AI Leadership Insights: Impact of GenAI on Retail

Amanda Razani: Hello, I’m Amanda Razani, and I’m excited to be here today with Vipul Vyas. He is the senior vice president of Go-To-Market at Persado. How are you doing today?

Vipul Vyas: Not too bad. Thanks for asking.

Amanda Razani: Glad to have you on the show. Can you speak about Persado, and what services do you provide?

Vipul Vyas: Persado is a decades-old generative AI company. So we’ve been, well-established in the space for a long time, working with mostly Fortune 500 companies, and with a lean towards Fortune 100 companies, where we really help them optimize their marketing activities. Specifically, one of our key taglines has been, “Words matter,” and they do. So we specifically focus on how to drive conversion activity, whether it’s for financial services institutions, retailers, insurance companies. We run the full gamut. Focus on about seven different key verticals, which represent a significant portion of the economy. And again, our focus is on performance, and so driving actual conversion activity.
And this is a little bit different than other generative AI companies, which are more focused on providing something that is intelligible and relevant. We kind of add to that layer cake by really focusing on things that are evocative. We kind of go to that reptilian part of the brain that evokes emotion, creates an itch that needs to be scratched and drives action, which is one of the reasons why we call our flavor of generative AI and motivation AI.

Amanda Razani: Interesting. And it’s funny that you say y’all are a decades-old company as far as generative AI, and yet I really had heard nothing about generative AI until about a year ago. And it’s so interesting how now we hear, “AI, generative AI,” and yet it has been around for a long time, but now it’s just getting so much mass attention by the enterprise. Why do you think that is?

Vipul Vyas: A few reasons. One, obviously OpenAI unveiled ChatGPT last September. And this has been going on for 50 plus years in terms of the development, and I’ve been in a variety of industries that have had a similar arc. I was in the speech recognition space, which had been worked on since the sixties and the seventies, and then really came to the fore in the late nineties, early two thousands, and that had to do with natural language processing as well. In a similar vein, generative AI has been something that people have been working on for a very long time, and gradually then suddenly kind of thing.
And I believe the larger players that weren’t OpenAI were holding back because they weren’t sure how the public would receive something along these lines because it’s a little scary in terms of its capability and what the implications are, and I think we still don’t fully understand. But with OpenAI being a little bit of a, scrappier, less advantageous market position player made the move, they launched ChatGPT, made a splash in the world. Now, everyone’s having to react to that, and everyone could see as from a consumer perspective and an enterprise perspective for that matter, what the power and capability of these types of tools are. And once you’ve seen it, you can’t unsee it. And so, that was a watershed moment for what was happening in the coding cubicles of many companies now was on the forefront of people’s minds. And very quickly, that’s really the sea-change that took place.

Amanda Razani: So, you said that your approach is a little bit different in how you are harnessing this generative AI service. Can you give a use case example or a journey of a company and what were they looking to achieve and how did they accomplish that with this generative AI implementation?

Vipul Vyas: Sure. There’s a variety of players we work with in the financial services space. I mentioned retailers, telcos, et cetera. There are examples of retailers such as Tapestry, the collection of brands such as Kate Spade, Stuart Weitzman, Coach, et cetera, Gap, and those folks wanted to essentially increase sales. It’s as simple as that, ultimately: most businesses are fundamentally lemonade stands, so you’re trying to sell more lemonade. And that objective was best met by doing a few things. One is optimizing engagement, so the one-on-one ongoing electronic communication or digital communication you’re having with the customer, getting that engagement so that you’re developing a relationship that is ongoing in what is essentially a transactional model, especially with a retailer. So, you want to create that recurring relationship that sometimes is exclusively the domain of people who have repeating billing relationships such as your cable company, which many people may not want to have a relationship with, but they do, right? So for a transactional business like a retailer, you want to keep that engagement up, drive business back either the store or onto the website. And so, that engagement was something we focused on.
And then, the second thing we focused on was actually a journey. So once I have someone engaged, someone’s on my website for example, and they’re browsing, they’re looking at things, how do I get them motivated to go from the product description page to really entice them into the product, and then move them to the cart page, and then the checkout and actually make the purchase? That’s a journey in which there’s plenty of pitfalls, and so to summarize this, we use the language to get engagement to draw people in, and we evoke emotion to do that: 60% of our decisions are emotion-based, are driven by emotion, not our intellect, and the same thing when it comes to journeys once we have them and their attention.
And when we do this, when we optimize for words and the right words, it’s not any words, it’s the right words, so we use specific tagged words and phrases that we know evoke the right types of emotions, 96% of the time we’re better than a human copywriter. So, almost always. And when we are better, which again is almost always, we’re about 44% better, so notably better. And that’s not because of a specific deficiency, it’s more just that we have analyzed what works over a massive body of language and employ that. That’s something that a person just can’t cognitively do.

Amanda Razani: That is so amazing.

Vipul Vyas: So, we help people do that. Yeah.

Amanda Razani: That’s amazing. And I know that I definitely make decisions on impulse and feelings, and things like that too, especially when it comes to shopping. And what are some of the roadblocks maybe that these business leaders have when they are trying to implement this? Is there any sort of advice that you can give them? What are some of the key concerns or the issues they face when they start implementing this technology?

