AI's New Era Innovation Strategies for Entrepreneurs

In this episode of the Beyond Normal Podcast, Kenny Groom sits down with Mbere Monjok, Technical Program Manager at Google, to talk about building real value in this new AI wave.

Mbere shares the biggest mistake she sees founders make: leading with "we're AI-powered" instead of the problem they actually solve. She digs into where AI is delivering real ROI right now (turning unstructured, messy data into dashboards and decisions), why founders should treat AI as a thinking partner rather than just an input-output tool, and what trustworthy AI actually requires when the tech gets things wrong.

The conversation also covers opportunity for Black founders in this moment, why distribution can outperform product, and why founders should choose quality over speed for the long game.

If you're a founder trying to figure out what's signal and what's noise in AI right now, this episode will hit home.

Transcript

Kenny 0:49

Welcome to a special quick time edition of the Beyond Normal podcast. I'm your host, Kenny Groom. Short conversations, fast pace, big ideas for the founders out there. Today, we're joined by Mbere Manjok. She is the product strategy leader at Google, working on the intersection of AI, media, emerging technology overall. And in addition to that, in her career, she's also consulting and doing some amazing work bringing, knowledge, to founders on these hot topics. So without further ado, I'm gonna bring Mbere to the stage. How's it going?

Mbere 1:27

Going good. It's a good Friday so far. excited to be on this podcast to share more about my background and my journey

Kenny 1:35

Love it. Love it. So I, I gave a quick intro for the folks, right? I did high level, 10,000 feet in the air. Let the folks know what you've been focusing on and what, what your passion is and why we're having this conversation today

Mbere 1:47

For sure. So I'm Mbere. I am a program lead at Google, where I've been working on shipping and launching gen AI features at the intersection of, just product strategy as well as digital commerce. been very excited about building an innovation. I've always worked in that space. Started my career off in consulting, building large enterprise scale opportunities for clients that wanted to do something new in their industry,

Kenny 2:13

And then now it's connected to Google to build things within the AI space. So I love working with

Mbere 2:19

new

Kenny 2:20

new and exciting technology. So

Mbere 2:21

working with

Kenny 2:21

with founders is

Mbere 2:22

my

Kenny 2:23

passion on top of my day job. So spending time consulting on AI strategy, growth hacking, you know, branding are some of the things that I spend,

Mbere 2:32

time

Kenny 2:32

doing. I love it, I love it. And so you being in that, in, in that role you're in right now, you see a lot of things what's going on in this space, specifically around business, the use cases for some of these cool technologies. The challenge we're seeing now when I talk to other founders, I myself, there's just like, it's hard for us to determine like what's real and what's noise. There's a lot of noise, there's a lot of ads coming at us. So what's a, what's a big mistake, you know, you see founders making when it comes to this new way of, of AI technology today?

Mbere 3:08

No, that's a good question, and it's something that I talk through with a lot of the founders that I've spoken with. I go to hackathons or see them present their product or their use case, it's always about the capability of AI and, like, "We are AI powered," or, "We're

Kenny 3:23

Providing an AI solution as opposed to what's the true value that your product is bringing to the consumer. I think we're getting away from thinking about what

Mbere 3:31

people

Kenny 3:32

really need and why we're building something for them in the first place, as opposed to talking about the tech stack as the most impressive part. When you are building something,

Mbere 3:41

the

Kenny 3:42

highest value you can provide is what lead your--

Mbere 3:45

what

Kenny 3:46

you're alleviating for your

Mbere 3:48

user, and

Kenny 3:49

and I think leading with that is more important than what I've been hearing about lately. I love that. it's so easy or it seems so easy what you just said, right? To whatever you're doing, the solution you're providing to say, "Hey, I am the AI-powered version of that." Right. Right? But it's like what you're solving is what's probably going to bring in the money and not just saying you're AI-driven by itself. It seems intuitive, but for some reason as a founder, it's hard for us to get, you know, wrap our head around that. And on the flip side of that, I know you're seeing some use cases out there where AI is adding a lot of benefit by default. Can you talk about maybe a use case that you've seen where AI is really creating some, some real value for the businesses, and maybe we could pay attention to that, you know, as founders and business owners to see how they're doing it in that industry. Anything sticking out?

