Nawell and Simon invented the ‘Thinking Machine‘. The first program that simulated the human quality of solving complex problems. This is the first program with Artificial Intelligence. This development is special, of course, but not easy to understand. The question remains: is society ready for this development?
The first chatbot went online in 1966, her name was ELIZA and she was born almost 20 years (!) before the internet as we know it came to be. After ELIZA’s creation came success stories where everyone was seemingly enthusiastic about the ‘new’ technology. Consumers start to hear about it and interest rises. A.L.I.C.E. (1995), a natural language processing chatbot based on ELIZA, won the Loebner prize three times. A prize based on the Turing test, a test that determines how human a chatbot is. In every round of the Loebner Prize, a (human) jury member will have multiple conversations with a chatbot and a human via a computer. Based on the answers, the jury will try to determine which conversation partner was a computer and which was a human being.
In the short time (15 years) after A.L.I.C.E., all kinds of chatbots start to appear. Do you remember Clippy? Every time you opened a Word document, Clippy would come to help you and tell you where the recycling bin was. Since then, Microsoft have found the recycling bin themselves and Clippy disappeared into it.
During the age of MSN, it became clear how chatbots could also be used for fun or as a social tool. Dutch mobile phone provider HI promoted Chatman heavily and the commercial is still stuck in my head. This allowed us to get to know chatbots in a fun, approachable way. We could also chat to Marco Borsato in chatbot form through MSN.
Despite the fact that the Hype Cycle denotes a clear line, this line can fluctuate during the cycle. In April 2016, Facebook Messenger launched a platform for developers which allowed them to create a chatbot. Through this, more consumers come into contact with companies by chatting with a bot. This was kind of a trigger which allowed people to get closer to the technology, causing interest to grow.
Around this period, Watermelon began researching and developing. This happened around the same time that Ecreation was taken over by Watermelon. Watermelon created the first generation of the product. Randstad was our first success story; we built a custom made chatbot. We learned a lot from this. We learned what we at Watermelon want to offer and what the business market is looking for. Parties such as IBM and Microsoft have already offered possibilities for building chatbots. At Watermelon, we want to make this technology available for everyone, without the need for knowledge of coding. We launched the first version of our Chatbot Management Platform in 2018.
A trigger like the one caused by Facebook Messenger can cause a peak. A peak during the starting phase of product development almost always means the same thing: the way down is easier than the way up. The media exacerbates this hugely. Watermelon’s (and other providers’) software was in its starting phase at that point. There is a large amount of interest in chatbot building software. But with all of this interest comes a lot of critique: headlines such as ‘Why are all of these smart chatbots still so dumb?’ start appearing.
This negative attention always seems severe, but the demand from the consumer remains. Think about that video that Frans Bromet made in 1998. In the video, various people comment on how they considered a mobile phone to be ‘unnecessary’ or ‘a waste of money’. Can you imagine a world without smartphones today?
We at Watermelon realised this. Despite a number of critical headlines, we still believe in our product and what it can offer both businesses and consumers. We believe that Watermelon can make customer service better, faster, and more fun. We’ve been successful in this since the launch of our Chatbot Management Platform, and that’s been confirmed by our clients. Nummer Een was one of the first clients to try the first version of our platform. Chatbot Robin is still working as part of their customer service department. BKR Foundation was one of our first clients and their Chatbot improved their customer service noticeably. BKR’s chatbot has 300 conversations per day. Conversations that used to need to be done by an employee. Interest is rises and Watermelon’s place on the market is clear. A variety of small and large companies from different branches now know how to find us. From banks like Chaabi Bank to insurance companies like VGZ to schools such as the HAN and TIO. AFAS Software also communicates through a WhatsApp chatbot that they made with Watermelon’s platform. Soon you’ll be able to get all the information you need about pizza with Domino’s chatbot and employees of PepsiCo will be able to get all of their HR questions answered by their HR chatbot.
Koen Gebbink and Maureen van den Bergh from Stichting BKR, read the business case here.
The hype is over?
When you hear the words ‘the hype is over’ you may think it means that there is no longer any interest. This is not the case! It simply means that the technology is now stable. Watermelon, the market and consumers have learned their lessons from the past. Society is ready for this technology and the market connects seamlessly. We are now reaching the Plateau of Productivity after scaling the Slope of Enlightenment.
