It’s an interesting time to be in the world. As innovators head to the Consumer Electronics Show in Las Vegas, throwback politics have returned to reign in Washington. Cutting-edge technologies like artificial intelligence and machine learning will be center stage in Vegas, while much of America revealed at the ballot box that they aren’t comfortable with how fast things are moving forward.
Sal Fuentes, founder and CEO of Berkeley, Calif-based Decision Counsel, joined Spotlight in the first week of 2025 to talk about both phenomena – all the ways he thinks AI, machine learning and other technology will continue to revolutionize business and marketing in 2025, how many industries are moving toward subscription-based economies and why not everyone is so eager to embrace change and evolution.
Spotlight: Educate me on what Decision Counsel is, where it sits in the marketing ecosystem and what you consider to be its unique selling point? What differentiates Decision Counsel from other firms like it?
Sal Fuentes, founder and CEO, Decision Counsel: We define ourselves as a go-to market partner. We went from a strategic sales and marketing firm to a go-to market firm because we found you couldn't keep those two things separated anymore. Between the technology evolution and the way consumers and B2B buyers behave, it’s all one motion.
Our biggest differentiator is the fact that we actually do product positioning, product creation and product marketing. That makes us a little bit different, and that goes directly to our Silicon Valley roots.
Spotlight: You have an in-house studio. Has that always been part of Decision Counsel?
Fuentes: No. Five years ago, we were producing just two to three hours of video content and that was almost 100% aligned to events. Last year, I think we produced around 200 hours of content. Much of it is corporate videos, demos, etc. But we also created eight to ten hours of AI-driven visuals. And 100 percent of our voice narrators now – and this is probably very controversial – are completely digital.
Spotlight: Yes, considering that 2023’s writers and actors strikes were at least partly about AI usage, what has the reaction to that been?
Fuentes: I'll give you a specific use case. We produce radio commercials for a national network of broadband providers. Some of them operate in rural Tennessee, and some of them operate in Long Island, and some of them operate in Chicago. So the cadences, voices and dialects we need to use need to be different. Most of our clients can’t afford to hire 50 voice actors to do the same script, but with AI, I can literally create 50 dialects of the same script – load the script up once, and then get literally 50 variable recordings. So we can be super specific and the quality is good enough for classic radio broadcast and streaming as well.
Spotlight: This is probably a bad example, but where I'm familiar with AI voiceover is TikTok, and I find it so off-putting. How are you using AI to create voice and video that seems human?
Fuentes: My team has used AI to create a video of me speaking. To do that, they record a reference video of me with our set of tools. We then create a visual face model, for which I have to grant permission because people need to have control over use of their likenesses, and I record my voice. Now I can give a team in China a two-page script and they can record and create a video without me being anywhere present.
This is what Pete Townsend of The Who was talking about when he said you don't want to see 80-year-old me, but you’ll be able to watch the 1975 avatar of me for as long as you want to. It's quite a breakthrough.
We can give talent and brands complete control of their experience, where they control the AI model, as opposed to somebody manipulating their AI model without any kind of input. It’s this concept of having ethically created and edited AI. I think by the end of the year we’ll be doing it at scale and it could be up to 50% of our produced content.
Spotlight: What tools are you using to do this?
Fuentes: One of the providers of our primary tools is a company called runway.ai, and we also use Adobe’s tools. Adobe specifically has allowed legal indemnification. We have to be very careful. We use publicly available tools to create images based on what we call public reference, but you can't be really brand or persona specific. I can't say something like ‘give me George Clooney in front of a tequila bottle,’ because obviously, I'm violating copyright. Instead, we use what you call RAG (retrieval-augmented generation) models, which is where you control all of the data that’s in there. So it would then be okay for me to take all of the designs ever done by my designers and start from there.
I think most marketing agencies and most brands will have to create isolated models. For example, say you are Amazon and you own the James Bond brand. And then you want to use that brand to create a game, that’s not a big deal. It’s when it starts to derive that it can become a problem. If you’re dealing with your own intellectual property but referencing public data, that’s when things start to break. A lot of our work is just teaching brands about this.