Mbere 4:45

There's a couple things. So one, the big areas that was really hard to in on was unstructured

Kenny 4:53

data, taking the information and then creating some sort of view or summary so that you can act upon it.

Mbere 4:59

So

Kenny 4:59

that could be review data, that could be, you know, what you're getting from customer requests, like sales information that is not necessarily in a format. Understanding how to get that data

Mbere 5:13

scrubbed

Kenny 5:14

in a format, a view, a dashboard has been really valuable for founders who didn't have time to either hire somebody to build this code and to be able to visualize it in the way that they want. So the, the vibe coding era has given founders a way to really think about their analytics a little more and find the types of data that was hard for them to even like look into in the beginning. Mm. That's dope. So they-- it's helping synthesize some of those things on the back end so th-those decisions can be made quicker.

Mbere 5:43

Yep.

Kenny 5:43

I love that. I love the sound of stuff like that. anything where I don't have to spend my day in a spreadsheet and I can actually talk to customers is powerful for me.

Mbere 5:53

Yeah, you can build that really quickly just by, like, taking the information you have or even prompting for what additional sources haven't you considered. So w- a lot of times we think about giving AI the inputs that we have. thing that I have encouraged founders to do is, like, you ask it what you should be doing. Like, think about it more as, like, that additional

Kenny 6:12

partner, like you don't always have to use what you know as the baseline. It could maybe think about things that you haven't thought about. Obviously, it's AI,

Mbere 6:21

You

Kenny 6:21

you know, you are the steerer of the ship, so you still control the narrative. But sometimes if you don't know how to do something, prompt it to

Mbere 6:28

give you

Kenny 6:28

you a direction so that you think about it the right way. Mm. I like the product lens that you're bringing to this conversation. I don't typ- I don't always wear that hat because I'm trying to, I'm trying to find a solution right in that moment that serves a need for our customers and make that transaction a lot of time as a business owner, as a founder. So I'm curious, you know, what's-- for you, what separates that great product from, say, the g- showing that great demo, that shiny, you know, ad to a customer that gets them to buy for you? Like, w- what does a great product look like separate from just showing somebody a demo or something where it looks cool and they want to buy in?

Mbere 7:10

Yeah. I think a great product has specificity that is really targeted to you know, th- like, there's the user and then there's, like, the sets of problems that that user may have

Kenny 7:22

Depending on the position that you're in. The more you can say like, "Hey, here's a journey I know you can walk through, and here's how this particular thing can

Mbere 7:32

fix

Kenny 7:33

it for you," whether that demo is, you know, is robust or even short. You know, it doesn't necessarily need to be a demo that has like all the bells and whistles, but that specificity of like, "Here's

Mbere 7:42

one

Kenny 7:43

way we can, like, fix this issue, and then here's the vision for this." Demos are meant to be short, crappy, quick, and we can do that now. But the part that you said, using the storytelling about what's next after this piece and how is it continuing to build upon the

Mbere 7:57

the, the

Kenny 7:58

pr- problem you're solving. Mm. You mentioned storytelling in there. I lo- I love that. I love that. And, and so going back to what you've seen founders maybe... I think for me, I'll speak personally, right? For me, when it comes to, like, the product strategy and like you said, laying that all out, sometimes it can be over-- it can be complex to somebody, right? You got all these flows and all these things on the back end and stuff is talking to each other. So I'm curious, like, like what have you seen founders overcomplicate when it comes to delivering value for their product? Like, is there a certain piece of, of that process where you think founders are just thinking about it too much, and they should re- should really simplify that for their customers?

Mbere 8:45

Yes. So a couple things. I think, one area has been on, what, do I get data from users? You know, there's surveys, you can go out and do focus groups. just ask people in your, like, immediate world really quickly. So getting...

Kenny 9:04

Or ask social media. I know it's not always representative of the large base, but you need to find the quickest path to getting the information about whether people like your product, like your offering, and sometimes overthinking the, nuts and bolts of, "I gotta create this survey, I have to send it out." You can just do a quick poll on Instagram and get

Mbere 9:24

immediate

Kenny 9:25

signal. y'all prefer this date or that date? Would you like to have this color or that color? If I did it in this shape or that shape?

