Check our Ultimate Chatbot Guide
5 Steps to Successful Cooperation with Chatbot Clients

5 Steps to Successful Cooperation with Chatbot Clients

Adam Szewczyk 28.03.2019
Adam Szewczyk

Whenever we start cooperation with a new client, we make our best to deliver the ordered chatbot in the shortest possible time. Imagine that creating a chatbot is easier than building a website, that all the necessary tools to make the bot are ready at your disposal in only several days. That’s how it will look like if you follow our steps. The process we have created (not to be confused with procedure ;)) will allow you to quickly and effectively cooperate with clients during the chatbot creation and implementation process.

Step 1. Welcoming Brief

Customer service starts even before a given person becomes our client ? We treat this step as a handshake. When we meet a new brand that needs a chatbot, we try to clarify any doubts and uncertainties. We use a specific brief to do this. The brief allows us to walk into the shoes of our client and to understand the brand’s products or services. Besides this, it also informs us about the client’s requirements and expectations towards the chatbot. At this point, we want to understand the problem we need to solve, get to know the target chatbot users, and determine whether the tool will work in a communicator, intranet, or on the brand website. We also ask about the expected deadline and budget. All these questions have one specific goal – to find out what will fulfill the client’s business needs. That is, what chatbot features will be helpful in attaining brand strategy goals, and which elements should be omitted to optimize the time and costs needed to finish the project. This is also a good moment for the client to ask us technical questions. With smooth cooperation, we can have the brief ready in just 1 day.

Step 2. It’s Time to Make an Offer

With the knowledge we have derived from the brief, we can prepare a personalized offer and a demo chatbot for the client. We can have the demo ready within an hour thanks to our own platform for creating chatbots. The client can quickly find out how the chatbot works, even before he makes his final decision about cooperation. This allows both sides to immediately verify their expectations, and assures them that the project is going in the right direction. Our offer also includes an estimated implementation budget and work schedule. One day is enough for us to make the first document proposition.

Step 3. Debriefing and Contract – Developing Closer Relations

This is the stage where we establish the scope of work and terms of cooperation. If needed, we also ask the client’s IT, marketing, or social media departments additional questions. We want to make sure that we have clarified everything that could affect the implementation of the project, especially with regard to GDPR. Experience in cooperating with big brands will definitely pay off at this moment, because from now on, you’ll know that everything will go smoothly.The only thing left to do is to sign the contract, which includes the final budget, scope, and schedule of work. This can be done even within 2 days.

Why is everybody talking about chatbots?
Because brands who use them in business are getting results
Let's talk

Step 4. Now We Know Everything – Let’s Make the Chatbot

We start the project with a kick-off – a meeting where the client receives all materials necessary to finish the project. Usually, they include content, photos, all the necessary passwords, and API (Application Programming Interface) data exchange documentation with the client’s IT systems.

Already at the beginning of our work, we tell our client how we promote chatbots and why it is important. Thanks to this, you can plan marketing activities in advance, as soon as your chatbot is be ready.

Starting with a demo version, we create the chatbot in iterations (subsequent, improved versions of the bot). We write user conversation scenarios, add more content, create necessary connections, and intelligent responses – the messages, which the chatbot will use to answer users’ questions. At the same time, our developers implement dynamic elements and create mechanisms that will exchange data with our client’s IT systems. All this is to make sure that the chatbot we provide will have features specified by the client.

You’ll learn more about building chatbots if you read this article by Ola: How I create chatbots – practical guide

After tests, we show the current versions to the client to keep him informed about our progress and to let him asses our working line. Close cooperation and feedback are essential at this stage. They assure us that we’re going in the right direction and allow us to instantly respond to comments and quickly introduce necessary changes.

Of course, the time we spend on producing the chatbot depends on its complexity. For example, a simple chatbot that responds to questions and sends offer messages, without the need for complicated IT integrations, is created within 2-3 days. A more advanced API integration can lengthen the time by another 1-2 days.

Step 5. Putting Our Cooperation to a Test – Launching the Chatbot

We launch the finished and tested chatbot on a chosen communicator with just a couple of clicks. In order to launch the chatbot on our client’s website or intranet, it takes us several more clicks to generate an appropriate HTML code. And that’s how, a moment after we have finished production, the chatbot is ready to meet the rest of the world. Only several days have passed by ?

What Next?

Building and launching the chatbot is only the beginning. After this, KODA BOTS clients receive full support – our chatbot team monitors how well the chatbot holds conversations with users and constantly makes necessary improvements. Each month, the team receives a report, which shows how people use the chatbot and provides recommendations for further development and changes. In the meanwhile, designated members of the Customer Success team stay in contact with clients, helping to address their needs.Thanks to this, we can move forward to the next iterations … ?