Vipul Vyas: Well, when it comes to generative AI more broadly, some enterprises are concerned about the source of the training data and whether there’s any kind of copyright infringement and whatnot. With technologies like Persado, that’s not an issue because we use all of our own training data that we’ve generated through being in the market for so long. So, we’re not scraping the internet, there’s not a future lawsuit sitting around the corner when people use our work, unlike what people feel may be the case for other options. So, there’s that one point of trepidation that we remove. The other thing is just going to be, do I have the capacity to do this? And now, I think that equation has changed: you almost can’t afford not to do it because you have competitors that are moving forward. And so, if you’re too much of a laggard, you’re going to pay the price.
And there’s two dimensions of that. One is these technologies give you a efficiency or productivity boost in the same resources you have today, and I don’t think there’s going to be much resource displacement, even though I gave that previous example around we can perform better than human. It’s really not about that. A bulldozer can perform better than a shovel and a human, but you still want a human in that bulldozer. And so, it’s really augmentation and you’re going to be able to do more with the people you have, and so your cost structure essentially effectively goes down.
Then on top of that, you’re going to be able to, for the example I gave, perform better. So, your lifts or your clicks or your conversions are going to be 44% higher. And so, for the same unit of work or even less as we described for less effort, you’re going to get more juice out of that squeeze. And so, that is going to create a competitive advantage that people aren’t going to be able to ignore. So, I’d flip the question a little bit on the head is, you’ve got to remove the barriers and they’re nominal, to be honest, because you’ll get left behind from a cost structure perspective if you don’t.

Amanda Razani: So, that return on investment is really phenomenal. And it brings me to my next question. We’re full swing in the middle of holiday season right now: For business leaders and companies that want to try to hurry and implement this generative AI technology, do they have time? Can that journey be done that quickly, or are they just going to miss this holiday season and have to start planning for next holiday season?

Vipul Vyas: So, both. On the first question, not necessarily, it’s not too late because fortunately, many of these tools can be implemented in a fairly lightweight fashion. More of what the limiting factor is how they already set their campaigns. Is that already in motion, already planned out? And it’s not going to be a technology limitation. It’s going to be a, “I’ve already locked in my spend. I’ve already locked in my media buys, et cetera,” such that the die is cast, but it’s not going to be a technical limitation by any means. You can get these types of systems up and running within a couple, two, three weeks.
Most of the stuff is pretty prescriptive in terms of how you can move forward. A bit more complex an implementation certainly would be a little bit longer. If you want to drive easier use, there may be some integrations, but none of those things are must-haves to get going. Now, I think having a generative AI strategy going into 2024 is a separate point, but also very important. There’s no too soon, if you will, to start thinking about that because all your competition is doing that right now.

Amanda Razani: Absolutely. And so, for those companies who are probably past that implementation for this holiday season, what advice do you have for them as far as getting started for 2024, and what do you envision this rapidly advancing technology, the impact that it’s going to have in, say, a year from now?

Vipul Vyas: A couple things. One is most people feel they’re always perpetually or perennially understaffed. And so, using technologies such as generative AI are going to take some of that pressure off. So, all the projects that forever fell sort of below the line that just didn’t make the cut because of prioritization, there’ll always be a line, but that line can go lower, meaning you can start to do more of the things that you just couldn’t get to before because not enough hands. That’ll change.
So in terms of specific tactics, I would look at your universal campaigns and use cases and look at the campaigns in terms of size, value, that’s going to be a function of the conversion value and the audience size, essentially what campaigns are worth more than others. And then, look at your generative AI strategy in terms of, where do I want to invest a lot of pre-planning effort because this is a big campaign? I know in January I’ve got back to school, I’ve got Black Friday, et cetera, eight, nine months down the road, and so start planning in January, which everyone does already.
But then the new added wrinkle or dimension is, layering in what tools am I going to use to make that either happen more quickly or inefficiently and more productively in terms of the performance? I really need these campaigns to perform because I can get 5% more here, that’s dramatic. And so, how can I leverage AI to get not only something faster out the door? That’s one simple dimension, but, which campaigns really have to hit the mark for us to hit our numbers? And that’s where I want to make special investments in generative AI to ensure that I can get the sales I’m going to need to get. And that’s especially true in 2024 because the level of economic uncertainty at the macroeconomic level is only growing. It’s not diminishing. You have a lot of factors: there’s instability in the world generally, you have inflation, you have a consumer that is more in debt than ever. All these factors are going to drive, companies have to fight for consumers who are going to feel more under duress than less, more likely than not.

Amanda Razani: Right, so if there’s one key takeaway you can leave our audience with today, what is that?

Vipul Vyas: Don’t fall behind because you are correct: this is a fast-moving business or a space, rather, technology. But that also means that early adopters are going to reap the benefits. And the good thing is typically you have with early adoption, risks like I might do something before it’s ready or right, I don’t think compared to other technologies in the past that’s as much of an issue because as I mentioned, we’re around for 10 years. This is not novel.
It may feel novel to some because as you said, “I didn’t know about this. It came out of the blue,” when in reality, many of these technologies like Persado, for example, have been tried and tested, and the risk is not in the execution or whether this concept works. It’s a matter of just making the decision. That’s probably the biggest angst-ridden, hand-wringing thing is, what do I choose? Because now, there’s so many options in the market.

Amanda Razani: Absolutely. Well, thank you so much for coming on our show today and sharing your insights.

Vipul Vyas: No problem. It was a pleasure being here.