Mbere 9:31

you

Kenny 9:31

You know,

Mbere 9:32

Quick

Kenny 9:32

signals. Sometimes it's a little hard to get your head around easy. You should still build the infrastructure to get data and synthesize it over time, but

Mbere 9:44

not

Kenny 9:44

thinking about,

Mbere 9:45

you

Kenny 9:45

you know, not hindering yourself from moving forward if you have just quick signals at hand. So that's one area. Mm-hmm. I like that. And it's-- And l- you're, you're speaking, you're speaking, you're, you're painting this picture for me where it's like, as a founder, you say, "Hey, I wanna get it to maybe 100%." It's not gonna be 100% from day one, right? You may get it to 50, 60, 70, and then that feedback from the customer is really what gets it to where we hope it gets to 100%. It probably ne-never does. Yeah. but that's a conversation for another day. But I'm curious with that because you're, you're taking us... this segue is amazing, by the way. You're taking us into the idea of building trust- Yeah right? Early and often, getting that feedback. So in this age that we're in now, like, what does trustworthy AI even mean?

Mbere 10:37

Oh man, that's a loaded question because it's a, one AI in itself is prone to giving you things that may not be true.

Kenny 10:46

Mm-hmm.

Mbere 10:48

but we have to work within the confines and the constraints of that reality

Kenny 10:53

While doing our best to overcome that with the information we have, with the training of the model that we have. I think being transparent about how you're overcoming those challenges

Mbere 11:03

is,

Kenny 11:04

number one, really, really important. If you're using, you know,

Mbere 11:08

data

Kenny 11:08

that's missing information for Black and Brown people, then, you know, say that this is how we're hoping to close that gap, but this is where we're starting right now.

Mbere 11:19

and

Kenny 11:19

and like not just saying that, but having a true plan and action to close it. Having those disclaimers, like the privacy disclaimers, making sure people are aware that like this is a risky territory to enter depending on what your product is. Different industries have different standards, so you need to take that into consideration. But the honesty piece has to be there. A true plan to close the gap has to be there. And then also having people share what they're seeing that is incorrect. So you let your customer be part of the process with you. It's like, "Hey, if we sent you information that you know, you knew that was like this is incorrect or this image wasn't exactly what it should have been, give us that feedback and let us know what happened because it only helps, the factor." That is powerful. it's, it's so weird that you said that. To me, it seems we're at this point where we all have said, "Hey, we've got too much data." But then, like you just said, like there's gaps in the data, right? Like there's points where Black and Brown people may not truly be represented, and then you got people building proxies and building a-- making assumptions on that, on these, on these groups of people, and those, those assumptions at times can be harmful.

Mbere 12:35

Yeah

Kenny 12:35

Do you think we're at a point now where the trust, having that trust is actually more important than the capability of these tools itself?

Mbere 12:45

I do think so. I do think that having a mechanism for letting people be comfortable with AI, and sometimes that is really leaning into the change management. At the end of the day, if you're in the

Kenny 12:58

AI bubble, and I suffer from this too, where I'm--

Mbere 13:00

Because

Kenny 13:01

I work in the space every day and I work with people who are really excited about the change,

Mbere 13:05

and

Kenny 13:05

some healthy cynicism, I think everyone's at the same level that I'm at. And the reality is that most people don't have that level of exposure and are unsure of how to even get to that level. So change management around like introducing things, giving education around

Mbere 13:23

having

Kenny 13:24

people understand here's how you can use it. If you are skeptical, here are some, you know, protocols you can do. Maybe you give people the ability to like confirm before things move forward,

Mbere 13:34

you

Kenny 13:34

you know, before you connect it to your Gmail or your bank account. Like you need to have something that helps people along every journey, with, with, with AI, and make sure that, like, you know, we can handhold those who are not as close to it, to get there. That's a great point. I'll be honest, I connected quite a few apps, and very, to, some of these tools.

Mbere 13:58

It's, that's okay

Kenny 13:59

You got me thinking like, what exactly did I sign up for?

