The 3-Step Guide to Scoping a Chatbot Project That Gets Results

We´ve all tried to contact a company looking for support or help with an issue, only to be met with a conversational bot (chatbot or voicebot) that simply doesn’t fulfill its purpose. You end up closing down the chat or ending the call frustrated and no better off than when you started. 

Is this really the experience you want your customers to have when they interact with your conversational solution? Well, unless you have followed the advice shared below, it´s likely your customers feel the same way…

The reason that so many companies fail with their chatbot project, is that there are so many individual issues that need to be planned for. 

  • The proper platform to work on 
  • The right team 
  • Project growth and expansion plans in terms of languages, channels, and geographies 
  • Scalable infrastructure

However, one of the most fundamental ingredients is defining the content that needs to be included in your conversational bot, or in other words determining the scope of your chatbot project.  

All too often projects fail because too much effort was put into implementing features and content that in the end, nobody used, because they failed to define the scope of their project from the get-go.  

Let’s dive into some insights and tips on how to scope your project to make your Conversational Bot truly helpful. 

1. Define Your Chatbot Project Objectives 

Before you start planning your chatbot content, you must first answer the question: 

“Why do you want to implement a Conversational AI solution in your company?” 

Until you can answer this question, your project can’t move on to the next step. 

Successful projects are usually those that have identified a clear use case and drivers aligned to company or department objectives – so all energies can be funneled in one direction.  

Objectives may vary between companies but typically there are three non-exclusive but complementary objectives: 

  • Increase customer satisfaction through lower waiting times and quick resolutions to customer issues 
  • Drive cost efficiencies in the call center by freeing up valuable agent time 
  • Boost sales conversions by helping the user more effectively and reducing abandonment rates

chatbot project

Let’s take a look at an example 

Throughout this post, we will use the example of a retail company. 

This retail company is in an expansion phase and wants to avoid a growing customer service cost. 

To achieve this, they plan to implement a conversational bot that automates the handling of certain user questions, thereby freeing up the time of their customer service agents.  

This will secure customer satisfaction by giving an immediate answer to those using the bot so agents can better serve the customers with complex queries. 

So, their objective in its simplest form is: “Expand without growing customer service costs by implementing a chatbot” 

Still, this is not enough to start their project.  

The next step is to evaluate if there is a business case that justifies the investment 

2. Business Case Evaluation 

To build a business case, you will first have to list out all the different questions (user stories) asked in the Call Center.  

You don’t need complex tools here. A simple spreadsheet will do the job.   

Once you have the list, you need to map them out based on three factors: 

  • Frequency: How often is the question asked? 
  • The process: Are you able to automate the answer or its resolution? 
  • Complexity: How many resources are required to automate it? 

chatbot project

Let’s take a look at each factor in turn.  

Frequency 

Generally, all call centers have statistics on the type of questions usually asked, so this information should be readily available for you. 

If you don’t have this information though, don’t worry. A workshop with a few agents and team leaders will give you all the information you need. 

When analyzing call center statistics, you’ll quickly realize there are a few questions that clearly stand out over the rest. This is due to the Pareto Principle, also known as the 80-20 rule, whereby 20% of the calls drive 80% of the volume  

You would be surprised to see how this rule is consistently met irrespective of industry. 

The process 

The second question is a very fundamental one:  

Will the conversational bot be able to provide the same answer as your Call Center Agents would?  

If the bot provides a general answer that doesn’t resolve the query, the customer will eventually call the contact center and you are back to square one – with frustrated customers. 

Following our retail company example, this completely goes against the objective of reducing customer service costs.  

This is not a situation you want to be in. 

During this step, you’ll quickly realize that some answers will be straightforward to provide and others will require additional backend or 3rd party integrations to fetch the right information.  

Imagine a customer on our retail company website asking the bot: “are there any in size M?” 

The Conversational Bot will have to check stock to provide a final answer, which is the only way to serve the client and save the company a call. 

Complexity 

This leads us to the final question. You need to understand the complexity of automating an answer that is connected to back-end systems.  

In other words, how much effort is required to get the bot ready for this question: 

  • What integrations (if any) are required? 
  • How difficult is it to put them in place? 

There may be some cases where you don’t want to automate responses, but you still require integrations to be in place. 

For example, if a user complains about your service, you should hand over this conversation to a real agent. So, even though the bot isn’t dealing with this query itself, it needs the ability to connect the user with the live agent provider. 

Or if the bot can´t answer a query, you still want it to record it in the ticketing system. 

With Teneo, get support with API integrations when you need it, and make those high complexity questions that little bit more achievable.  

Assigning savings and value  

By now, you’ve listed out all the questions asked in the call center and assessed them based on the frequency and effort to automate.  

To close your business case, you now need to map each of them with the monetary value you think they are going to provide.  

This is now an easy task as you’ve done all the preparation work already. Here are some examples:  

  • If you are fully resolving a client’s question: You should associate savings that are aligned with the cost of one call center phone call 
  • If you are not resolving but directing the query to cheaper channels such as live chat: You should associate savings that are aligned with the cost of live chats 
  • If your objective was to increase sales on your website: You should assign a value generation based on closing a purchase on your website 

In our retail company example, they would now have the full picture of how to expand their customer service footprint without needing organic growth.  

They now know how much they need to invest and the return on that investment (ROI). 

3. Plan the Chatbot Project 

Having completed steps 1 and 2, you have essentially created the backlog of your project.  

The project objectives have been defined. The business case has been evaluated. And you have a list of questions categorized with the value they will bring should you automate them.  

Now it’s time to plan the project. 

At this point, you want to involve a broader team to prioritize the backlog and to make sure that you are splitting up all the backlog items into different phases.  

We have a full article on building teams for each phase of a conversational AI project. But, our recommendation is to start small. 

Start your project with a small amount of content – the high frequency, low complexity area in the bottom right of the diagram above, and work your way through to the top left in stages.   

Focus on the low-hanging fruit. This will allow your organization to learn by doing while providing measurable quick wins 

It’s important to have clear KPIs that prove the value, reinforcing the benefits of developing an intelligent conversational AI solution. 

This will also help get full internal support for further phases. Check out our article on identifying the value of an AI chatbot to support you with this. 

Now the chatbot project is ready to kick off. 

If you’re implementing a conversational AI project for the first time, you may find that once you go live the behavior of some of your customers changes.  

This is a dynamic effect where some customers will stick to traditional channels, while others shift to the new ones or a combination of both.  

On top of this, expect the overall number of contacts in all channels to slightly increase as you will be reaching out to more people. 

Conclusions 

This article has outlined the basis you need to make informed decisions on where you should invest your resources to implement a conversational AI solution – from identifying the use case to mapping each user story and assigning value to it.  

The result is a clear business case that will allow you to start a solid conversational AI journey by aligning the organization’s resources to drive the project in a clear direction. 

This is the only way to be successful. 

When you build with Teneo, get the support you need to execute your first use case and start providing exceptional automated support to your customers.   

Get in touch to find out how we can support your organization with your chatbot project.