← The Agile Brand with Greg Kihlström®: Expert Mode Marketing Technology, AI, & CX
The Agile Brand with Greg Kihlström®: Expert Mode Marketing Technology, AI, & CX
#558: When SMS delivers real results with John Wright, TrueDialog
00:26:49
Loading summary
Transcript36 lines
- [00:01]
A
The agile brand.
- [00:09]
B
Welcome to Season six of the Agile Brand where we discuss marketing, technology and customer experience, trends, insights and ideas with enterprise and technology platform leaders. We focus on the people, processes, data and platforms that make brands successful, scalable, customer focused and sustainable. This is what makes an agile brand. I'm your host Greg Kilstrom, advising Fortune 1000 brands on martech, marketing operations and CX best selling Author and speaker. The Agile Brand Podcast is brought to you by Tech Systems, an industry leader in full stack technology services, Talent services and real world application. For more information go to teksystems.com before we get started, I wanted to let you know that my latest book, Priority is Action seven Principles for Better Strategies, Decisions and Outcomes is now available. In it, I give ideas and insights for leaders and teams that need to make meaningful progress on their priorities. After all, our priorities are what we do, not what we say we'd like to do. You can find Priority as Action on Amazon or learn more on my website greggkilstrom.com now let's get on to the show. With a 98% open rate, the that's right, 98%. SMS marketing can be incredibly effective, but deliverability is key and AI can help here. Today we're going to explore the transformative
- [01:29]
C
role of AI in SMS marketing strategies.
- [01:32]
B
Joining us is John Wright, CEO of TrueDialogue to discuss how AI driven tools
- [01:37]
C
are enhancing SMS deliverability and overall marketing effectiveness.
- [01:41]
B
John, welcome to the show.
- [01:43]
A
Thank you for having us on your pod today.
- [01:45]
C
Yeah, yeah, looking forward to talking about this with you. And you know, before we dive in, why don't you get started by telling
- [01:53]
B
us a little bit about your background
- [01:55]
C
and true dialogue and you know, what led your team to focus on leveraging AI to tackle SMS deliverability challenges.
- [02:03]
A
Well, thanks so much. A little About Me My background is I've been in technology, specifically messaging technologies, for the last 20 years. The first startup I was involved with was an email marketing company called Flow Network out of Canada. We were later acquired by DoubleClick, which is now part of the Google portfolio. TrueDialog is an enterprise CPaaS platform. CPaaS, for those maybe not familiar, is called Communications Platform as a service and we provide software for our customers that gives them the ability to connect with their customers, prospects and employees via our Enterprise UI or one of our native integrations. In the last seven or eight months, or seven or eight years rather, our fastest growing channel for connectivity and messaging has been sms and specifically 10 DLC. And for those unfamiliar with the vernacular of SMS and messaging. There's really three types of codes that customers have been using in the last 10 years. First is a short code which is that five to six digit code that you probably get your two factor authentications from companies like Salesforce and Google. We also have long codes which are your traditional phone number, like what mobile phones are you and I talk with using our Verizon or AT&T or T mobile connection? And then there are 10 digit long codes which is a relatively new application that the carriers have implemented over the last two to three years in an effort into reducing spam. In the early years of sms, numbers were issued without a lot of visibility into who they were being issued to and it caused a problem with spam. And over the last two and a half, three years, the carriers have implemented a system called the 10 DLC and Campaign Registry which customers are now forced to go through, register their brand, how they're going to be using sms. So if there are problems, they know to who go, they know who to go speak with with regards to, you know, any issues with regards to a number, particular number that has been sending messages that their user base has said, hey, this is spam. And specifically that's what true delivery is helping overcome, which is anytime the carriers implement these technologies, there's also filtering involved. And along the way all of our clients who are Fortune 500 and large, sports teams, education and healthcare, all different types of verticals, but they get caught in the filtering that 10 DLC provides in trying to protect their customers.
- [04:38]
C
Great, great. Well, yeah, let's dive in here. So start by talking about the role of AI in enhancing SMS deliverability. Again, as the open rates are so high, it's delivered here is key. Let's talk a little bit first. But you know, what exactly are the stakes involved in SMS deliverability?
- [05:00]
A
Well, I mean the most obvious one, kind of off the starting line is that if a message is not delivered, meaning that an application or a company like ours which is initiating messages on behalf of our customers into the carry networks, if it's not delivered and it's flagged erroneously or falsely by the carriers as spam, the customer can never get it. If you're doing marketing efforts to close a sale or to get another sale or to get a customer to take action, obviously they're not able to because they never get the message. So really the stakes are pretty high in that there could be potential lost revenue if you're communicating with a customer along. Like for instance, our education clients, if they are trying to enroll and they don't get a message saying you're financial aid application is due next Friday and they don't complete it on time. It's not only a poor user experience, but that university is losing tuition revenue and you know, just generally lost opportunities. Our customers, you know, that can cause, you know, customer attrition, added strain, you know, inside of our business, you know, and for our customers who are using it, you know, this lost productivity, but by their customer service or call center agents who are, are using this and causing higher support costs because all of a sudden they're like, I sent you this message and they're like, I didn't get it. Trying to figure out what happened to it. And kind of along with that, all of the same stakes that are, you know, lost revenue, poor user experience and overall missed opportunities are, are the same for our customers as well.