Mbere 14:03

Yeah. Well, I, I think there's going to be a point where we need the level of connectivity to be able to be as productive as we want. Now, yes, there's still risks that you need to take into consideration, but we also can't test out... You know, it's a double-edged sword, right? You want AI to get better at certain things like finance and healthcare, but we're also

Kenny 14:25

Sensitive about how quickly we move in that area because of the level of like privacy and personal information that is available, and the, ability for people to like take advantage of that. So a, a lot of people in like AI space are kind of grappling with how do we push and pull, like push people enough to say like, "Hey, we need to get more comfortable," but also keeping the guardrails in place because we know that there's real risk in play, so. Mm-hmm.

Mbere 14:52

an

Kenny 14:52

Except for that, it is a-- There-- You know, we're moving forward, and that's the

Mbere 14:57

the

Kenny 14:57

best that we can do is taking one step at a time every day. Yeah, I love that. segueing there, you, you mentioned moving forward and right. So a lot of founders, a lot of business owners are just trying to figure out like where the opportunities are gonna be in the next few years, right? For themselves specifically. where do you think founders should be paying attention over the next few years in terms of getting value from these tools?

Mbere 15:22

Hmm. Getting value. So there's, there's a range. There's using it for, like, input and output. Like, I give you informa- I give the AI tool that I'm using information, and it gives me something back. It's like, "Here's an image, here's an audio. I created a slide deck, a book." You

Kenny 15:41

You want to get to a place where you're training.

Mbere 15:44

and I

Kenny 15:44

I know everyone's been talking about agents, and I

Mbere 15:47

hate to

Kenny 15:47

to bring this point up again, but having workflows, tasks, and capabilities that can run outside of you providing input constantly

Mbere 15:59

is

Kenny 16:00

is key to

Mbere 16:01

like,

Kenny 16:01

productivity, right? So if you know that this is something that happens on a reoccurring basis and you need to take that off your plate, training an agent to be able to, like, understand that workflow really well, putting the checks and balances in place, obviously, and, like, human review is still important at certain points. But if you want to unlock true productivity, getting past the initial input-output phase into

Mbere 16:26

having

Kenny 16:27

someone kind of be an assistant, and by someone, I mean an AI tool. Mm-hmm. Yeah, that's a, that's a good point. There's so-- I think there's, there's so many tasks that, like, they just end up piling up on a founder's plate, and they never truly get done. And so whether it's tech debt, whether it's just, you know, business- Emails or like-

Mbere 16:49

understanding that you're sent, summarizing conversations, things like that

Kenny 16:55

Mm-hmm. And it just help, it, I, I see the benefit, like the efficiency already, which I can tell it, "Hey, just like remove the gunk from my email box." Like just remove the stuff where I've never, I don't know who this person is. Unsubscribe me. Unsubscribe me. Oh my gosh, the unsubscribe, uh... it's so crazy. Like I, I used to download, and it shows where we are with technology, I would always download those apps that would be like, "Hey, here's the email chain that you, or the, the company that you haven't, you, you haven't interacted with them for years. You should probably unsubscribe." I probably don't even need someone a lot of times to lay out the list for me versus just unsubscribe me, right, from some of these things and, and

Mbere 17:42

sort of scheduler that just does that automatically

Kenny 17:45

Yeah, yeah, I like that. I'm gonna look into that, because I've already given these tools all the information to, all my systems, like you mentioned earlier. I really need to check that. But I, I wanna s- I wanna stay on this topic, around how businesses, founders specifically, can get value opportunities from this space. I wanna dive into Black founders specifically.

Mbere 18:09

Yep, yep

Kenny 18:09

are, i- is there something like a specific call out that you may have for, for Black founders to make sure we're represented in this, this AI wave? Like what, w- what should we all be doing in this moment?

Mbere 18:22

That's a good question, yeah, because what I have seen, at least, like, when it comes to, like, where AI fear is and AI understanding, you know, the, the numbers show it's, like, Black founders being behind.

