All Things Interactive Voice Response

Posh AI
6 min readApr 9, 2021

--

By: Larry Ng (Business Development Analysts) & Jack Mcgrath (Customer Success Manager)

Have you ever had to navigate through a series of options through the touchpad of your phone? “Press 1, Press 2, Press 3” and so on? If so, you’ve used an IVR.

IVR, known as Interactive Voice Response, allows companies to funnel callers to the right menu options to ensure they get their answers and processes executed without needing to wait to speak to a live agent. .

How the Process Works:

  1. Callers dial the customer service number provided by the company.
  2. A pre-recorded voice greets callers and explains the different menu options. “Press 1 for checking, Press 2 for savings, Press 3 for fraud etc”
  3. The caller selects an option and another pre-recorded audio clip (usually pulled from the company’s database) pops up giving you commands to help you complete your requests. You can either use your voice or the phone’s touchpad to enter the required information.
  4. If a caller does not find the solutions to their requests, they usually press 0 or automatically get routed to a live representative.

As you can see, the IVR is an adequate solution that has streamlined customer service centers for decades. Many popular vendors such as Voice Cloud, Plumvoice and ConvergeIT have led the charge here. What did companies use beforehand, though, and how did IVR even get its start?

IVR Timeline

  • 1930s

Homer Dudley, a electronic and acoustic engineer at the famous Bell Telephone Laboratory (“Bell Labs,” owned by Nokia), experimented with electromechanical devices to mimic human speech. Working with another engineer, Robert Riez, he created the VOD (Voice Operating Demonstrator). The VOD mimicked human speech through a series of keys that produced a buzzing noise, mimicking nasal and vowel sounds. When played together, phrases could be heard. However, the product did not receive wide adoption as only a few trained operators could fluently navigate the complex keyboard and foot pedals.

  • 1960s

Throughout the late 50’s, most houses had rotary dial phones. However, their complex dialing process and need for human intervention when making long-distance calls, quickly pushed for technological development. In 1963 Bell Systems unveiled a phone system that used DTMF technology (Dial Tone Multi Frequency). The DTMF had 4 rows and 4 columns. Each key sent a unique combination of signal tones that would notify the phone company of the combination of numbers that were being dialed. As more consumers found the convenience and reliability of DTMFs, the rotary phone soon became irrelevant.

  • 1970s

As the mass consumer age roared through the 60s, companies were trying to figure out a way to manage the influx of callers. In 1973, Steven Schmidt deployed the first commercial application of IVR as an order entry inventory control system. Other small inventors deployed similar voice response systems using DSP (digital signal processing) technology but its vocabulary was limited.

  • 1980s

Hard Drive technology had reached a cost effective price point at this time, opening the door for Leon Ferber from Perception Technology to modify digitized voice data. He built a new system that could store digitized speech on disk, play the appropriate spoken message, and process the human’s DTMF response. This piece of technology allowed multi-channels to be controlled by a single application processor which allowed for single touch inputs to playback the recorded speeches in the database.

  • 1990s

In the late 90s, more call centers began to utilize IVRs to provide convenience for customers and reduce servicing costs. IVRs became vital for call centers as they helped with queuing, routing, and data collection points. Further improvement in the technology led to speaker-independent voice recognition of limited vocabulary, allowing people to migrate towards a semi hands-free approach; it also generally sparked more creative solutions beyond call centers. Perhaps one of the most relevant examples was a product called Moviefone, which allowed individuals to find information about a movie just by inputting the zip code and the first few letters of the film!

  • 2000s

In the 2000s, increased CPU power and the use of the VXML standard made IVR common and cheaper to deploy. Some of today’s modern-day call center IVRs have come from software built in this decade. Additional features were added depending on the vendor, which included SMS messaging, call tracking and analytics features. At this time, callers could do things such as pay bills, ask basic FAQs and check balances and account statuses.

  • 2010s

With the complex abilities that the internet and web development opened, IVRs became web compatible and voice-driven. These improvements led to the development of a new model called VoIP (Voice Over Internet Protocol) which was quickly adopted by many companies and call centers. It’s been a popular choice among consumers and small businesses because of its cost-effectiveness compared to traditional landline services. Consumers could make requests over phones but IVRs were still mostly limited to handling basic queries.

  • Present Day

IVRs have come a long way since nearly a century ago! With the rapid shift towards online and mobile banking (propelled even further by the mass adoption of convenient customer service during the pandemic), IVRs have been encouraged to evolve to the next level in order to engage and retain customers. In order to improve, they have significantly harnessed Artificial Intelligence (AI) to unlock the full potential of a human-like conversation without having to escalate to an actual call center representative. Yet not all AI-powered IVRs are equal: while some surge ahead, their more poorly engineered counterparts can still leave callers frustrated with their limited capabilities.

Enter Posh, a cutting-edge conversational AI company built by MIT AI Lab graduates who studied under an early inventor of Siri. In addition to intelligent chatbots, Posh offers a Phone Not (Conversational IVR) that replaces traditional IVR with a natural conversation-driven experience. When someone calls into a Posh IVR, instead of the typical “Press 1 for hours and locations, Press 2 for checking, Press 3 for mortgages,” etc, they’re simply greeted with, “Hi! How can I help you?”

Posh’s conversational approach to IVR is powered by Natural Language Understanding (NLU) that is trained on a deep corpus of banking and financial information. This means that it’s able to handle a huge range of banking procedures, from transferring money to reporting fraud.

Posh’s Conversational IVR authenticates callers through the preferred methods of financial institutions, from traditional PIN-based methods to more cutting-edge solutions, such as voice biometrics. Posh’s state-of-the-art AI then answers questions and executes digital banking workflows over the phone such as balance checks, payments, transfers and travel card freezes to name just a few. When an agent is needed, Posh’s Phone Bot routes to the most available and relevant contact center rep, contextualizing the conversation so far for that rep so that a customer doesn’t have to repeat herself and the rep can quickly and successfully resolve the call. Posh’s Phone Bot can also pull up relevant information about a customer through “Screen pop” and even provide sentiment analysis (i.e. is the caller calm or frustrated). The Conversational IVR integrates with any API-friendly system, meaning it easily works with broader telephone vendors and banking core systems.

Posh’s Conversational IVR has garnered some pretty impressive success metrics:

  • Posh is able to handle up to 97% of telephone banking calls without ever having to involve or escalate to an actual human contact center rep
  • Last quarter, Posh handled over 316K calls for Chartway (a credit union in VA Beach), saving their contact center team over 7100 hours on the phone and their broader organization over $136K.

In addition to voice-driven Phone Bots, Posh has deployed text-based chatbots with over dozens of financial institutions. If you’re interested in learning how Posh can provide your organization the next generation of customer service, let’s chat!

--

--

Posh AI

Conversational AI Company that spun out of MIT in 2018 🚀 We create Intelligent Chatbots and Nex-Gen Phone Technology 📲 www.posh.tech