SATI (Sentiment Analyzer and Tone Interpreter)

SATI is a new, optional Artificial Intelligence tool in SARA. With SATI, you can analyze client and staff text, email, and Messenger conversations in real-time, escalate client conversation priority if they turn sour, or find opportunities to improve client engagement, dialog strategies, and client journeys.

SATI allows you to monitor text-based client service and support conversations so you can respond to your clients appropriately and at scale, find out if clients are satisfied or frustrated, and if staff are polite and sympathetic to client needs.

The language tones evaluated by SATI include frustrated, sad, satisfied, excited, polite, impolite, and sympathetic.

Why use SATI?

As staff use the interactive SATI communications tools, they will see how the client feels about the communication in real-time and be able to respond accordingly. Specific training can be developed on how to deal with frustration, anger and sadness, as these emotions often stand in the way of client success. 

Tonal information feedback generated by SATI can be summarized by staff, office, region, and agency to develop benchmarks for continuous quality improvements in client services.

For managers and trainers, knowing whether a client is frustrated or satisfied with their interaction is critical to assess client satisfaction. It is not simply the tone of an individual statement that’s important, but the progression of tones throughout the conversation that should be tracked.

Optionally, SATI tonal interpretation can be included in SARA case notes, dashboard, performance reports, and analysis applications to improve client engagement performance and outcomes.

Use Cases:

If a client is still frustrated when the conversation ends, that's a problem. However, just knowing how the client felt at the end of the call alone doesn’t tell the whole story. Was the client frustrated at the end of the conversation, because they didn’t like the given resolution? Or, was it because the case manager didn’t show excitement when resolving the problem? Was the case manager impolite or not sympathetic enough to the client’s situation?

Tracking these tone signals can help managers improve how their teams interact with clients. Do the case managers need more training in content or communication style? Are there any patterns in the tones of successful case managers? If so, what can be learned from it to replicate it more broadly?