Kenny 18:37

and we have a lot of opportunity at hand to build things that are sustainable long term and also create, like, like being a founder is also being able to create wealth for the future for yourself because you have like a, you have something that you own. So if you want to use one-- I mean, that's also a

Mbere 18:57

prompt

Kenny 18:57

that you can do is like how can, as a Black founder, build something that is sustainable, build something so that others can use. Like reaching back and also pulling people forward in your industry is really important. There's always not that many of us. so when you learn something, can you create communities? Can you get together and share? Sharing knowledge and information is also something that I know that doesn't happen as often as it should. and we should have more conversations around building businesses, building wealth, closing deals,

Mbere 19:27

what type of partnerships. Like a lot of times I see Black founders building silos and not reaching out for help. Mm-hmm. Having partners who are better at something other than you, and it takes a little bit of humility to like maybe recognize that. It's like, "Hey, I'm good at this part in my industry. Maybe someone else is building something and they're doing well with partnerships. Let me tap in and maybe get some information." I love that. I love that. It, it, it's interesting 'cause the AI does uncover, like we can use it to uncover parts of our business Yep

Kenny 20:02

where we may be, you know, like me for example, accounting, anything where it's like the back end and making sure this account talks to this one, I ain't your person. I need to have an accountant-

Mbere 20:14

well.

Kenny 20:14

Someone

Mbere 20:15

really well

Kenny 20:15

who does that really well. Yeah, someone, yeah, somebody does that really well and, and we gotta be comfortable, to your point, having that, you know, fig-figuring out what our supply chain is, right, for our business. that's gonna be key.

Mbere 20:25

about, like, an accountant. Like, someone probably who is a podcaster has figured out how their accounting system is dope.

Kenny 20:33

Hmm sharing and being deliberate about that is, like, really important because we're-- if we're gonna learn from others, we've gotta learn from each other. Yeah, I see what you're saying. I see... It's interesting that you say that. I haven't seen-- I've seen some of that, but there's definitely room to grow in terms of like, "Hey," like to your point, "fellow podcasters, I built this tool. Hey, here, here's access to it. Feel free to make it your own prompt and, like, do it your own way, but here's how I set the system up." I've seen, I've seen some of that, but to your point, people are still, like, tinkering,

Mbere 21:10

And

Kenny 21:11

know? And like-

Mbere 21:12

Someone's

Kenny 21:12

done this before.

Mbere 21:13

We

Kenny 21:13

Mm-hmm.

Mbere 21:14

hard. We're working too hard.

Kenny 21:16

It's too hard. Mm. I like that. Yes, I do not wanna work too hard on stuff I don't have to. That's a very-- That's, that's my takeaway right there.

Mbere 21:23

you're working

Kenny 21:24

It's too hard. Yeah. All right, so now we're gonna enter a little bit... Well, th- this makes sense for our quick time session. We're gonna enter a lightning round. Whatever's the first thing, first word, first item, first sentence that comes to mind, just spit it out. All right?

Mbere 21:40

Okay, okay, I'm ready.

Kenny 21:41

All right. most underrated founder skill

Mbere 21:44

Ooh. definitely I think, yeah, while it is something that you need to do ongoing, I think people don't spend enough time to think about like, "Who do I really wanna be long term?"

Kenny 21:56

Got it. Most overrated founder skill

Mbere 22:02

Overrated founder skill

Kenny 22:05

Is there one in your mind?

Mbere 22:06

I was like, you kind of need all skills.

Kenny 22:10

That's a answer too. That's a answer too

Mbere 22:13

But the most overrated skill is, like, you spending time on the one that's not your bread and butter

Kenny 22:18

Mm-hmm. Okay. Okay. You just being able to do everything and be a jack of all trades. Yeah. Okay. Yes. I love that. I love that word. Best AI tool you're using right now

Mbere 22:33

Best AI tool I'm using right now. So I'm spending a lot of time in Gemini, doing a lot of, like, image creation, using it for, you know, infographics, doing some, like, image editing work as well.

Kenny 22:45

So, you know, we do have a great image generation tool, so- Mm-hmm almost vintage. You know what? I actually, I went to my favorite restaurant a couple days ago, and they talked about using Gemini's, image, generator. They typically would have to hire graphic designers and kinda go through this proce- this long, drawn-out process.

