Episode 6
Building an AI Value Prop Without the AI Washing
Doug Straton, CMO of Bazaarvoice, joins Erica Seidel on The Get podcast.
You'll hear about:
- What makes an AI value prop actually defensible
- Moving from a 'selfish' AI approach, where individuals automate their own jobs, to an approach that puts the business and its processes first
- How relentless simplicity in org design wins, especially in an environment where CMOs need to defend their org investments to the Board ("Can I explain this to someone in 30 seconds?")
- The charter of the new role of 'Head of Marketing Strategy and AI Transformation' -- to make sure the processes serve the business, not the people on the team – THEN automate the process
- What it's like to go from the B2C world to the B2B SaaS world – and when to wear your 'insider' hat versus your 'outsider' hat
- A context-heavy interview approach that gets candidates to dig in and explain 'This is where I can help you'
Notable Quotes:
- “There are SaaS companies that are fully powered by engineers that have been able to design very clever solutions but have not necessarily really understood their customer. It takes years for them to figure out product market fit. They've got something that sounds like an interesting idea, but they have to test it over a very long period of time. That modus operandi is extinct."
- "Probably the biggest move I made is I put in a Director of Marketing Strategy and AI Transformation. The job spec is very much like a marketing engineer, and that person makes sure that for all the work streams that we have, there are processes built, and the processes are strong and the processes serve the business, not the people on the team. And once we know that the process is strong and it serves the business and not the team, THEN we automate it. And only then, because if you automate chaos, you'll just get automated chaos. You'll get more content, for example, if that's what you're trying to create, but it won't be any good."
00:00 Show Intro and Guest
01:20 Guitars and Woodstock
02:27 From B2C to SaaS
04:51 Staying Client Centric
07:00 AI Value Prop Strategy
12:40 Proof and Metrics
15:36 Avoiding AI Washing
21:52 Rebuilding the Org
25:08 AI Transformation Role
30:16 Hiring and Interviewing
32:31 Early Mistakes Learned
35:31 Talking to PE Boards
38:49 Wrap Up and Outro
The Get is here to drive smart decisions around recruiting and leadership in B2B SaaS marketing. We explore the trends, tribulations, and triumphs of today’s top marketing leaders in B2B SaaS.
This season’s theme is focused on how AI is reshaping the CMO role: how you lead, how you hire, and how you get hired."
The Get’s host is Erica Seidel, who runs The Connective Good, an executive search practice with a hyper-focus on recruiting CMOs and VPs of Marketing, especially in B2B SaaS.
If you are looking to hire a CMO or VP of Marketing of the ‘make money’ variety, rather than the ‘make it pretty’ variety, contact Erica at erica@theconnectivegood.com. You can also follow Erica on LinkedIn or sign up for her newsletter at TheConnectiveGood.com.
The Get is produced by the team at Simpler Media Productions.
This podcast uses the following third-party services for analysis:
OP3 - https://op3.dev/privacy
Transcript
Welcome to The Get, the podcast that's all about recruiting and
Speaker:leadership in B2B SaaS marketing.
Speaker:I'm your host, Erica Seidel.
Speaker:I recruit CMOs and VPs of Marketing for B2B SaaS companies, from
Speaker:scale-ups to larger companies.
Speaker:My tagline is that I place the make-money marketing leaders,
Speaker:not the make-it-pretty ones.
Speaker:This season on The Get, we're looking at how AI is reframing and reinventing the
Speaker:CMO role, especially how it relates to hiring, getting hired, and leadership.
Speaker:Today, we do a deep dive with a CMO who comes from the B2C
Speaker:world and pivoted into B2B SaaS.
Speaker:We'll talk about how to build an AI value prop without
Speaker:being perceived as AI washing.
Speaker:My guest today is Doug Straton.
Speaker:Doug is CMO at Bazaarvoice, the largest provider of consumer-generated content
Speaker:about products, with a network of over 13,000 brands and retailers.
Speaker:If you've left a rating or a review on a product or used ratings before
Speaker:buying something, it's likely you've interacted with Bazaarvoice in some way.
Speaker:Doug was previously Chief Digital Officer at Hershey and VP at
Speaker:Unilever, covering digital, e-commerce, omnichannel, and data.
Speaker:Doug, welcome to the show.
Speaker:Thank you, Erica.
Speaker:It's nice to be here.
Speaker:I'm glad to have you.
Speaker:Uh, let's start with a fun fact about you that is not on LinkedIn.
Speaker:Well, I play guitar.
Speaker:I play and collect guitars.
Speaker:As such, you obviously have to be a big music fan, and I live in
Speaker:Woodstock, New York, the Woodstock.
Speaker:Uh-huh.
Speaker:Um, so most of my neighbors are actors, musicians, directors,
Speaker:documentarians, producers.
Speaker:There's always somebody famous that you're gonna run into in town.
Speaker:Um, also bears.
Speaker:We have
Speaker:a lot of bears.
Speaker:That's great.
Speaker:How many guitars do you have?
Speaker:Ooh.
Speaker:At last count, 11.
Speaker:Okay.
Speaker:Wow.
Speaker:Do you have a room just for the guitars?
Speaker:Um, well, I just built an addition, and yes, the guitars will gradually migrate,
Speaker:though I've been kind of curating down my collection and, uh, 'cause
Speaker:there's a lot of ancillary equipment.
Speaker:It's mostly electric.
Speaker:Yeah.
Speaker:Um, so I'm going digital, so most of my stuff is … Most of the amps and
Speaker:whatnot are being sold off, and I'm going all digital, which is great.
Speaker:It's very freeing, to be honest.
Speaker:And it's very on brand, you being a, you know, digital native kind of guy, so, um-
Speaker:I'm not afraid of it.
Speaker:Yeah.
Speaker:That's cool.
Speaker:All right.
Speaker:I'm picturing you in Woodstock.
Speaker:That sounds very bucolic.
Speaker:Um, you are in some ways kind of an unlikely B2B SaaS CMO 'cause
Speaker:you were CDO at Hershey, and again, you were at Unilever before.