Spotlight: The Consumer Electronics Show is taking place in Las Vegas this week. In light of what you're saying here, what are some trends you are seeing in the ways that artificial intelligence and machine learning will be used this year in marketing?
Fuentes: I think the public is still figuring out the difference between what machine learning artificial intelligence is and generative artificial Intelligence is. I think we're all really comfortable with our Netflix and Spotify algorithms that recommend things to us. That's mostly machine learning. I’m interested in this idea of AI-generated private networks. I can definitely see a model where instead of paying Comcast for my 200 channels, I just pay Comcast $100 a month, and I get my own personalized network of shows. I think it's a little bit of what you saw with the Disney-Hulu-Fubo announcement. That merger moves us toward a place of more consolidated content experiences.
I think the second thing you're going to start to see is device specific creation, where you walk up to a Samsung television and say, ‘I really like these characters. Can you get me some original plots based on these characters?’ I think we’re going to start to see the first dynamic, AI-driven storytelling.
I've also been really fascinated to watch how search behavior is changing. You're starting to see Apple and Google substitute the Siri and Google Assistant interfaces with Apple Intelligence and Gemini. You've started to see voice-based navigation where voice is the primary interface for both input and output. You won’t look at your screen for your answer, you’ll just listen to your answer. And taking that idea one step further, let’s say you listen to a restaurant review and then you just ask the interface to make you a reservation. It’s that idea of taking the next adjacent action – you’ll get a direct result from your personal inquiry and then you’ll be given the choice of whether to take further action on it.
Spotlight: What are the obstacles to moving more toward that model?
Fuentes: The issue is attribution. Let me give you an example. What if you wanted to create the world's best business avatar? So you want to include people like Malcolm Gladwell, Steve Jobs, Warren Buffet, Sheryl Sandberg, the best business minds in the world, and you want to be able to query them with questions like ‘should I make this acquisition?’ or ‘what are my best hiring policies?’ The problem isn't the technology. The technology exists.The problem is, who do I pay? How do I give Malcolm Gladwell his .7 cents, and then how do I give Sheryl Sandberg her .4 cents? The problem is the attribution of the economic model to the different entities who all have an economic stake. That's the thing that's most interesting to me – existing business models are currently barriers to things that could change the world.
Spotlight: How do you think these concepts apply specifically to marketing?
Fuentes: So I gave you one example in which we created a business content product for business leaders to use at scale around the world. How do we use AI to create that model within content? And then how would we bring it to market?
I'll share a real lesson I learned from two engagements, one with Audible and one with Spotify, which is that there are people who want to own content and there are people who just want access to content. The key is understanding which model you can go to market with based on your content ownership and your customer base. Figuring that out is a perfect example of what we do at Decision Counsel. That’s one of our fundamental processes – taking new technology, applying it to existing business models and understanding the core engagement proposition that a consumer or business user wants.
Another thing we do is work with companies to help them understand what we call their ‘true audience’ – how many people can really afford what you sell? And how much should you spend to sell it? We’ll see a company with a $10,000 annual product spend $6,000 to get one customer, but the reason it works is because they keep that product for five to ten years on average. So it’s doing the acquisition math and the lifetime customer value math, which seems like such a simple discipline but it so rarely gets executed at scale. And that model applies to every content company in the world. Every entertainment company is like ‘how much do I need to pay to get a subscriber’ versus ‘how long will I keep that subscriber?’
Spotlight: How do you think machine learning will advance audience analytics this year?
Fuentes: I have always been jealous of Spotify. We worked with Spotify’s sales team on the launch of their advertising project. What we learned is that it's pretty easy to sell wedding destination travel to a woman who's creating her wedding playlist. The ability to target and segment when the user is proactively giving you intimacy points is invaluable. Now take that Spotify data and overlay it with another dataset, like where they do their grocery shopping and they buy, and you have learned quite a lot about that person.
Spotlight: Specifically, how does Decision Counsel work with brands? Do you operate like an agency or where do you live in that ecosystem?
