Caller identification can give you a significant competitive advantage and boost CX in inbound environments. According to research, more than 1 in 3 customers expect agents to be knowledgeable about their previous brand interactions and past purchases. The first step for being knowledgeable and providing a personalised experience basis what agents know is identifying who’s calling in the first place.
In this article, we discuss the four key methods for caller identification in an inbound contact centre – caller ID on the agent UI, DTMF ID, speech recognition and voice biometrics.
- Caller ID on the agent UI
A caller ID system equips agents with caller information by sending the data to their desktop UI. the incoming phone number is matched with the integrated contact database as per your CRM, and the agent can view it either before the call or during the interaction. Before the call, the agent will see only basic customer details and some background information. During the call, the agent can access more detailed insights on the customer and even share them with colleagues for on-the-fly support. After the call, the caller ID system will give the agent the option to enter new data based on the interaction or update existing details.
- Dual Tone Multi-Frequency (DTMF) ID
DTMF, also called touchtone, is one of the foundational technologies for caller identification. Essentially, it receives the caller’s keystrokes as input data and enables identification via PSTN ecosystems. In most cases, DTMF will be able to surface the incoming caller’s phone number but not their name or other details. That’s why DTMF is typically used along with integrations with your active directory or CRM to equip agents with comprehensive caller information.
- Speech recognition-based identification
This is among the most cutting-edge caller ID mechanisms out there, and technically, it refers to voice recognition technology. While speech recognition is useful for transcribing what’s said into structured text (without necessarily knowing the identity of the caller), voice recognition aids in identifying and optionally verifying the caller’s identity. The caller speaks into an automated system – typically an IVR, or an intelligent voice assistant. This audio is used as a sample to compare against your existing database of voice records, identifying who is speaking.
- Caller identification through voice biometrics
Voice biometrics is related to speech recognition, in that it also uses uttered audio as a marker of identification. Every human being has a unique set of vocal acoustics and intonation characteristics, commonly known as a person’s voiceprint. Voice biometrics takes a sample of the caller’s voice and typically a pre-specified keyword or passphrase, to verify and authenticate their identity.
Pros and Cons of Each Type
Every approach to caller identification comes with its own pros and cons. DTMF is near-universal, but gives limited information and is open to fraud risk. Agent UI-based identification is comprehensive but relies on other underlying technologies. The last two methods are the most secure and convenient but are prone to errors as they are yet to become fully mature.
