Background

Over the past 40 years, diallers have become so prevalent in the consumer credit industry, that many of us rarely ever bear them a second thought.

This led me to think about the parallels and similarities between diallers and decision engines:

  • Both products have been around for approx. 40 years
  • Both have transformed greatly over the years, from expensive on-premises software integrations to low cost, lighter touch cloud instances
  • Due to the decrease in costs, smaller lenders can now take advantage of the benefits they both deliver
  • Both deliver significant benefits to companies and become integral to a lenders plans
  • Both are highly flexible in order to rapidly adapt to changing volumes and other criteria
  • Both can be used to run A/B tests to explore the most optimal strategies
  • Both can be used across a wide range of use cases within an organisation (collections, up-sells, cross-sells, limit increase campaigns, just to list a few)
  • Both require little/no intervention from the client’s IT department once they are up and running

Predictive Diallers

What is a predictive dialler?

(Note: In Queen’s English, dialler is spelt with 2 letters ‘l’. US English uses only 1 letter ‘l’).

From Wikipedia:

“A predictive dialler dials a list of telephone numbers and connects answered dials to people making calls, often referred to as agents. Predictive diallers use statistical algorithms to minimize the time that agents spend waiting between conversations, while minimizing the occurrence of someone answering when no agent is available.

When dialling numbers one at a time, there are two sources of delay. First, only some fraction of dials is answered; for example, if 1 out of 3 dials are answered, a predictive dialler might dial 3 lines every time an agent becomes available. Second, even dials that are answered take some time before being picked up. If it typically takes 10 seconds for someone to pick up, and conversations typically last 90 seconds, a predictive dialler might start dialling at 80 seconds.[1] A predictive dialler does so by discarding all tones and only passing “Hellos” from the lead to the agent.

Dialling one number at a time, only when an agent is available, typically keeps agents utilized for 40 minutes per hour (33% idle time). Predictive dialling can increase utilization to 57 minutes per hour (5% idle time).

Predictive diallers may be standalone hardware devices, cloud-based, or they may be integrated in software with call centre or contact centre platforms. A cloud-based predictive dialler provides the functionalities of a traditional predictive dialler without requiring installation on the local device and reduces the dependency on the hardware.[citation needed] The dialler can often also perform less aggressive dialling modes such as, power, progressive, or preview dialling.

This very comprehensive description touched upon a number of considerations:

  • Agents
  • Best Time to Call Algorithms
  • Answer Rate
  • Hosting
  • Dialler Types
  • Dialler Modes

Agents

The staff who work on a dialler are called ‘agents’ and they are typically focused on collections or sales. A call centre is split between:

  • Inbound calls
  • Outbound calls

Inbound calls are customers or prospects returning left messages made by agents or they could be customer service type calls. Outbound calls are typically:

  • Collections calls
  • Sales calls

Call centres typically operate on a 12 hours a day, 6 days-a-week basis. This will vary by country and business model (some BPOs operate on a 24-hours basis, 6-7 days a week).

Based on the operating hours, agents work on a shift basis and once again, this will also vary from country to country. In some markets call centres use part time agents, working just 4 hours per day. In some countries agents work 8-9 hours per day.

Best Time To Call Algorithms

These should not be confused with Dialler Algorithms, which control how many calls are dialled, in order to optimise the wait time of the available agents.

Best Time To Call models (BTTC) are used to predict the best time to call a specific customer, based on previous information. These are typically run monthly/weekly/daily and essentially are telling the dialler to “call John Smith between 1700-1800.”

Answer Rate

It is important to define what the Answer Rate is. A model predicting the likelihood of a number being answered is NOT predicting if the mobile phone number will be answered by the customer. The latter outcome is described as the right party connect (RPC) rate.

Hosting

There are 2 hosting types for diallers

  • On-premises
  • Cloud

In many markets on-premises diallers are still the norm, but as with decision engines, the cloud option is becoming rapidly more popular.

Dialler Types

There are 2 main types of diallers

  • In-house developed
  • Dialler vendors

With the proliferation in open source software, in-house developed diallers are common in some markets and industries. Other markets purchase diallers from software vendors. There are now many dialler software vendors, which is similar to the situation with decision engines.

Dialler Modes

There are 4 types of dialling modes in a call centre:

  • Predictive Dialling
  • Power Dialling
  • Progressive Dialling
  • Preview Dialling

Predictive Dialler

predictive dialler automatically places a call by predicting the availability of agents to take the call. The dialler uses algorithms and ratios that calculate metrics such as average call duration. It dials multiple numbers simultaneously to ensure that as soon as an agent ends a call, they are connected to the next one.

Predictive diallers are a huge time saver and as a result, they can save organisations a lot of money. Your dialler team won’t waste a second dialling a number and waiting for a reply, receiving busy signals or reaching a voicemail service. And there will be no lengthy gaps between their calls.

Predictive diallers are often used in telemarketing, debt collection, customer service follow-up and market research. Predictive diallers maximise the time agents spend on the phone with customers. 

Power Dialler

This term is often used to refer to all types of outbound diallers, but it also refers to a specific type of dialler. A power dialler automatically calls the next number on a contact list as soon as an agent ends their call. If the number is busy or goes to voicemail, the dialler moves to the next number on the list. It does not call multiple numbers simultaneously as a predictive dialler does.

A power dialler means that agents do not lose time locating contact details, dialling out or having long gaps between calls. Agents can focus on live calls knowing that their next call will be set up for them.

A power dialler only calls a number when an agent is free to take the call, meaning a customer will never be left talking into ‘dead air’ which is a risk with predictive diallers. It can often be a more cost-effective solution than a predictive dialler.

Progressive Dialler

Progressive diallers operate similarly to predictive diallers, but the number is not dialled until an agent is free to take a call, eliminating the risk of silent calls and frustrated customers. The system manages the dialling process and automatically moves on from numbers that are busy or unanswered.

Sales agents can take a high volume of calls and no time is wasted on the dialling process. There is no risk of customers being left waiting for an agent to respond as the call is only placed when an agent has indicated that they are available.

If your objective is to improve customer experience, increase customer loyalty and gain more revenue from existing customers, a progressive dialler is an excellent choice. This type of dialler is a simple and effective way to upsell products or services to customers.

Preview Dialler

Preview diallers offer a slower pace of outbound calling. Numbers are automatically dialled but only when an agent has marked themselves as available and then only after a specified time. So, an agent may indicate they are free to take a call and the dialler starts a timer for five minutes, after which the next number is dialled.

This system gives agents plenty of time to review customer information before they are connected. This opportunity to prepare ahead of the call can improve the service a customer receives and increase the chances of a positive outcome.

If your business involves the resolution of complex or sensitive cases, such as debt collection or customer complaints, staff may need the extra time a preview dialler allows them to familiarise themselves with a customer before speaking to them. The steadier approach of a preview dialler may be the most suitable for your needs. 

Summary

Diallers and decision engines often work hand in hand in a lender’s account management arena. There are many parallels and similarities between these two indispensable tools.