The contact center industry has experienced three distinct generations of AI & automation.
First came hand-built speech recognition engines. Second, around 2012, natural language processing (NLP) emerged with tools like Google Dialogflow and Amazon Lex.
Then, the third generation surfaced with the rise of generative AI (GenAI) two years ago.
Critically, this has accelerated the return to value for AI investments. For example, a speech analytics project that may have taken months to fine-tune two years ago now takes minutes.
Yet, alongside this greater ease in configuration, use, and deployment, the third generation has enabled new AI use cases to reimagine cradle-to-grave contact center experiences - from pre-call insights to post-call automation.
As Thomas John, VP of EMEA at Five9, explains:
“AI Insights equips agents with customer data and suggested responses before the call, and AI-generated summaries speed up post-call wrap-ups – it’s the entire lifecycle of an interaction.”
Given this newfound opportunity, more contact centers are asking their vendors for help starting with AI, managing AI, and delivering streamlined conversations.
In line with this, they’re demanding responsible AI policies, care about how their data is used, and seek assurance that AI models aren’t biased.
Thankfully, Five9 has stepped up to the plate for its customers by launching Genius AI. Here’s how it works.
Five9 Genius AI
Five9 Genius AI is a four-step process for implementing AI, centering on the Five9 data lake. Customers can action that process – as outlined below - by leveraging elements of the Genius Suite, which centralizes the vendor’s AI solutions.
Step 1 – Listen
At this stage, businesses capture information generated on the Five9 CX platform and store it in the data lake. There, contact centers can better understand and enrich their data.
Step 2 – Analyze
Here, contact centers can assess where their pain points lie, using tools like large language models (LLMs) to reduce each interaction down to the core contact driver.
Interestingly, Five9 calculates an impact score to show how each of these drivers impacts the contact center’s workload.
With other Genius Suite tools, contact centers can also consider: is this a self-service issue? Is it an assisted service issue? Could we have avoided the conversation entirely? And more.
Step 3 – Tailor
In this step, business ground AI models – including third-party LLMs from OpenAI, Google, Meta, etc. - to ensure optimal performance within the contact center environments and guard against risks, such as AI hallucinations.
That process involves overlaying the model with data and insight from the CRM, customer conversations, and various knowledge sources.
Step 4 – Apply
Now, contact centers can select and action AI solutions, harnessing their tailored AI model and delivering new-look experiences.
Whether through Intelligent Virtual Agents (IVAs), agent assist, workflow automation, or other forms of AI, targeted implementations guided by that analysis work from step two will help drive success.
Finally, measuring that success is critical, isolating improvement opportunities, and revisiting this cyclical process – which the contact center can do as frequently as possible.