• By 2020 experts predict that more than 85% of all customer interactions will be handled without the need for a human agent (Gartner)
• 88% expect seamless transition from automated self-service to agent assistance, without the need to repeat information (Aspect Consumer Experience Index Research 2016)
• When companies engage and respond to requests over social, those customers end up spending 20-40% more (Bain & Company)

You’d be forgiven for thinking brands are facing an impossible task to deliver an efficient and effective response to each enquiry. But, this is where automations can help.

Automations enable the triage and organisation of incoming messages, taking on the heavy lifting and enabling the customer engagement team to focus on what’s really important – responding quickly to customers.

How automations can help

• Pre-emptive customer service – Identify those customers who need help before they ask, providing a great brand experience.
• Always-on – Automations don’t need holidays, take breaks or have sick days.
• Reliable – Automated processes always perform to designated parameters.
• Cost saving – Once the initial investment is made the service can be scaled to suit needs.
• Speed – Handle multiple queries and help several customers at once.
• Self-service – Point customers to the information they need to resolve common questions themselves.

Types of automations

Automations can be used to assist with a range of customer service challenges, helping to improve agent productivity and increase customer satisfaction. Here’s our top 5 automations for  customer service:

#1. Reduce noise in your inbox

A noisy inbox can mean missed messages. Use automations to automatically route away comments about marketing posts or campaigns, leaving customer service agents to focus on the messages they need to respond to.

#2. Web chat with a chat bot

A Chatbot is software designed to simulate conversation with human users. Think of it as a virtual assistant that communicates with humans through instant messaging interfaces and text messages.

Chatbots combine AI technologies — such as Natural Language Processing and Dialog Management — which allows customers to ask questions more naturally and have the bot resolve simple issues or carry out simple tasks. Chatbots are designed to quickly and efficiently understand intent and provide simple answers to common questions such as product recommendations, password reset, account access, and troubleshooting.

#3. Remove spam/profanities

Automations can also be used to remove profanities on public pages. This type of automation can be set to activate if a profanity is used, for example on a Facebook post, and then remove it from the public page so it’s no longer visible.

Another second automation could be set to trigger after this, to flag the comment as high priority and escalate it with a “profanity” tag to a complaints team who can then respond privately to the customer before they repost to the public page.

This ensures any profanities are minimised and difficult complaints can be dealt with offline if necessary.

#4. Skills based routing

Routing messages to agents with the right skills to resolve them helps ensure a fast customer service response. It means your customers aren’t kept waiting while agents forward the enquiry to a different team or colleague. This type of automation will automatically route your incoming messages to the right agent, right away, depending on the type of enquiry or keyword used by the customer.

Image showing example triggers.

#5. SLA management

Setting and meeting SLA response times is key to driving customer service efficiency. This can be measured both for individuals or across the whole team. A smart SLA message distribution will enable you to control how conversations are assigned to agents, based on proximity to SLA breach. This helps prevent cherry-picking and ensures agents always deal with the highest priority message first – and can’t just pick the messages they prefer from the queue. 

Once the agent has cleared the messages assigned to them their inbox is automatically populated with new messages.

This can be really useful in improving agent productivity and enabling them to provide a consistent, high-quality service.

How do automations work in Sentiment?

1. To set an automation you select a stream, a social account or a page that the rule will be applied to.
2. Select a role to allocate the comment to, for example, “Buyer complaints team” may be a team set up in the platform.
3. Choose a trigger for the automation.
4. Choose an action or automatic process to be activated once the trigger conditions are met.

Example Triggers

• URL – The source of the message is from a specific URL. For example, a marketing post will contain a specific URL and can be used to route marketing feedback directly to the marketing team rather than via customer services.
• Phrase – Words or Phrases can be used as a trigger. For example: delivery, late delivery, waiting for delivery.
• Channel – The channel of the message can be used to set automated processes.
• Sentiment – The sentiment of the message can be used to set automated processes.
• Language – The language of the message can be used to set automated processes.
• Country – The country of the message can be used to set automated processes.
• Match all – This is used for round-robin distribution, so any comment from the stream in question will be processed.
• Comments on own posts

Example Actions

• Note – Add a note to the comment that will be visible in the customer history.
• Sentiment – Provides the option to override the automated Sentiment setting.
• Mark As – Irrelevant posts can be removed from the queue and marked as complete or no action for reporting purposes.
• Delete from Source – Comments can be deleted from source on Facebook (if for example they contain profanities) but will remain in the Sentiment platform for agents to respond. The post is only removed from the public Facebook page.
• Set Priority – A priority flag can be set on the comment to allow agents to filter or see a comment that is high priority in their queue.
• Tag – A tag can be added to classify the contact reason and assist in reporting and issue identification. The agent can also filter by tag. For example, if a website issue occurs they can filter by the tag “website issue” and answer all the questions as a high priority.
• Last Agent Routing – A comment can be assigned back to the agent that last dealt with that customer on the specific social channel. If they are unavailable the comment will be reassigned back to the agent’s role.
• Fair Comment Distribution – Assign comment to the next available agent.

Image showing the actions against the automation. The automatic processes show that the comment will be tagged as a high priority, sentiment will be overridden to “Negative” and a tag added to show the comment is payment related.

Automations can be a powerful tool for increasing agent productivity and improving customer satisfaction levels. They can act as an extra helping hand, helping you do more with less, and assisting agents in delivering a great customer service experience.