Speaker:So there's this arc from like outsider to insider, and I'm wondering, like, in
Speaker:your mind, like how does this make you different from the dyed-in-the-wool SaaS
Speaker:CMO, you know, profile, and what have you had to learn and what was easy for you
Speaker:kind of g- you know, fulfilling your role?
Speaker:Yeah.
Speaker:Well, the, I, I think the key thing for me, and, and it was a transition that
Speaker:I had to kind of like lean into over a period of time, was just, you know,
Speaker:rethinking the vocab there and the lingo.
Speaker:So marketing is marketing, but the, the lingo of a B2B SaaS business is very
Speaker:different than the lingo of a B2C company like Unilever with big brands like
Speaker:Dove or, or Hershey with like Reese's.
Speaker:But ultimately, once you get through the vernacular and the
Speaker:vocabulary, you can figure things out.
Speaker:And I think maybe what I've brought to this role that a, you know, a typical
Speaker:B2B SaaS CMO wouldn't, is that I just, I had the experience as a client.
Speaker:And even as a client, I've had like a very unique career in that I started
Speaker:very early in the beauty industry, literally selling and then marketing
Speaker:to independent salons through wholesale distributors, and then ultimately
Speaker:went into, into the mass market.
Speaker:So I've, I've called on from independent salons all the way
Speaker:to Costco, Walmart, Amazon.
Speaker:Those have all been my clients, both domestically and internationally.
Speaker:So I've, I've seen every route to market you can think of and, and dealt with
Speaker:very big and very small, and helped actually, you know, build a brand and
Speaker:then ultimately sell a brand to Unilever.
Speaker:So I think just a lot of experience across like small businesses, big
Speaker:businesses, ultra big businesses.
Speaker:And I think the, the, the other thing that's maybe a little bit different
Speaker:is, you know, as I built that business and ultimately helped, you know,
Speaker:r-run it and then sell it to Unilever, I became very adaptable, you know,
Speaker:'cause I went from big company to small company back to big company.
Speaker:That's helped me just have a unique perspective stepping into this
Speaker:role where I can very easily see copy or a description of something
Speaker:that's very technical and say, there's no way the clients are
Speaker:actually gonna understand that.
Speaker:You could simplify it, and here's why.
Speaker:And oh, by the way, your brand client is gonna look at this differently from,
Speaker:say, your e-commerce persona within that same client, and you have to market
Speaker:to both and be very clever about it.
Speaker:Yeah.
Speaker:Was there a tipping point where you lost a bit of the external client lens and
Speaker:drank too much Kool-Aid, or drank enough Kool-Aid so that, you know, you were, uh,
Speaker:I guess, y- you know, equally thinking of Bazaarvoice and the customer side?
Speaker:Boy, I hope not.
Speaker:Yeah.
Speaker:Um, I think-- I can tell you when I, when I took the CMO role, one of the things
Speaker:I did was I told the entire marketing team, you need to get out in the field,
Speaker:and I want you in s- I want a store.
Speaker:We're gonna walk a store like one of our clients would walk a store, and I want
Speaker:you to walk the digital stores as well.
Speaker:And I literally was doing guided tours.
Speaker:It was so helpful for me.
Speaker:Number one, it showed me that I hadn't forgotten everything, but it also
Speaker:showed me that I was a little bit rusty.
Speaker:I had to kind of pull from memories that I hadn't, or, or, you know,
Speaker:knowledge that I hadn't used in a while.
Speaker:But literally walking, walking into a store and saying, okay, what do you see?
Speaker:How is this merchandised?
Speaker:Why is it merchandised this way?
Speaker:And then, like, literally walk into every section in the category aisle.
Speaker:It's, you know, the general merchandise versus grocery versus drug.
Speaker:And then point out, you know, from top of the shelf to bottom of the shelf,
Speaker:you know, like, how, uh, as you're looking at this, what kind of science
Speaker:and analysis do you think took place to, to make the shelf look the way it looks?
Speaker:And why are these certain products in a checkout lane, but they're
Speaker:not in other parts of the store?
Speaker:And, like, really forcing my entire marketing team to think it through in
Speaker:a way where they just, they can't do that behind their desk and on a screen,
Speaker:thinking that they're just marketing technology, and they're just gonna
Speaker:focus in on this one element of digital marketing and suddenly understand
Speaker:where our clients are coming through.
Speaker:I'm like, get out in the stores-
Speaker:Yeah
Speaker:… and talk to people.
Speaker:Yeah, and s-surprisingly, you know, that wasn't a muscle that
Speaker:we were, we were really using.
Speaker:So it was helpful for me to, to make sure that I wasn't losing it,
Speaker:uh, losing my, my mojo and, uh, making sure that my, my team had it.
Speaker:And I think that the other thing too is I, I make sure that I, I talk to,
Speaker:you know, as many clients as I can.
Speaker:So at the beginning of this year when we were doing the big AI push and we were
Speaker:talking about core messaging and making a lot of our marketing changes, I probably
Speaker:talked to 40 clients in about 60 days.
Speaker:Right.
Speaker:Right.
Speaker:So let's talk about, um, building a new AI value prop for, like,
Speaker:a product marketing standpoint.
Speaker:So Bazaarvoice is responsible for a lot of content that AI surfaces, obviously.
Speaker:And so, uh, you know, my question is kind of like how do you build a new AI
Speaker:value prop, and can you tell us about that process where you're kind of trying
Speaker:to move at the speed of the market, but also not being perceived as AI washing?
Speaker:So I think there's this, like, threading of a needle that a lot
Speaker:of companies are thinking about.
Speaker:Can you, can you tell us your experience with that?
Speaker:Yeah.
Speaker:Well, the, I think the first thing was, um, to our previous conversation,
Speaker:you know, I, I used a really practical approach to, to think about, you know,
Speaker:what we were gonna need from a core messaging perspective, how that was gonna
Speaker:play in terms of PR or with analysts.
Speaker:Yeah, I had a vision for what sales assets would look like, and then
Speaker:ultimately, you know, campaigns, and it's all like really, really pragmatic.