Mbere 23:10

Exactly

Kenny 23:11

a picture of their burger, their sandwich, and they said, "Give me a flyer that says blank, blank, and blank." And before you know it, by the end of the day, they took, sent it off to the printer. Like, I'm talking, like, the big stuff. So you can, like, see if it's not right. And visualize it. Yeah, and visualize it. Prompt and make the changes, so-

Mbere 23:28

I'm currently doing a lot of, like, infographics, so the image generator is, is, is helping me out

Kenny 23:35

Yeah, like the, the flyers for like clubs and stuff like that, I'm noticing a couple years ago they weren't good, but now they're like, they're, they're legit. Like they look really

Mbere 23:46

say on that point, infuse a little bit of personality and specificity.

Kenny 23:51

Yeah, for sure.

Mbere 23:53

they're

Kenny 23:53

Starting to look the same. Yeah, good point, good point, good point

Mbere 23:56

might be using the same prompt. I don't know, but, you know, please

Kenny 23:58

Good point.

Mbere 23:59

fits

Kenny 24:00

right. All right, one thing founders should stop doing tomorrow, 'cause the day's over

Mbere 24:07

Yeah. Stop doing tomorrow. I think they should stop overthinking. I think they should definitely, if you have an idea, test it. Test it quickly. You don't have time to sit on something and realize, like, down the line someone else has either done it or, you know, you could have moved a little faster and been the, been, been the one in front. AI is letting people move really quickly. So if you have something, test it, get the feedback really quickly, and kill it if it doesn't work Okay. Which one of these ranks higher, for you for a founder? product or distribution? Hmm. You know,

Kenny 24:44

Out the gate

Mbere 24:46

I would say you, you need a good distribution channel, 'cause there are some mediocre products that they're get- they're getting real- they're getting everywhere.

Kenny 24:53

Mm-hmm.

Mbere 24:53

People are

Kenny 24:54

we, we all know what you're talking, and we all, we all know those products. We all see them every day, and it's like, how is this product... Oh, they've got the distribution. They've got the, the channels to make sure we,

Mbere 25:05

channel strategy, go

Kenny 25:06

all see it. We all see their product every day, all day. And with that, they're gonna get some sales of,

Mbere 25:11

Yeah.

Kenny 25:12

of their product

Mbere 25:13

that back end is really, really important

Kenny 25:15

I love that answer. And then, rank these two, speed or quality out the gate for a founder

Mbere 25:23

You're gonna need some quality. I, I'm a big quality person. I think speed is important, but it will kill you in the long run. And if you're thinking long run, just focus on that quality

Kenny 25:32

Got it. Got it. Well, appreciate you, for taking part in our first lightning round. Hope it wasn't too stressful, but you gave some incredible answers for us. I just wanna say, you know, thank you for having this conversation. I took some notes on my end of some things I need to do, but I know our listeners, they're definitely gonna have some takeaways from this. you're doing some amazing things in this space, and so appreciate you for, having this conversation and coming on our platform today. Final, final question,

Mbere 26:03

Okay

Kenny 26:04

the hard truth about innovation that people, founders need to hear in this moment?

Mbere 26:12

If you don't hop on the bandwagon, it will leave you behind

Kenny 26:15

Hmm. That's a mic drop.

Mbere 26:19

It,

Kenny 26:20

It

Mbere 26:20

you behind. So you can't be scared, and you can't say that it, the time is not now.

Kenny 26:26

Any little thing you do to

Mbere 26:27

move

Kenny 26:28

forward, overcome, try something, test something, build it,

Mbere 26:31

gets

Kenny 26:31

you one step closer. Everyone around you

Mbere 26:34

is,

Kenny 26:34

is, is also kind of feeling it. So,

Mbere 26:36

you

Kenny 26:36

you know, find community,

Mbere 26:38

and

Kenny 26:39

build something for yourself. Yeah, we all figuring out, it out right now in, in real time.

Mbere 26:44

Yeah, exactly

Kenny 26:45

yeah, we all figuring out in real time. Well, and Barry, thank you for joining us. if you enjoyed this quick time conversation, make sure you subscribe to the Beyond Normal Podcast and share this with a fellow, another founder or builder. until next time, thank you for tuning in to this great episode of the Beyond Normal Podcast. Peace.

Mbere 27:07

Peace. Bye