All-in-one conversational platform, start your free trial here
The first platform only offered the possibility of adding Frequently Asked Questions. With the knowledge that we gained form our first platform, we managed to launch our second generation in 2018. We added the possibility to build conversations. We learned a lot about how to further develop our product. We discovered that it’s not really about how smart your AI is, but rather about the content that you add to it. This is one of the reasons that we start a Customer Success plan with every client. Sometimes the lessons seem very simple, but they make a huge difference. We made the decision to make each chatbot reply take a few seconds, making it appear as though the chatbot is typing. The chatbot immediately comes across as friendlier.
The most important lessons contributed to the development of the Hybrid Model, the most distinguishing part of our platform. We realise that a computer cannot answer all questions and that sometimes you just need to speak to a human employee. To solve this, Watermelon developed a Hybrid Model which allows a live agent to work in tandem with a Chatbot.
The success of a chatbot is essentially determined by the content that the Chatbot is made up of. You can look at it like a new employee that needs to be coached. You teach your new colleague everything about his or her new workplace. When the employee starts to work independently, questions will keep coming: ‘Where can I print?’, ‘How do I address someone when answering the phone?’. It’s sometimes difficult to guess what will be asked, and the same goes for a chatbot working in customer service. Whilst you may be convinced that a customer will ask what the estimated delivery time is for their new couch, he may actually want to know whether he can come pick the couch up from himself. You haven’t taught your chatbot how to answer that yet. Using the Hybrid Model, this question will automatically be relayed to an employee. You can then tell your customer that he can indeed pick up his couch himself. In this manner, your customer is helped quickly and there’s a good chance that you’ve sold that couch! Win-win, then- both for you and for your customer.
The hybrid model is especially interesting in today’s market where there is a large amount of automation. When someone is expecting an answer immediately, the conversation can be started by a chatbot. More complex questions are passed on to a live agent.
The latter target group in particular is much larger than many companies think. Research by Capgemini Research Institute shows that 70 percent of consumers will gradually replace a visit to a shop or bank with a conversation with a Chatbot within three years.
It’s clear what the consumer wants and, with a good, stable platform already available, the question is: what are companies waiting for?
A glimpse into the future
If we think about the future, we realise that there are a huge number of great developments coming up. Gartner has already indicated that by 2020, 85% of customer service will be handled by virtual assistants. This means that your customer service department can truly focus on the consumer and help them as well as possible. Doesn’t that sound great?
The third generation of our Chatbot Management Platform is going to make a big impact. We realise that a chatbot can be of interest to any company, big or small. That’s why we’ve decided to make changes to our business model. This means that you can build a chatbot for just €45. We are much more than just a Chatbot Management Platform, so we’ve also changed the name to ‘All-in-one platform for online conversations’. We offer four plans: Forever Free, Watermelon Essentials, Watermelon Team and Watermelon Professional. The product is optimised to the point that every user can start using the product completely independently. We offer our Customer Success Plan with Watermelon Team and Watermelon Professional plan. This allows us to offer our user-friendly platform to every company, big or small.
Prices of an annual license, view all pricing plans here.
Our Development team has been working super hard to develop free fields. By adding free fields to our platform, the chatbot is able to pick up all kinds of data on its own. Data which you used to have to call or email about. Do you want to know the location of your winter tires? By verifying your license plate number and postal code, a chatbot will be able to help you! Or do you want to know which of your university courses are cancelled? Send your student number to the chatbot and find out right away. This allows an even larger part of customer service to be automated.
Chatbot Pat helps customers even better with Custom Fields
I get asked very often if Watermelon is capable of being combined with various Voice providers. Despite the fact that Voice is still in the starting phase of the Hype Cycle, we have been working with it a lot recently. You can integrate Watermelon with Google Home for example, which opens up new possibilities for many companies.
Are you ready for the conversation?
Watermelon has only existed for three years, but looking back you can really see how much our product has grown and developed. We have a real, tangible place in the chatbot market. We understand how the market works. We hear what companies want to see and we understand the needs of the consumer. With this knowledge and the lessons we’ve learned from the past (as well as the lessons we are learning on a daily basis), we’re confident of a bright future. We now have a stable product, meaning that we are continuing our growth. Watermelon is available for every company, there’s only one question remaining. We are ready for the conversation, are you?