Speaker:We were kind of known for having a, a very smart but complicated message that
Speaker:was very difficult for our salespeople to convey and for our clients to understand.
Speaker:So a lot of it was like whatever we do in terms of AI or not AI, you know, if,
Speaker:depending on what the hypothesis was gonna be, we needed to make sure that the
Speaker:company, as I was … That was my main focus is, like, just simplify everything.
Speaker:Focus on just a few things, make sure the execution is really great,
Speaker:don't get fancy, and make it just super simple for our salespeople
Speaker:to communicate to our clients.
Speaker:So that was, like, the first thing that was on my mind.
Speaker:For Bazaarvoice and specifically, you know, how did we arrive at a value
Speaker:prop, it's a really good question.
Speaker:To be frank, we were a little bit lucky.
Speaker:We did have a hypothesis, and the, the hypothesis that, that we had was
Speaker:that AI platforms need a lot of scaled sources of structured data, and data
Speaker:that's really, you know, that reflects what real humans are interested in and
Speaker:what they're thinking, not necessarily what a machine has generated.
Speaker:Right.
Speaker:Which is in, you know, the experience of others, you know, in their
Speaker:interactions with products and services.
Speaker:It's verified as authentically human.
Speaker:Once again, had endorsement signals of some sort or not endorsement signals.
Speaker:It doesn't matter.
Speaker:As long as it's an authentic product experience, we think that's valuable.
Speaker:As soon as questions to an AI platform goes into that realm of product or service
Speaker:and use and experience, you know, th- they need a lot of data which speaks to that.
Speaker:And we just thought that we're at the center of that, which was a bit of luck.
Speaker:It was important to shopping anyway, and very important, obviously, in e-commerce,
Speaker:particularly in the early years.
Speaker:And we just happened to have a huge, scaled data corpus.
Speaker:And so as we are crafting our strategy around, you know, thinking that we
Speaker:think this is a really huge resource and very beneficial to AI platforms,
Speaker:and then therefore the people that were using them, Yale and Columbia
Speaker:came out and put out a joint study that basically said AI platforms were
Speaker:using star ratings, review volumes, and recency as input as to what to surface.
Speaker:And that really validated that we were on the right track.
Speaker:And so I think the other thing that we believed is that the AI
Speaker:platforms needed guardrails that they couldn't get on the open web.
Speaker:So if they're just scraping for information, and they could only scrape
Speaker:so deeply into, say, a review corpus, they would end up saying things,
Speaker:'cause they're probabilistic models, they're like fortune tellers, that were
Speaker:inaccurate and potentially harmful.
Speaker:And so we actually know how to manage those things.
Speaker:All the, the human sentiment and unstructured data that we collect
Speaker:is actually connected up to a very structured spine of product information.
Speaker:And we know, for example, that you can't talk about alternate uses for
Speaker:Tylenol, like it, like it cures cancer, because that's fundamentally not true.
Speaker:It is not a claim that anybody's going to endorse.
Speaker:In places like Reddit, things might be said that are like that,
Speaker:and you don't… The AI crawlers don't know the difference, right?
Speaker:So we handle thousands of categor- categorical nuances, and so we
Speaker:thought that was gonna be, you know, part of our story as well.
Speaker:And so ultimately we, we approached the AI platforms.
Speaker:So we talked to the Googles of the world and the ChatGPTs, and
Speaker:we said, here's our hypothesis.
Speaker:We know that you're using star ratings, volume, and recency to surface.
Speaker:We think you have a problem because if you don't go deep enough in, beyond
Speaker:those superficial trust signals, um, you're gonna start surfacing very strange
Speaker:things that aren't necessarily true, and you want them to be true, and you want
Speaker:them to be… you want some guardrails.
Speaker:And we also think some of the places that you're finding the information
Speaker:to fill in the gaps that you have are just not good places to go, but we are.
Speaker:And they said, "You're right." So, so that's where we were.
Speaker:We had a good hypothesis.
Speaker:We had research that was independent, third party from very prestigious
Speaker:organizations saying it's an input, and then ultimately said, well,
Speaker:there's a deeper level here beyond just the, the superficial trust signals.
Speaker:We think they're valuable.
Speaker:We went to the platforms, and they said, you're right.
Speaker:They are, they are valuable.
Speaker:So we were lucky in that we didn't have to make anything up.
Speaker:Yeah.
Speaker:Um, they, they verified what we thought.
Speaker:We knew we were on solid ground.
Speaker:Then it went back to, okay, team.
Speaker:We understand these things to be true.
Speaker:Let's make sure that our messaging is really, really tight.
Speaker:What do we do?
Speaker:You know, we're the world's most trusted network of authentic consumer voices.
Speaker:You know, we source, display, and amplify ratings, reviews, photos, and
Speaker:videos at massive scale to drive reach, traffic, and conversion, and we help
Speaker:our clients' products be discovered, be trusted, and chosen by shoppers and by AI.
Speaker:That's, that's literally the entire messaging system in 30 seconds, and we
Speaker:use particularly that "be discovered, be trusted, be chosen" as kind of a core
Speaker:of our, uh, our, our marketing message.
Speaker:I like it, yeah.
Speaker:"Be trusted, be chosen." Any success metrics that you could share compared
Speaker:to any, like, pre-AI related, um, kinds of launches you guys did?
Speaker:Yeah, sure.
Speaker:So I think there's two things.
Speaker:The, the first thing is we were already using AI, uh, internally as, as
Speaker:part of a product platform for, like-
Speaker:Yeah
Speaker:content coaching and analytics and those types of things.
Speaker:So part of it was consumer-facing and, and interact with, uh, the
Speaker:people that were leaving the reviews.
Speaker:Some of it was more around, uh, analysis.
Speaker:The, the piece that we didn't have was like, okay, now, now we have something
Speaker:that really speaks to the AI platforms, and it wasn't just a feature, right?
Speaker:It was a, it was an output that someone could, could understand or an input
Speaker:into a system that drove an output, which is your product's gonna get,
Speaker:it's gonna get discovered on a shelf, which is a very small shelf, you know?
