The folks over at AWS have had a very busy start to 2022, releasing 20+ new Amazon Connect features so far! So lets take a look at these features, with a focus on their new ML-powered Amazon Connect forecasting, capacity planning, and scheduling features which went live in eu-west-2 (London) this week!
ML-Powered Workforce Management (Preview)
According to a recent report, 75% of contact centre costs are made up from staff – so it’s important to ensure that your contact centre is running efficiently – forecasting expected demands, accounting for staff holidays/sickness, and planning for contingencies. This is often done through spreadsheets which can cause headaches for managers trying to plan the schedules of hundreds of contact centre staff. Furthermore, it is hard to understand data in this format, and glean insights to predict future work patterns.
Amazon Connect forecasting uses machine learning to analyse historical contact centre data to predict future usage patterns. Short-term forecasts generate an 18-week horizon with 15-30 minute granularity, whilst long-term forecasts can produce a 12 month horizon with daily granularity – both of which are updated on a daily basis.
Forecasts are displayed in a graphical or tabular manner broken down by each queue channel, furthermore these can be exported to CSV format for offline use, or ingestion into CRM’s etc.
Once a forecast has been reviewed, and published a capacity plan ‘Scenario’ can be run against the forecast to understand staffing requirements over the next 12 months.
A weekly or monthly calculation of the forecast contact activity will display
- required FTE,
- available FTE,
- and any overtime (OT) or voluntary time off (VTO) required to cover the gaps.
The next step is to use the capacity planning to automatically produce fine-grain schedules complete with your business rules built in (such as number of breaks, lunch time duration, weekly hours worked etc.).
Shift Profiles define the standard working hours/patterns across your contact centre, whilst staffing groups allow you to group agents with similar working patterns. Staff Rules can be set at the staffing group level, and even overwritten at the agent level to provide flexible schedules. Once automatically generated, and published – agents can view their schedules in the same single pane of glass view that they have become accustomed to with Amazon Connect.
Over the last six months, Amazon Connect has been slowly rolling out new API’s that allow organisations to programatically automate/control processes within Amazon Connect. As Amazon Connect has matured, more of these recent API’s are focused on update, and delete operations which increase the flexibility and integration options for contact centres.
Recent API updates allow you to programatically
- transfer calls to agents/queues,
- retrieve real-time active user data,
- change an agents status,
- search for users by name, agent, tags,
- claim, and configure phone numbers.
For a full list of available APIs, click here.
You can now create custom task templates, making it easy for agents to consistently capture the relevant and required information to create or complete tasks.
It’s now possible to store your own prompts within Amazon S3 and use Connect to play them to customers at any point (such as voice recordings, or different wait music).
You can now have up to six different parties on a call, the agent, caller, and four additional users.
Rich Messaging for Chat
Agents and customers can use bold, italics, bulleted lists, numbered lists, hyperlinks, and attachments
Contact Lens Custom Vocabulary
Improve the accuracy of speech recognition for product names, brand names, and domain-specific terminology, by expanding and tailoring the vocabulary of the speech-to-text engine in Contact Lens.
As a relatively new AWS service, Connect is now becoming extremely competitive as a contact centre offering and these new features remove the need to integrate with third party software to supplement Connect features.
As always, new features are available for everyone, and using the pay as you go pricing model, customers can try features and easily disable them – without tying in to a license or long term commitment.
As an Amazon Connect Service Delivery Partner, Mobilise have experience delivering contact centres for local government, police forces, and the private sector. If you would like to learn more, please get in touch.