My Undiminished Hatred of Predictive Dialers ... By The Numbers

A new contact at one of our clients stepped into a decision making role, and reached out to us, trying to get acclimatized to their collection process, started asking questions … they asked about our call volumes, talk times, wrap up times, and times between calls. 

You could tell from the questions he was asking, he wanted to know how we used a predictive dialer.

I told him we don’t use one.  Which prompted a Teams meeting.  Which allowed me to tell him face to face that I hate predictive dialers the passion of a thousand burning suns.

Why do I hate them?

This is not a new topic for me, I wrote an article four years ago about my hate.  Let's revisit that topic and break it down to show from an operations level why predictive dialers don't make sense any more.  Well, aside from the fact they are brute force engines, they are increasingly ineffective in today’s digital communication environment, and they don't provide meaningful consistent, professional communication with consumers ... It's not just about my feelings, I've got math to back up my position.

When I built a predictive dialer with my IT Team at my old employer back in 2009, at that point we had a pretty good 0.4% penetration rate.  That means when we used a team of 24 people to call 130,000 files daily, which resulted in roughly 480 right party contacts, or 20 contacts per day per FTE.  Sounds not completely terrible, right?

Well, at the time, our telephone bill for our eight bundled T1 lines ran about $12,000 a month.  But in the world of VOIP today, let’s assume $0.009 per minute, your average call is going to be about 45 seconds, and we’d get billed in 6 second increments, and let’s also guess that half those dialed numbers were no answer or not in service and thus unbilled, so in 2025 economies of scale we would be running 45,000 billable minutes for a cost of $405 per day, which works out to $8,500 per month.  That’s a little bit better than 2009, but still pricey.

I won’t even get into the issue of dropped calls, consumer aggravation, a pooled environment only collecting to the lowest common denominator of skill in your team, and oh, and the soul-crushing monotony of being jacked into the Matrix like machine that makes your employees surf Indeed every night looking for another job, causing a rotating manpower pool that costs you thousands of dollars in training, lost productivity and expertise, and manpower shortfalls.

Now, when I send a first collection notice through email, on my primary assignment paper, I get about a 20% response rate (either call ins, responses by emails, or website chat requests).  There’s a diminishing rate of return as you send subsequent 2^nd, 3^rd, or 4^th messages, so if I send 6,000 emails in a day with an average 10% response rate, I’ll get about the same 600 right party contacts as a dialer, and it costs me …. Nothing.  Just pushing electrons out from my mail server, and it’s a push of a button in our collection database platform. Instead of taking an hour to call this many files, I can send out these emails in about two minutes.  And every email has a call to action payment link, not just a request to please hold to speak to someone.

Or let’s alternatively talk about SMS?  I can send 3,000 SMS in a day, for an initial 20% response rate (again, by call ins, responding text messages, emails, or website chat requests), so again about 600 right party contacts.  This has about a fringed cost of $0.03 per text, so call it $90 per day or just under $2,000 a month.  Again, an SMS can have a call to action hyperlink payment option, once identity and consent are managed.

By focusing on emails and SMS, and building the right workflow you can throttle output, run campaigns for individual staff members who are named in the messages, creating consumer-collector rapport and connection, give a call to action with that hyperlinked payment URL, *and* give the consumer the option of what communication channel they feel most comfortable using.  You can manage inbound communications easily and get the 20-30 right party contacts per collector.

And for a human touch?  Believe it or not, an individual collection team member making their own calls is more effective than a dialer.  They’ll get about a 10% penetration rate, so 100 outbound calls will connect on average with 10 debtors, resulting in about 7 payments a day.  The telephony cost per user is about $1 per day.  There’s the manpower cost on top of that, but supporting your team with SMS and email, either allowing them to send individual messages or batch contact requests, have consistent communication with the same collector through the life cycle of the collection account, and you can still easily hit the 24 right party contacts per FTE as the dialer example above, plus allow debtors to self-cure accounts.

So, why do I hate predictive dialers?

It’s 1990’s technology, built for a world where people didn’t operate 90% from smart phones with call display.   That 0.4% penetration rate from 2009 is probably down to 0.3% now, and continuing to dwindle.   We now live in a world where the average person checks their cell phone 58 times a day, or roughly once every 15 minutes, for an average total screen time of 2.4 hours a day.  We live in a world where consumers actually get a dopamine hit from responding to the blinky red light on their phone when they get a text message or email (not kidding, watch this video: https://youtu.be/4ReDYMBeHZ4?si=uglq6r_ujZsMbHv4)  

So why are most collection agencies still stuck on using dialers, even four years after I wrote my first article and told everyone that they should shut them off?  It’s still readily available, scalable technology that lets you spin phone numbers.  But is it worth it to use in today’s world?  I’ll let the numbers above speak for themselves.

Got questions about using modern digital communication tools?  Happy to share what I know and how to implement it.  Drop me an email at blair@receivableaccounts.com

Blair DeMarco-Wettlaufer
KINGSTON Data & Credit
Cambridge, Ontario
226-946-1730
blair@receivableaccounts.com\