Speaker:Physical shelves, 400 products.
Speaker:Digital shelves, infinite number of products.
Speaker:But you put a s- filter on it, it's like really the first page, it's 10 to 12.
Speaker:Well, now with AI, it's between, you know, three and five that they're gonna surface.
Speaker:So being able to influence that because of the data source that you
Speaker:have i- is, you know, uh, a strength.
Speaker:And so that became part of the message.
Speaker:We said, you, you know, you need lots of data.
Speaker:You need authentic, accessible, and abundant content.
Speaker:This is gonna help you drive discovery.
Speaker:In other words, you're gonna be surfaced in those environments.
Speaker:And so the most natural thing for us to do is, is that, that, that
Speaker:leans into our sampling business.
Speaker:Yeah.
Speaker:And so people started buying a lot more sampling, so that was
Speaker:one thing that we saw right away.
Speaker:We also then answered with an innovation roadmap that leveraged what we knew
Speaker:to be true, which was the AI crawlers will only superficially scrape
Speaker:a certain amount of information, then they'll just kind of fill in.
Speaker:And we knew through experiments that we had did and that, that, that they
Speaker:also validated that if you, if you actually gave them our entire review
Speaker:corpus, the results that they would give Erica, for example, let's say
Speaker:you're shopping for a pair of sneakers.
Speaker:The, the result, the answer that they would give you, not only would they give
Speaker:you a recommendation of the different sneakers, b-but they would also give you
Speaker:a lot of contextual information around why this is the right sneaker for Erica.
Speaker:Because there's just so much more data to pull from.
Speaker:So what we did is we built something that allowed AI's, uh, crawlers to
Speaker:access a structured version of the data that they weren't deeply crawling.
Speaker:I know this is technical.
Speaker:Yeah.
Speaker:But it's just giving them a lot more information besides star rating, volume
Speaker:of reviews, and recency to go on, and then this influences both the product
Speaker:they're gonna surface and then how they actually talk about that product.
Speaker:Right.
Speaker:And as we've launched that product, we've done dozens upon dozens of
Speaker:implementations at this point, and, you know, within a couple of weeks,
Speaker:they'll start to see single-digit improvements in AI search visibility.
Speaker:Um, and over, you know, a longer period of time, upwards of 40% lifts, uh, in
Speaker:their ability to get, uh, surfaced.
Speaker:So it's a real product that works, and that's starting to drive the
Speaker:other part of the business, the software part of the business.
Speaker:Yeah.
Speaker:That's fabulous.
Speaker:Thank you.
Speaker:Um-
Speaker:Yeah, super cool.
Speaker:Any tips for other CMOs that may be hired to bring to market an AI product
Speaker:or capability or, or message, um, especially somebody who feels like
Speaker:they're kind of being asked to, I don't know, AI wash, you know, something
Speaker:that is not as natural of a, of a fit as Bazaarvoice, you know, has been?
Speaker:Yeah.
Speaker:Th- I mean, ultimately, you know, I, I've been, you know, pretty successful
Speaker:in my career, but luckily, you know, the reason I was successful is I believed
Speaker:in the products that I was selling.
Speaker:Um, I-- some people can sell anything and market anything,
Speaker:and I'm not one of those people.
Speaker:I have to fundamentally believe that the, the product I'm selling and
Speaker:the service that I'm selling, you know, has some sort of real benefit.
Speaker:And I knew there was a benefit when I came to Bazaarvoice because I had used
Speaker:it, obviously, at two other companies, both Unilever and, um, and Hershey.
Speaker:Right.
Speaker:And so I'm a big believer in content in general, and I'm a really big believer in
Speaker:that, you know, word of mouth is usually the best endorsement, so it's kind of a,
Speaker:this word-of-mouth company, essentially.
Speaker:Authentic consumer voices is exactly what, what the company's founded on.
Speaker:Bazaarvoice is about the voices in the old market bazaars of y- of, of old, and
Speaker:you trust what other people are saying.
Speaker:Go to this vendor.
Speaker:Don't go to that vendor.
Speaker:So fundamentally, I just really believe in it.
Speaker:I believe that people are fundamentally smart and that they can smell a rat.
Speaker:And, um, I think that you're trying to sell something where you don't really
Speaker:maybe have a proprietary data set or some sort of scale that allows you as a
Speaker:company, a software company, that to do something with your AI-powered product,
Speaker:um, that others literally cannot do.
Speaker:Your clients figure that out, and your clients also believe, and they're
Speaker:running enough experiments, that some of the things that you might be doing,
Speaker:they might be able to do themselves.
Speaker:Now, I also think, though, for those of you that I had just freaked out by,
Speaker:by uttering those words, that there is power in having domain expertise.
Speaker:And so I think SaaS companies that are just fully kind of powered by engineers
Speaker:that have been able to design very clever solutions, but not necessarily
Speaker:really understand their client.
Speaker:In other words, you know, it takes years for them to figure out product market fit
Speaker:because they don't-- They've got something that sounds like an interesting idea, but
Speaker:they have to literally test it over a very long period of time with their clients.
Speaker:So I think that kind of modus operandi, I think, is, is extinct.
Speaker:I think people can experiment very quickly now with these platforms, with
Speaker:ideas that they have about their clients.
Speaker:And so unless you have very specific domain expertise that is not
Speaker:easily replicable by your clients, then that expertise, I think, can
Speaker:also be a competitive advantage.
Speaker:So you could have something that, yeah, maybe on a technical level they could,
Speaker:they could create, but at a, at a fun-- at a deeper level, they don't have the
Speaker:experience or the know-how to actually use that tool to its greatest end.
Speaker:So I'll give you an example.
Speaker:Back at Unilever, I had a team that, uh, my IT team, and they were like:
Speaker:"Well, we can build ratings and reviews. We'll just build it." And I
Speaker:said, okay, how are you gonna get the content that we ingest to the retailers?
Speaker:And even more importantly, who's gonna moderate that content for safety?