- [06:37]
C
Yeah, yeah, well, and, and for those, those marketers out there that are, you know, whether it's increasingly relying on SMS or just, you know, rely on it because of its high open rates and things like that, you know, I'd imagine this is, this, this affects the potential reach of, of their marketing campaigns as well. So, you know, how, how should marketers be looking at, you know, deliverability and ensuring deliverability?
- [07:06]
A
Well, that's a great question. And I, I really kind of look at this industry and it's very similar to what happened on the email world that I was involved with for a long time. And if you remember, in the early days of email, several years after that, there were several companies that came along, return path, email on acid, pivotal veracity that helped companies understand the overall deliverability. As you know, email spam engines started filtering legitimate messages from marketers. And the same thing's happening here. And I think that the statistic that you use, 98%, which is what Pew Research produced on an open rate, if it gets delivered to the handset, they're seeing 98% opens. If you don't get it to the handset because it's being blocked, you're missing out on a significant amount of opportunity for your consumers to interact with you.
- [08:00]
C
Yeah. And so with the introduction of true delivery, which you briefly mentioned, your team has reported up to an 85% reduction in messages being falsely flagged as spam. So can you elaborate? There's some AI behind this. How does AI contribute to achieving such significant improvements in deliverability?
- [08:24]
A
There again, this is something that came out of really our own needs internally, which was as 10 DLC continued to grow. And really it's not even grow it was something that was being forced by the carriers and that if you want to be able to play in their sandbox, you need to go through and register much like, you know, in the banking regulations and know your customer. The carriers want a very similar thing in place and they do not want what happened in email to happen inside of the carrier networks. And we, as they started implementing 10 DLC and implementing filtering across the 10 digit long codes, you know, we started seeing an increase inside of our customer service department with customer inquiries saying, hey, my customer said they never got this message. And we started doing our own internal research and we deliver hundreds and hundreds of millions of messages on behalf of our clients. And as you can imagine, with the vast volume of data that we're dealing with, it's finding a needle in a haystack. What's causing it, what's causing it to happen. And we started seeing this at the, you know, right as ChatGPT and other AI tools started hitting the market. And one of the challenges that we faced specifically was is that, you know, we deal with a lot of higher education, our data and our customers data. We can't let that outside of our ecosystem. We've got a responsibility to protect that data. So we started going down the path of using AI to look at all messages being sent that our clients were sending and then looking at the term codes that the carriers provide us with, saying, it was delivered, it was not delivered, this message was not delivered because the number no longer exists or the cell phone is out of minutes, it's a prepaid cell phone, or there's a code that comes back and saying block spam likely. And we started doing evaluations of different AI tools. We ended up settling up on Microsoft's AI tool inside of Azure, which is our entire tech stack is inside of Microsoft Azure. And we started testing and along the way we started using AI and machine learning to basically run through and build algorithms that determined here's what was delivered, here's what was not delivered, and here's what was not delivered because of spam and assigning scores and building algorithms and models built off of that. From that. The way AI works is it goes through and it looks at every individual word, assigns a score, and it starts looking at similarities across those. And because of our large volumes, that gives us a really good read as far as what is and what is not likely to get screened at the carriers. We also started breaking it down by carrier because every carrier has different filtering rules that are in place and that allowed us to deliver a product that our customers can use prior to delivery. So unlike email where you would get a, a list of domains and email accounts that you would send to to see what your deliverability into Gmail or into Hotmail or into these different mailboxes and tell you what is and what is not going into these mailboxes, ours looks at things prior to delivery. Also with our ongoing, you know, generative AI learning because they're changing the rules every day, what is and what is not likely to get filtered.
- [11:47]
C
Right.
- [11:47]
A
Our system is easily able to update and give our customers a preview as far as the likelihood of it getting block or getting delivered by carrier.
- [11:56]
C
Yeah, and I think I know the answer to this or at least part of the answer to this. But just to confirm, it really is the volume of messages that makes this so powerful. Right. So you know, in other words, a company that sends, you know, a few thousand or even, you know, in the low, low millions of messages could probably do some analysis. But you're the breadth of and depth that you have and the ability to do this really helps the AI make better predictions and all that stuff. Would you say that's true?