Fuentes: I'll tell you my fundamental secret to what we've been doing for 24 years: It's determining what conversation your brand can own. That's what I think the best companies in the world do. Figure out what conversation you can own and then how you can amplify it and magnify it at scale.
Spotlight: How do you think that marketing and sales teams should work together to create revenue? How accountable do you think marketing teams should be for revenue and do you think that marketing can create revenue?
Fuentes: I believe that the compensation plan of the VP of marketing and the compensation plan of the chief revenue officer should be exactly the same because it all goes to revenue performance for the company, and while the metrics are different – how many leads have you created, or how many impressions are created – in the end, every business in the world comes down to how many customers pay me how much for how long. That should be marketing's problem, and that should be sales’ problem. I have clients where the marketing analytics tech stacks and the sales tech stacks are run by different teams who don’t report up to the same person.
Transparency, alignment, compensation: If CEOs made those three things completely the same between marketing and sales teams, I believe you would have higher performing companies across the board. It's so culturally installed not to do that, it scares the hell out of me.
Spotlight: Why do you think that’s the case?
Fuentes: I think it’s because people don't fundamentally understand the disciplines universally, because they haven't worked across them. Many CEOs come from finance or revenue teams, while very few CEOs come from marketing. You have people who come from different worlds who are like how do I measure passion or brand loyalty? Brand loyalty should manifest in purchasing behavior or in loyalty behavior, and the metrics are pretty advanced now so you can actually watch that stuff.
I would be overly exuberant if I said the actual art of it is easy. It's a very proven discipline, but the companies who do it well have resources and time and have done the hard work of calibrating it. Imagine if sales and marketing metrics had the same discipline as medical equipment and medical devices like blood pressure monitors and glucose monitors, because the level of accuracy required on those devices is life or death. What if you took that kind of disciplined calibration and instrumentation and added it to marketing and sales data? You would get a whole different view of the world. It's hard though. It takes talented people, and the needle keeps moving as well.
Spotlight: What are some best campaigns or best strategic approaches that you've seen in the past three months or so?
Fuentes: I saw ChatGPT come out, and they were so far ahead of the game with everybody, I was like, wow, these guys are going to dominate the landscape. But AI technology moves so fast now that catching up is easier. I think there's a great lesson there, that just because you missed wave one or wave two of a technology trend, you can catch up pretty easily. That's going to be even more true of the tools that we buy outside of those main platforms.
The second thing is that I have been surprised by the fundamental willingness of people to pay for the convenience of access. I think it applies to everything. For example, instead of paying for a lawyer by the hour, I'm just going to have a $2,000 a year subscription that gives me all-I-can-access legal whenever I want. People are already going to subscription-based medicine, where you pay a fee to have access to your doctor. You are starting to see this idea of the subscription economy apply beyond what we’ve always confined it to.
Spotlight: Considering that we already have things like Uber or Airbnb, which are sort of a version of the subscription economy in that those services pool existing resources to provide services, do you think that this subscription economy is evolving into something new?
Fuentes: I think we always talk about innovation within the context of tech. But there's innovation in business models. There's innovation in applying technology to new markets. I think it's that idea of continuing to leverage things and go from there. What's interesting to me is where are the gaps in the market? Where are people just going to keep innovating?
Spotlight: Do you think a subscription-based economy – like the idea of subscribing to a car service rather than owning a car – will help solve some of society’s overall problems? Like maybe less car ownership will be better for the environment, for example.
Fuentes: I think we are actually seeing a backlash against innovation, and I think you're starting to see a backlash against parental concepts – like we used to argue that you should own electric cars because they're better for the environment. Or that you should eat at Whole Foods because natural foods are better. I think there is a backlash against the parental tone of some products. I think part of that is also a backlash against this idea that everything new is better.
I think the most interesting thing for brands right now is whether or not they are going to make the decision to message to the middle. Football is the middle. The Oscars are the middle. The Golden Globes are the middle. This year, I think you’re going to see more mainstream messages and more universal marketing.