Speaker:And who's gonna make sure that it's actual human content, number one,
Speaker:and, and who's gonna make sure that, for example, we don't do something
Speaker:stupid like publish a review that says Tylenol cures cancer, right?
Speaker:How are you gonna do all that?
Speaker:"Oh, we don't know."
Speaker:And I said, don't compete with a company that has domain expertise like that.
Speaker:If you don't have that expertise, then, yeah, you can build the widget,
Speaker:maybe, but you're not really gonna be able to compete and innovate
Speaker:and, and, and start to build on that product in a way that's compelling.
Speaker:Very interesting.
Speaker:And so I, I would think that if you're a CMO looking at one of these jobs,
Speaker:that gives you some ways to vet.
Speaker:Fundamentally, AI is really fickle.
Speaker:It, it's just not reliable, right?
Speaker:So y- the-- I think when you first start playing, and you're like, this is magical.
Speaker:This m- it knows everything.
Speaker:And then you start to find out it doesn't know everything, it's making
Speaker:stuff up, or it's trying to fill in gaps in its knowledge that it just doesn't
Speaker:have, and it sounds really great.
Speaker:You know, you have to be able to really work those problems in a deep,
Speaker:deep way consistently over time.
Speaker:It's not just like I stood this thing up and then I'm gonna
Speaker:go stand up something new.
Speaker:It's, it's like these things take consistency of execution, and they take
Speaker:real knowledge and understanding of what's the expectation of the output of any
Speaker:given tool, um, or service or product.
Speaker:I think that's just fundamental.
Speaker:As long as you can defend those things, like s- sure, you, you
Speaker:could stand up a loyalty widget, but, like, who's gonna do this?
Speaker:Who's gonna do this?
Speaker:Who's gonna do this?
Speaker:Oh, by the way, we have 20 years experience doing this, and you don't,
Speaker:and you've got it being run by, you know, a, a brand manager or an IT person, and
Speaker:they're gonna rotate out of that job.
Speaker:Who's gonna take care of that product when they leave?
Speaker:Do you have the right processes in place?
Speaker:All those things, you know, need to be built into products too.
Speaker:And so AI can seem very magical as… and a shortcut to certain things,
Speaker:but it doesn't replace know-how, good process, knowledge of the
Speaker:industry, knowledge of the space.
Speaker:You know, it doesn't, doesn't do those things very well-
Speaker:Yeah.
Speaker:At least not yet.
Speaker:Yeah.
Speaker:It's a great point for people who are wondering about that build
Speaker:versus buy decision, you know, within their own organizations.
Speaker:Certainly, there are things that we looked at, um, like content production,
Speaker:like our own content supply chain, which is something that I, I talked
Speaker:about at Bazaarvoice for a couple years and, and helped build at Hershey.
Speaker:And, you know, we were like, ah, we really need to kinda, like, better build
Speaker:this, and maybe we have the tools now.
Speaker:And we went out, and we talked to agencies about building it,
Speaker:and there was a huge price tag.
Speaker:And, you know, I had somebody on my team who's very confident, and she's like, "I
Speaker:can build this. I can build this, and I can make sure that it gets maintained."
Speaker:And I said, "Okay, have at it."
Speaker:And so you, you, you are gonna have a few, I think, use cases that, that really you
Speaker:probably could do as long as you have, you know, a foundational layer of very
Speaker:structured data that's well-governed.
Speaker:But a lot of other things are gonna be, you know, experiments that just
Speaker:never really get followed up on.
Speaker:They don't become part of the day-to-day.
Speaker:They're just nonsense.
Speaker:Yeah.
Speaker:Right.
Speaker:Right.
Speaker:So that was kind of the external view.
Speaker:You-- Let's look at the internal view.
Speaker:Can you talk about your organization and how you've built it and what's different
Speaker:now versus a couple of years ago?
Speaker:You know, whether there's like AI teammates showing up in the org
Speaker:chart or humans with new titles showing up on your org chart.
Speaker:What d- what does that look like?
Speaker:Yeah.
Speaker:Well, this is, this is fascinating.
Speaker:So, you know, I looked at my… Originally, when I took the, the
Speaker:seat about a year ago, I looked at the, the organizational chart, and I
Speaker:didn't know if it was right or not.
Speaker:I was used to an org chart, you know, out of the CPG, a
Speaker:big matrixed org chart, right?
Speaker:So the first thing I, I tried to do is just take a look at my org chart, and
Speaker:I, I recommend everybody to just stand back from it, pretend you don't know
Speaker:anything, and look at your org chart and ask yourself, "Could I explain
Speaker:this to somebody in about 30 seconds?" And they would go, "Yeah, I get that."
Speaker:And when I looked at my own org chart that I inherited, I'm
Speaker:like, "I don't understand this."
Speaker:So, um, I'm like, I don't understand why this is tucked over here.
Speaker:I don't understand why this is tucked over here.
Speaker:You know, these things naturally go together, but they're split apart.
Speaker:And so the-- A- again, back to my team, I'm like, okay, we're gonna have
Speaker:to redesign this to a certain degree, and, and here is my expectation.
Speaker:My expectation is that anybody, particularly a board member, could look at
Speaker:this and instantly go, "That makes sense."
Speaker:Mm.
Speaker:And if, and if we, if we can't actually do that, then it's not simple enough.
Speaker:And why did I want it that simple and, and, and, you know, like some
Speaker:of these things that were kind of stuck together that weren't natural?
Speaker:Because what I had the, the team do is I said, "You need to define for me what your
Speaker:teams do." And so they went, you know, we, we did a whole RACI exercise, and
Speaker:everybody defined all their work streams.
Speaker:We defined, uh, 473 work streams.
Speaker:And I said, okay, great.
Speaker:Based on the work streams, tell me what your charter is, and then we're gonna bump
Speaker:that charter up against what best in cloud looks like in software, and we're gonna
Speaker:see if there's even remotely a match.
Speaker:And of course, it was really messy, you know.
Speaker:Everything was overlapped.
Speaker:So we, we pulled out the overlaps.
Speaker:We, we got very specific about who was responsible, who was a combo, who
Speaker:would be consulted, who was informed.