- [12:31]
A
Without question. You're right on point. And I think that the second thing is that the other kind of side benefit in that having the customers in healthcare and government and higher education and our dependencies in that we don't let our information outside of our own network and share it with a public AI like ChatGPT is that we get smarter and smarter and smarter. And it also is very proprietary to us. And allowing us, the law of large numbers allows us to not only identify trends very quickly, but also as they're adjusting things inside of the carrier networks for us to identify those as they happen.
- [13:17]
C
So I want to talk a little bit about integration of AI in marketing strategies here. And you know, a recent Forrester report indicated that businesses attribute almost 13% of their online revenue to SMS marketing. So you know, given, given this, you know, that's a, that's a pretty large share for some pretty large companies. How should companies be looking at integrating AI tools like yours and others to protect and increase this revenue stream?
- [13:50]
A
Well, I think it's paramount, right. And I think that we've seen over the last six months significant decline as Google and Microsoft and others have really turned the filters up on email. And we've seen, I think that number is actually growing because of the filtering that's happening at the carriers or I'm sorry, at the ISPs with regards to email deliverability. We've seen a sharp decline in the overall effectiveness of email, which has been constant since I got in the game back in the late 90s when email was kind of in its infancy and seeing how it grew, you know, today, as email becomes a more difficult mechanism to use profitably, text is really increasing. And as the carriers continue to streamline, you know, their detection tools, the landscape, you know, is just changing daily. Multiply that, you know, by. Let's just say there's five main carriers in the U.S. t Mobile, Verizon, ATT, U.S. cellular and Dish Network. You know, they each manage their own networks, apply their own rules. You know, it's not easy and, you know, getting around and building what in the email world were static models in order trying to determine deliverability. Ours are in real time and generative, so they're updating real time. If you're a large retailer and you're sending out a message to a million people who purchased from you in the last 90 days and you're doing, let's just say a Fourth of July sale, since we just passed Fourth of July recently, a decrease in that of, you know, 2, 3, 4% of your overall response rate is a significant amount of lost revenue. And it's using tools like this to ensure deliverability of your message. Not only that, you know, your, your, your messages are getting sent, but getting delivered to the inbox, I think is paramount to the overall success of any company who's engaging with their consumer, their customers or students or whatever via sms.
- [16:02]
C
Yeah, yeah. And, and so, you know, not only is it SMS marketing, you know, again, about 13% of, of online revenue currently, you know, the, the market itself is estimated to reach almost 90 billion, 89.2 billion by 2030. You know, as, as you kind of mentioned with email, you know, the, lots of, it's a popular medium and therefore there's a lot of people trying to, to exploit it. You know, SMS as it continues to grow, is AI the key to evolving and meeting not only the increasing demands of a growing medium and channel, but also in just the complexities of if there's more SMS messages going out, then there's going to probably be more potential fraud and more things to mitigate against. Is AI the key here?
- [16:59]
A
Yeah, I think that AI is definitely one of the keys to success here. I think that the carriers are doing a really good job as far as narrowing and mitigating risk as far as people who are trying to gain access to their infrastructure and send messages, whether it's smishing or phishing campaigns. That they're trying to run. And 10 DLC has knocked out a big chunk of that. I think they're going to continue on their end. They're applying AI in their own infrastructure networks and we work very hard. I think one of the things that I'm very proud of as a business that we do is we turn down, I don't know of our overall people who are trying to register and come in and try to use our system. We maybe accept 20 to 30% of our customers or 20% of the traffic that's coming in because we're using tools to not only help deliverability, but that's a powerful tool. We don't want that getting in the wrong hands of bad actors. But you know, we see that it's not only a 2Pmessaging, but P2Pmessaging is getting blocked as well. And using tools like, you know, how we've deployed it for our customers. You know, I, I think that with the volume of data that the carriers are dealing with that companies like TrueDialog are dealing with, I don't know that there's any other way other than to use AI in order to deploy it to handle the voluminous amount of data as well as the constant changing and tweaking that the carriers are doing. And that's each individual carrier. So again, multiply by five because they all have different settings and being able to identify those quickly is paramount to the overall success of our customers.
- [18:47]
C
Yeah, and real quick, terminology check. You use the term A2P and P2P. Do you mind just explaining for those less familiar?
- [18:56]
A
Not at all. P2Pmessaging is, if you and I were to communicate, it's called person to person. And that's like if you and I were having a one to one communication and that is not governed by the carriers, they, they don't require a 10 DLC registration for that. A 2P or application to person is business to person communication. So a business or a governmental, a government entity or a higher education institution that wants to communicate with their end users, that by the carrier's definition is called application to person. And that's what requires the 10 DLC registration. And requires. And that's where the filtering starts, if you will.
- [19:35]
C
Yeah, yeah, got it. No, that's helpful. Well, I want to, as we kind of wrap up here, I want to talk about a few, you know, future leaning things for those listening out there. You know, have a lot of marketing leaders and aspiring leaders listening out here. You know what, what strategic considerations should companies keep in mind when Thinking about how to, you know, adopt AI enhanced SMS tools and, you know, knowing that this market is, it's already big, but it's, it's continuing to grow. You know, what, what are some of the strategic considerations?
