Product Overview
I was recruited to assist in the development of streamlined processes that cater to the needs of the agents while handling interactions and providing a consistent experience.
The company has a contact centre platform that can be used as a stand alone platform or integrated with other CRMs and platforms. The redesign focused at the beginning on the integration part, where we had to make sure the agents could access all the functionality of the platform from a small I-Frame window provided by external CRMs like Salesforce. This project was created for desktop only, with the possibility of going mobile in the future.
Ancient legacy products
The company had an old virtual contact centre that was designed more than 15 years ago and was reskined a few years ago to adhere to the new branding.
None of the flows were clear or linear. The platform was an amalgamation of functions stuck together in a fairly organised layout; basically everything was visible on the screen at the same time, regardless of whether that information was needed or not at that specific point in time.
Doing some digging
This was not the first redesign initiative. Before I joined the team, there was another project in the works to revitalise the old platform, but it was shelved in the middle of development, and all of the designers who worked on it, left the company. But it was not all bad, as I could find most of the concepts and research that had previously been done.
In order to understand how the old platform worked and how it correlated with the concepts I found from the previous initiative, I had to reverse engineer everything by assessing the information architecture and re-mapping the user flows.
Analysing previous research
As the previous team already conducted user interviews and I also had experience on contact centre products and conducting research for this kind of product, I considered it enough to start working on the MVP.
Learnings
When designing products, we always have to keep in mind the environment in which the product will be used. In contact centers, agents are usually trained on a specific product, and with time, they automate most of the actions they take to become as efficient as possible. The agents usually use multiple platforms at the same time, each with its own specific role, and the work environment is fast paced, loud, and distracting.
Main objective
We needed to develop a solution that could be integrated with other platforms that the contact centre might be using and, at the same time, encourage agents to perform only one task at a time and avoid adding friction that could hinder the agent’s performance. To help streamline the agents' work, I broke the interaction into five simple steps:
- Improve the incoming interaction panel
- Improve the pick-up and navigation
- Solutioning
- Disposition
- Next status
After I identified those steps, together with a dedicated UX researcher, we created a couple of scenarios based on which I created prototypes to test key flows that agents have to perform on a daily basis in moderated interviews with customers and internal users.
When every interview was done, the UX researcher in our team would compile the data and discuss it with me. In most cases, I made small changes to the prototype based on the feedback received after each interview and compared the before and after results for usability and other factors.
The project started at the beginning of 2021, and we divided the work into four phases, one for each quarter of the year. At the end of each phase, we would present a high-fidelity prototype alongside an explanation for our design decisions backed by our extensive research findings.
In the first phase of the project, we covered voice and phone interaction; in the second phase, chat and email; and in the third and fourth phases, we refined and improved our initial designs. At the end of the four phases, we released an Alpha version, followed by a Beta version, to some of our customers.
Iteration and innovation
The Alpha and Beta releases took about a month each, and during this period we asked for and gathered as much feedback as we could from the customers enrolled in the program.
-
Improve the incoming interaction panel
I learned that all contact centres are unique, and we needed to build the incoming panel to be very flexible. Agents need to see on the incoming panel the name, email, or phone number of the person trying to initiate the interaction and a timer, as most contact centres have a limited amount of time that the agents have to pick up the incoming interaction (e.g., 15 seconds). I decided to show that timer as a line around the accept button, or a bar at the top if there is no accept button, that would change colour from green, yellow, and red as time moves on. Some contact centres need to display more data on the incoming panel, while others need less, so to cater for this requirement, I opted to have a "show more" or "show less" option to display data if there is any outside the mandatory fields.
-
Improve the pick-up and navigation
Agents can pick up multiple interactions, some from multiple channels, and most of them do handle multiple interactions at the same time, be it with customers, internal contacts (a.k.a. colleagues), or both at the same time. I decided to always display the customer interactions first, in chronological order, and the internal interactions below, with the option of resizing the panel the internal interactions were sitting in.
-
Solutioning
Efficiency in contact centres is one of the most important things, so the agent needs to have every tool that he needs to accomplish his task successfully at his fingertips. I made sure that all controls, actions, and pieces of information the agent needs are prioritised at the appropriate time and are easy to reach.
-
Disposition
Each contact centre has slightly different methods of handling dispositions the moment an agent completes an interaction, so we needed to be flexible here as well. One thing they have in common is the fact that agents do not have a lot of time to complete their dispositions after they finish an interaction. So, I took the decision to allow agents to complete their disposition and take notes while they were handling an interaction, as the disposition is basically a summary of that interaction that is saved for future reference.
-
Next status
When an interaction is concluded, the agent automatically goes back to the "available" status, which enables agents to receive another interaction. This is not always ideal, as agents are either not allowed to decline incoming interactions or they will be placed on the "Break" status if they decline, and this will take time from their actual break time even though they need the extra time to complete the disposition of the previously ended interaction.
To solve this, we offered agents the option to pick their next status while handling active interactions, and the moment they finished all active interactions, they would go to the chosen status.
Improving the adoption rate
As we were slowly replacing the legacy platform, one customer at a time, we were hit with the reality of the contact centre agents. The agents already have a hard life, and introducing major changes to their workflow, like the redesign of the platform they were working with, made their lives even harder. To help with this, I presented a proposal for a training mode for agents to the wider team and decided to start developing it further.
Our hypothesising was that a training mode with gamification elements like ranking and achievements would increase adoption and be beneficial for the agents, making the experience feel fun while also helping them streamline the process of adopting the new platform. To test our hypotheses, we organised multiple user interviews with users that had never used the redesigned platforms and tested a few key scenarios. The tests were highly successful and helped us get buy-in from the senior stakeholders.
Beyond omnichannel
After the beta release, we received a lot of feedback that we already had implemented, and we also picked up most features left on the back burner. As I ended my journey at 8x8, here are some conclusions I reached while working on this project:
Contact centres need to go beyond the current multichannel capabilities: voice, web chat, social media, email, and voicemail. To do something different, there is a need to develop omnichannel. This way, you will be able to create the perfect setting for highly satisfied users and increase adoption.
The direction of a single point of work distribution will enable agents to manage all customer interactions as well as internal interactions. By having a single decision engine, contact centres can create business rules that will serve their business at its best, and the needs of the customers can be matched with the best available resource.
Supporting the cell centre staff and enabling them to adequately empathise with their customers through AI-generated suggestions and allowing the agents to monitor the customer's needs, goals, and behaviour by mapping everything out clearly will likely create a more satisfying product.
Identification of negative emotions while an agent is interacting with a customer is crucial for the success of the interaction, but there is also a need to enable a supervisor to easily intervene if the need arises.
Having an integrated admin solution with real-time insights and analytics will allow supervisors to manage their resources seamlessly and pull the available resources when they are needed.