Speaker:We took any work that wasn't work that was traditionally marketing,
Speaker:and we put it back in the functions that were adjacent to us.
Speaker:So there were certain things that were really needed to be with our
Speaker:product development team, and we said, we're not doing these anymore.
Speaker:We, we don't know why we're doing them.
Speaker:If this is for you… And oh, by the way, you're doing three things over here.
Speaker:This is actually product marketing's role or, or growth or whatever.
Speaker:And we pulled those back in.
Speaker:And then we just made sure that there wasn't any redundancies within the team
Speaker:as well from a work stream perspective.
Speaker:And then we re-ran kind of an exercise around now what do
Speaker:the charters look like, right?
Speaker:Which we used AI to help us with.
Speaker:And we, we look at that every quarter or so.
Speaker:It's, I want very, very clean charters and great specificity on who does what
Speaker:and why, and I don't want extra work.
Speaker:So there's a lot of stuff on there like, we wanna do these 15 or 20 things.
Speaker:I'm like, we can't even do our core marketing functions very well, so I
Speaker:don't wanna do any of those things.
Speaker:By the way, that's a hallmark of Doug.
Speaker:Everybody's like thinks I'm a technical guru from a digital marketing perspective,
Speaker:but mostly what I usually do is I just correct the foundations, and then the
Speaker:rest will just take care of itself.
Speaker:And this was the same thing.
Speaker:I'm like, don't talk to me about holding a couple hundred thousand
Speaker:dollars over here to do experiments.
Speaker:We're not even doing our basic job right.
Speaker:So let's do our basic job right, and then we'll talk about the experiments later on.
Speaker:So, we did all that, and, and then probably the biggest move
Speaker:I made is I, I put in a Director of Marketing Strategy and, and AI
Speaker:Transformation, and the job spec is very much like a marketing engineer.
Speaker:And that person works with that RACI and makes sure that for all the work
Speaker:streams that we have, that there's processes built, and that the processes
Speaker:are strong, and that the processes serve the business, not the people on the team.
Speaker:And once we know that the process is strong and it serves
Speaker:the business and not the team,
Speaker:then we automate it.
Speaker:And only then, because if you automate chaos, you'll just get automated chaos.
Speaker:You'll get more content, for example, if that's what you're trying to
Speaker:create, but it won't be any good.
Speaker:It's still is gonna be subject to missed deadlines and poor
Speaker:usage, et cetera, et cetera.
Speaker:So it's like, make sure that process is good.
Speaker:And then the other thing that role does, which I think is unique, is it
Speaker:makes sure that to the degree that we are gonna use AI to automate and
Speaker:transform the function, this was the call of this person who's a genius.
Speaker:She's like, you know, I, I studied everything, and she's like, well, most
Speaker:people are using AI for is just to, to automate their own individual jobs.
Speaker:But what they're not doing is they're not thinking about like
Speaker:the whole team and that process, but only thinking of themselves.
Speaker:It's a selfish way of looking at it.
Speaker:So she said, I'm not gonna work on anything, or we're not gonna automate
Speaker:anything, and I'm not gonna work on anything that doesn't actually focus
Speaker:in on a superhuman task, something that automates a process to the degree that
Speaker:like no hu- that humans just can't do it.
Speaker:And so she, she's very much focused in on the things that like were gonna make the
Speaker:most impact across multiple sub-functions in marketing, and then to, by extension,
Speaker:our product development organization as well as our, our revenue organization.
Speaker:So she's just like not gonna focus in on automating reports or task
Speaker:list or this and I'm, I'm only gonna focus my time on automating things
Speaker:that are gonna make a concrete difference, and the focus is gonna be
Speaker:on quality, not on quantity of outputs.
Speaker:Um, and so that's, that's how we've arrived at a, uh, a team now, which
Speaker:probably, she's got probably 20 agents.
Speaker:So I've got about 55 people, but she's probably added 20, quote unquote, "people
Speaker:helpers" to all those people, essentially.
Speaker:I've also got an operating system, a CMO operating system
Speaker:that mimics my functional leads.
Speaker:It tracks all of our conversations, all of our one-on-one documentation,
Speaker:all the budget trackers, the RACI, and about 20 or 30 other documents,
Speaker:and it essentially says, "Are we living by our marketing principles?"
Speaker:Which is to keep things simple, to be value-centric and be
Speaker:client-centric, and our goals and objectives, are we living up to those?
Speaker:Are we meeting those?
Speaker:And are we swimming outside of our lanes?
Speaker:And to the degree I see those things come through my, my robot overlord,
Speaker:I just reach out to the team.
Speaker:I'm like, "I need you to update this, this, and this," or, "Tell me why
Speaker:these things are happening, because this isn't what we agreed to."
Speaker:And it just, it helps me.
Speaker:And, um, it's like having, you know, two or three assistants.
Speaker:Yeah.
Speaker:That's very interesting.
Speaker:When you christened this person, this Director of Marketing Strategy
Speaker:and AI Transformation, was the idea that she would automate
Speaker:things with people or for people?
Speaker:Like, is she, is she building it and then giving it to people on the team or, or
Speaker:kind of sitting down with them and, and they're co-creating what they need to?
Speaker:So this is just like being a consultant, right?
Speaker:So she goes in, she's doing stakeholder interviews.
Speaker:You know, she kind of like, she's basically my deputy, so she gets to see
Speaker:literally everything that's going on in the function, and I give her access
Speaker:to a lot of stuff that most people wouldn't, and I'm like, so this is how
Speaker:this operates, and this is runs, and here's why I'm looking at these things.
Speaker:And she's internalized all that, so she understands the
Speaker:why behind certain decisions.
Speaker:And then as she goes in, uh, she, she's done, you know, intake processes with
Speaker:all the, the different leaders on the team, and in some cases, their teams, to
Speaker:really understand what they're working on, why they're working on it, whether
Speaker:they think it's effective or not, whether they think it's superfluous, redundant,
Speaker:stupid, or whether they think it's smart, amazing, and it's just, you know,
Speaker:like it's making such a difference.