- [20:10]
A
Well, I think, you know, I've mentioned a couple of times during your podcast, but I think, you know, the first one that entities should consider is where do you want your data to end up? And I know that's a bit of a stumbling block for a lot of people, but what I mean by that is that by incorporating things like a ChatGPT great application, I think it's a terrific company. But you have to remember that as you submit data into an application like ChatGPT, that data can be used by anybody who's inside of the ChatGPT universe because it is using it to further enhance and learn and optimize its generative AI through your data. Our clients, because of their specific needs and their desire to keep their data out of public AI or open AI applications like ChatGPT or OpenAI, making sure that whatever application that they're using meets your data requirements. Everything that we do is closed. It's inside of our own knowledge base or AI knowledge base and is not shared with any public concerns. And it's also anonymized. So it's not at a client level. It's looking at the overall level of messaging and protecting. The other thing to consider is that we've been very careful in that our customers are not just concerned with their own outbound messaging activities going into public OpenAI applications, but even more than that, they're concerned about their customers information, meaning their responses going into an OpenAI type application and the impact that can have on their own privacy and the standards that they've set forth inside of their privacy policy. We screen out and don't use any incoming data. The only thing we're looking at is the outbound delivery data that is being sent from our clients to our, to their customers and the elimination of all inbound data, which would be catastrophic if customers were sharing PII over text and those got included into OpenAI or ChatGPT type application sets.
- [22:24]
C
Yeah, yeah, absolutely. Well, one last question for you here, looking forward. What are you keeping an eye on? You know, what emerging AI capabilities or other innovations do you think are going to further transform SMS marketing and customer engagement on this medium?
- [22:43]
A
Well, it's again a great question and probably the most obvious that's being deployed today are chatbots. Right? They're already out there, they're already growing and improving. And I think the same things that you've got to take into consideration as we just talked about is what guarantees are you making with your customers as far as what's getting shared. I think a lot of people have adapted ChatGPT and again great application but you know, just realizing that you're sharing data and that's outside of any agreement. So again, you know, I think that what we are going to see is, is that you know, a lot more private byob if you will bring your own bot where you're building your own chat index and it's only being built off of your data I, I think is probably a one of the things that it, you know, we're we're doing currently and we think that the market's going to be trending that direction. We also think that generative AI can be applied in many different aspects of our industry. Things like response models going through and things that are being used in other marketing channel like changing content at an individual level to optimize response rates because you're able to look at individualized generative learning behavior based off of all of your prior response data that is coming in out of sms. And again being able to update those in real time provides a real advantage to customers who are I think advanced and forward leaning enough to where they can take advantage of that and they have a good understanding of data and how to optimize data. Therefore optimizing the response rates by tailoring content but also eliminating messages to those customers who are not likely to respond. Which is a, you know, if you can cut out the cost component of sending out, let's say you've got a million customers inside of your database and you can identify 300,000 of them that only buy that one time a year and have never purchased anything else. In identifying those customers that maybe were in that 300,000 that don't reply that have suddenly started purchasing every month. Those are big benefits that can be taken advantage of by using AI type applications.
- [24:58]
C
Yeah, yeah, I love that train of thought there. Definitely it's something I'm seeing with other channels as well. Just the potential of that actual one to one personalization that many have been talking about for. I feel like we've been talking about it for decades but it seems like we're on the cusp of it with some of this gen AI stuff. So I love that we'll have to talk more about that, maybe have you back on the show and talk more about that.
- [25:25]
A
Well, thank you so much for having us on the Agile Brand Pod Greg, I really did enjoy speaking with you and as we've got more cool AI marketing applications cooking in our kitchen, would love to come talk to you about them.
- [25:38]
C
Absolutely. Well again, I'd like to thank John Wright, CEO of TrueDialogue, for joining us today and sharing insights into how AI is revolutionizing SMS marketing through enhanced deliverability and engagement. You can learn more about John and True Dialogue by following the links in the show notes.
- [25:56]
B
Thanks again for listening to the Agile Brand brought to you by Tech Systems. If you enjoyed the show, please take a minute to subscribe and leave us a rating so that others can find the show more easily. You can access more episodes of the show at www.greggkillstrom.com that's G-R-E G K-I H L S T R O M.com While you're there, check out my series of best selling Agile brand guides covering a wide variety of marketing technology topics. Or you can search for Greg Kilstrom on Amazon. The Agile Brand is produced by Missing Link, a Latina owned, strategy driven, creatively fueled production co. Op. From ideation to creation, they craft human connections through intelligent, engaging and informative content. Until next time, stay Agile.
- [26:45]
A
The agile brand.