Speaker:She takes all that information, and then she'll… You know, she's started
Speaker:kind of like in one area of content.
Speaker:It's the one area that I had her focusing on first.
Speaker:And she said okay, and she, she is building specifically for
Speaker:that part of the organization.
Speaker:And they're sitting down with her as she's building these things to
Speaker:basically test the outputs and how the whole thing runs so that they,
Speaker:they have veto power over that output, make sure the quality's there.
Speaker:And then a- as that starts to run, she's making sure that she's institutionalizing
Speaker:it, and in some cases, she'd get our back office, you know, our intelligence
Speaker:team involved and make sure that she can migrate the, the thing that she's
Speaker:built over onto something that's a sturdy kind of corporate structure.
Speaker:So then that's basically, you know, running on its own as opposed to just
Speaker:an experiment, and then she moves on to the next project and does it again.
Speaker:Got it.
Speaker:Yeah.
Speaker:That's really cool.
Speaker:Thank you for sharing.
Speaker:I love the specificity here.
Speaker:This is fabulous.
Speaker:And let's talk about interviewing.
Speaker:Do you have a favorite interview question when you're meeting people-
Speaker:I, I don't.
Speaker:-That could be fits for your team?
Speaker:No.
Speaker:Really?
Speaker:Yeah, some people do-
Speaker:You, but you seem so, you seem so structured.
Speaker:You seem so… You know, you have your RACI stuff.
Speaker:You have your, you know, rigorous data foundations.
Speaker:This is so important for you, and so I'm surprised.
Speaker:Well, it's in- it's interesting you say that because I, I, I don't
Speaker:actually think I'm that structured, but a lot of other people say, no,
Speaker:no, you're super clear what you want, and you repeat it all the time.
Speaker:So every staff meeting I'm like, "What are the marketing principles?" And,
Speaker:and then I say, "Okay, what are the corporate objectives?" And then I say,
Speaker:"What's the core messaging system?"
Speaker:And, like, literally it's at random.
Speaker:My entire team needs to know it.
Speaker:And I, I've told them, I said, "The reason I'm doing that is because it
Speaker:becomes like your guidebook for any decision that you're gonna make."
Speaker:The first thing is like, I just wrote something.
Speaker:Is this making it simpler for sales to sell what we do?
Speaker:And if the answer's no, then maybe you should rewrite it.
Speaker:You know, or whatever.
Speaker:And, um, so that's why, like, I literally just talk about it
Speaker:over and over and over again.
Speaker:At every all hands with like to the entire function I do the same thing.
Speaker:We have to start with the marketing principles and the objectives and whatnot.
Speaker:So I do that, and I do that in interviews.
Speaker:And then I, and then I basically have people reflect on it, and
Speaker:it, it does a couple things.
Speaker:The people that react to it really positively are like, "Oh my God, you know
Speaker:exactly what you want. And you're really clear in terms of your point of view."
Speaker:People will react to that, and that, that's kind of the question itself.
Speaker:How do they react to that kind of like intense kind of onslaught of like,
Speaker:"Here's what the company's doing. Here's what this function's doing.
Speaker:Here's what I'm concentrating on."
Speaker:One, two, three.
Speaker:One, two, three.
Speaker:Even my strategy is one, two, three.
Speaker:So I get reactions to that, and then I give people a lot of context
Speaker:of like, here's why, um, I wanted to speak to you about this role.
Speaker:And I give them a lot of context about the good, bad, and ugly at the company.
Speaker:And that usually could take… I mean, that whole foundational layer
Speaker:could be a third to a half of the interview, and then the rest is I just
Speaker:shut up and listen to them reacting to it and asking clarifying questions.
Speaker:And if they're not like really digging in and asking clarifying questions and,
Speaker:and saying, "This is where I can help you," then they're, they're probably
Speaker:underprepared or not experienced enough, or, or just the wrong candidate.
Speaker:Great.
Speaker:I know we only have time for a little more dialogue here, but could you
Speaker:share a mistake or a lesson from your experience so far, whether it's building
Speaker:an AI value prop for the external market or, you know, internal change
Speaker:management or internal hiring or any- anything that, that comes up for you?
Speaker:I would say like big mistake, I think at Bazaarvoice, I think, um I probably… I
Speaker:was, I was hired for a certain job as a consultant, like an internal consultant
Speaker:essentially, that was a client expert.
Speaker:And I think that the mistake I made that I clearly am not repeating as CMO,
Speaker:because I've learned my lesson very mu- you know, like, in a, in a big way.
Speaker:But when I first came on board, I just, I just assumed that e-
Speaker:everybody at Bazaarvoice had the same facility with how the, the software
Speaker:kind of ecosystem works at a client.
Speaker:All the system integration that had to happen, what was important, what really
Speaker:wasn't important, how to think about ROIs, how to think about alternate
Speaker:use cases, how to think about where Bazaarvoice sat in the broader ecosystem.
Speaker:I just assumed everybody really understood that.
Speaker:And so the big mistake that I made was I would, I would get deployed against very
Speaker:important clients with a very high-level stakeholder, and we'd have these amazing
Speaker:conversations and, and they'd be like, "Oh my god, you know so much about the
Speaker:space, and this is really helpful."
Speaker:And I'm like, "Great." And then I'd communicate it to the sales team, and
Speaker:the sales team are like, "We have no idea what to do with that information."
Speaker:Because I was speaking, like, C-level to C-level, and I wasn't doing what I've
Speaker:been preaching for this entire call.
Speaker:I wasn't simplifying it and saying, here's the, here's the next three steps
Speaker:for you, sales force, to, to continue to have the conversations with these people.
Speaker:I was kind of assuming they could watch me do it, and they would get it.
Speaker:And that's not what they're paid to do, and, uh, that's
Speaker:not really the expectation.
Speaker:And I underestimated how much work they have on their plate, and I underestimated
Speaker:how hard it is to close a sale in a B2B SaaS environment, which doesn't
Speaker:have, like, steady state demand.
Speaker:You create your own demand.
Speaker:I mean, I understood it intellectually, but I hadn't felt it yet.
Speaker:And so I, I really, it took me about a year to, like, really figure out that
Speaker:I could be just as harmful as I could be helpful if I didn't understand what,
Speaker:what to do with the conversations that I was having with, with our clients.
Speaker:Um, and really spell it out for the sales force, not because
Speaker:they couldn't handle it.
Speaker:It's just I needed to spend more time explaining, here's why I said
Speaker:these things, and here's how you connect it to Bazaarvoice, and this
Speaker:is why this would be a good thing.
Speaker:Yeah.
Speaker:Yeah.
Speaker:But a lot of that muscle now is being used in the new talk tracks.
Speaker:So I literally have the levers to put my brain and my client thinking into sales
Speaker:assets, into webinars, into our events, and all the … And now it's working
Speaker:because it's built into the system as opposed to kind of like an overlay.
Speaker:That's great.
Speaker:Thank you for sharing.
Speaker:Final question for you.
Speaker:You know, you're in a PE-backed setting.
Speaker:Any tips for other CMOs of PE-backed companies on how to talk to their boards
Speaker:about AI and what they're doing with AI, how they're moving the needle with it?
Speaker:Especially when, you know, kind of boards are worried about the value
Speaker:of their SaaS companies, you know, cratering and… Any thoughts there?
Speaker:Yeah, for sure.
Speaker:Well, as, as always, keep it really, really simple.
Speaker:So it doesn't matter whether it's a public firm, which I've
Speaker:had a lot of board experience.
Speaker:It doesn't ma- it doesn't matter whether it's Wall Street analysts, of which
Speaker:I've had the pleasure of interacting with a lot of folks on Wall Street.
Speaker:It doesn't matter whether it's private equity or even a private company.
Speaker:You're, you're fundamentally talking about something where their
Speaker:expertise lies in a different area.
Speaker:So PE folks, you know, typically the boards, they'll have a
Speaker:few subject matter experts.
Speaker:Yeah.
Speaker:But a lot of them are, you know, at heart, they're
Speaker:super-educated financial engineers.
Speaker:And they're just, you know, so bright.
Speaker:But, you know, they'll, they'll throw questions at you and, and if you don't,
Speaker:if you don't make sure that you're like very clear and simple and like,
Speaker:"This is what we can do versus what we can't do," you know, um, they're just
Speaker:as susceptible as, as any human to the belief that like, but these, these
Speaker:are magical tools that could do this.
Speaker:And it's like you have to kind of… You know, we had to talk
Speaker:them through our positioning and why we thought it was gonna work.
Speaker:And I have to say our, our, our firm, THL, is, they're very dynamic and
Speaker:very smart and very willing to change.
Speaker:Like, once they realize they need to change, they're gonna, they're gonna move
Speaker:really fast, and they're super supportive.
Speaker:But yeah, you have to take them along for the ride, and if you rely on kind
Speaker:of software speak and a lot of technical mumbo jumbo, you're gonna lose them
Speaker:just like you lose a client, right?
Speaker:And so you have to really simplify.
Speaker:Uh, and so that'd be my recommendation.
Speaker:Particularly, I've been in digital marketing now as kind of
Speaker:my, my specialty for 15 years.
Speaker:I can't even believe I can say that.
Speaker:And that was the, the first criticism people have given me, like, "You
Speaker:sound so smart, but we don't understand what you're talking about."
Speaker:So, and ask a lot of questions like, you, you know, and not in a condescending
Speaker:way, but like, "Do you understand X, Y, and Z?" And just making sure
Speaker:that they, they're really tracking.
Speaker:And then that will give them the confidence that you have the problem at
Speaker:hand, that you understand the problem, and whether you really have a problem
Speaker:or like from a PE perspective, whether you really have a problem on your hands.
Speaker:Like, this is just like anybody can do this now, and there's no value in
Speaker:this company, so we have to transform this company into something that does
Speaker:have value so that we can get a return.
Speaker:Or it's like, hey, we're, we're actually sitting on a gold mine,
Speaker:which I think is Bazaarvoice.
Speaker:That's, that's okay.
Speaker:If we're sitting on a gold mine, we do these three things, and
Speaker:all of a sudden it's great.
Speaker:So I can't take credit for the success that Bazaarvoice
Speaker:is having right now fully.
Speaker:Certainly the messaging I, I nailed some of the hypothesis, but I had a lot of,
Speaker:you know, help from, you know, our CTO and from our CEO and the board themselves,
Speaker:you know, a lot of listening and whatnot.
Speaker:It was de- definitely a group effort.
Speaker:But, you know, we figured out a way to like really simplify our message and
Speaker:understand where the value was gonna come from and start to move the company
Speaker:towards slightly different version or slightly different North Star.
Speaker:So- yeah, yeah.
Speaker:Great.
Speaker:It be- gotta be simple and truthful, I think is the key thing.
Speaker:Fabulous.
Speaker:Thank you so much for coming on the show, Doug.
Speaker:This has been great having you.
Speaker:Thanks, Erica.
Speaker:That was Doug Straton, CMO of Bazaarvoice.
Speaker:Some great themes here about the importance of relentlessly simplifying
Speaker:your AI story, of becoming both trusted and chosen, and of ensuring that your
Speaker:marketing and your AI initiatives benefit not just individuals,
Speaker:but your company as a whole.
Speaker:Thanks for listening to The Get.
Speaker:I'm your host, Erica Seidel.
Speaker:The Get is here to drive smart decisions around recruiting and
Speaker:leadership in B2B SaaS marketing.
Speaker:We explore the trends, tribulations, and triumphs of today's top
Speaker:marketing leaders in B2B SaaS.
Speaker:For more about The Get, visit TheGetPodcast.com.
Speaker:If you like The Get, please share it, and please leave a review.
Speaker:To learn more about my executive search practice, which focuses on recruiting the
Speaker:make-money marketing leaders rather than the make-it-pretty ones, follow me on
Speaker:LinkedIn or visit TheConnectiveGood.com.
Speaker:The Get is produced by Evo Terra and the team at Simpler Media Productions.
