Beyond Dashboards: Democratizing Data and Demystifying MOPs with Grant Grigorian

Episode Summary: 

In this episode, hosts Ken Roden and Erin Mills dive into marketing operations and analytics and discuss how generative AI is transforming MOPs. They share personal experiences with AI, such as using ChatGPT to solve tech issues and uploading personas for board meetings. Then, Ken and Erin talk to Grant Grigorian, a marketing operations expert and CEO of Mogi Technologies. Grant discusses the current applications of generative AI in marketing operations and the potential for AI to enhance reporting and decision-making. He emphasizes the importance of data literacy and the role of generative AI in democratizing data. Grant also shares insights on the future of AI in marketing and the challenges and opportunities it presents.

00:28 Real-Life AI Applications in Marketing

03:26 Exploring AI in Marketing Operations

04:05 Interview with Grant Gregorian: AI in Marketing Ops

08:00 Generative AI and Data Democratization

21:12 The Challenge of Data Overload

21:46 Key KPIs for Marketing Operations

23:29 Integrating AI into Marketing Strategies

26:42 Generative AI's Impact on MarTech

32:55 Practical AI Tips and Future Insights

 

Key Takeaways:

• The power of generative AI lies in its ability to explain and verbalize data, making complex data accessible and actionable for marketers.
• The future of marketing operations involves personalized communication and tailored insights based on individual data literacy levels.
• AI can enhance customer journey mapping by providing insights into customer behavior and predicting the best-case customer journey.
• The integration of generative AI into the MarTech landscape will lead to increased content ideation, data analysis, and personalized communication.

 

About our Guest:

Grant Grigorian is the co-founder and CEO of Mogi Technologies, a tool that simplifies marketing data analysis and delivers actionable insights and recommendations to marketing teams. With over a decade of experience in marketing analytics, Grant has a deep understanding of marketing operations and the power of data. Prior to Mogi, Grant worked at Engagio as the Director of Product Management, where he helped define customer journeys and track account stages. He also co-founded Path to Scale, a company focused on multi-touch attribution modeling, which was later acquired by Engagio. Grant is known for his expertise in marketing operations and his ability to demystify complex data analytics.

 

Notable Quotes:

• "Generative AI gives us a chance to create data literacy across our teams." - Grant Grigorian
• "We're sitting on huge amounts of data, but we lack the vocabulary to explain and interpret it effectively." - Grant Grigorian
• "Generative AI can help marketers at all levels of expertise by providing insights and recommendations tailored to their needs." - Grant Grigorian

 

Resources:

• Grant Grigorian on LinkedIn
• Mogi Technologies

To listen to the full episode and stay updated on future episodes, visit the FutureCraft Marketing Podcast. Please subscribe and give us a review if you enjoyed today’s content.

Disclaimer: This podcast is for informational and entertainment purposes only and should not be considered advice. The views and opinions expressed in this podcast are our own and do not represent those of any company or business we currently work for/with or have worked for/with in the past.

Music: Far Away - MK2

Episode Transcript

(00:19):
Thanks for tuning the FutureCraft podcast.
Let's get it started.
Hey there.
Welcome to the Futurecraft Marketing Podcast, where we're exploring how AI is changing all things from brand to demand.
I'm Ken Roden, one of your guides on this exciting new journey.

(00:41):
And I'm Erin Mills, your other co host, and together, we're here to unpack the future of AI and marketing.
We're going to share some insights, test the latest technology, interview industry pioneers, and talk to folks doing really cool things. 11 00:00:57,529.999 --> 00:01:07,900.001 Ken, what really cool thing have you done in AI recently? So this is a real scenario that just happened about, I don't know, 45 minutes ago. 12 00:01:08,230.001 --> 00:01:13,440.001 I used ChatGPT to help me Problem solve a tech issue. 13 00:01:13,440.001 --> 00:01:17,560.001 I was having with one of the applications we use to actually make this podcast. 14 00:01:17,930.001 --> 00:01:24,100.0005 And I was trying to go through the support hub or talk to one of their digital assistants and it just wasn't working. 15 00:01:24,100.0005 --> 00:01:27,579.9995 So I was like, whatever, let me try chat to BT and explain the problem. 16 00:01:27,580.0005 --> 00:01:30,640.0005 And it said, Oh, it may be one of these three things. 17 00:01:30,870.0005 --> 00:01:33,750.0005 And then gave me step by step what I could do to solve each of them. 18 00:01:34,120.0005 --> 00:01:38,760.0005 And the second one worked it saved me so much time and so much frustration. 19 00:01:38,960.0005 --> 00:01:52,389.9995 So do you think that, you would use that instead of going to the support center for other technology is like your starting point or what do you think? I have to laugh because the big joke about millennials is that we don't like talking to people. 20 00:01:52,679.9985 --> 00:01:54,949.9985 And in this example, it was totally true. 21 00:01:55,19.9995 --> 00:01:59,209.9985 I wanted the answer now and the support team wasn't able to get back to me fast enough. 22 00:01:59,209.9995 --> 00:02:02,429.9995 But this was actually something where I could get the answer I needed. 23 00:02:02,779.9995 --> 00:02:07,899.9995 if you start thinking about voice overlay onto these things, I can have a conversation with it. 24 00:02:08,264.9995 --> 00:02:12,324.9995 And actually have it walk me through what I needed to get done or explain what's not working. 25 00:02:12,634.9995 --> 00:02:16,924.9995 I'm actually curious to see how customer support's going to change due to chat GPT. 26 00:02:17,124.9995 --> 00:02:17,624.9995 Totally. 27 00:02:17,824.9995 --> 00:02:20,824.999 What about you? I took your advice from the last episode. 28 00:02:20,824.999 --> 00:02:27,384.999 last time you were talking about how you had uploaded your persona of a stakeholder, an internal stakeholder. 29 00:02:27,614.999 --> 00:02:29,214.999 I had a board meeting. 30 00:02:29,264.999 --> 00:02:32,134.999 So in prepping my slides, I actually wanted to. 31 00:02:32,329.999 --> 00:02:36,729.999 Get a sense of what would the board, really dig into based on what I was presenting. 32 00:02:36,729.999 --> 00:02:55,199.9985 So uploaded the slides and asked a bunch of questions about, what would they care about and what would they dive into? That's really interesting because, especially on a board, it's multiple perspectives, right? how did you think about personalizing it, but also keeping it kind of broad for a board? Yeah, I did a couple of different variations. 33 00:02:55,209.9975 --> 00:03:00,129.9975 I first said a broader scope of a board and the types of folks that are on it. 34 00:03:00,369.9975 --> 00:03:08,219.9975 And then went in and really crafted it to much more about Each person and what sort of I know about those folks that are on the board. 35 00:03:08,609.9975 --> 00:03:21,789.9975 Which really helped to mitigate some of the questions that were coming out know if I have an answer or something that you know Maybe I wouldn't have dug into on the slide It gave me just another data point to bring up during my talk track I really like that. 36 00:03:21,799.9975 --> 00:03:31,199.9975 It's just going onto this ongoing thread that I'm experiencing in kind of my AI journey right now, where I'm not using AI just for content creation. 37 00:03:31,199.9975 --> 00:03:35,159.9975 it's actually helping me prepare for some human interactions. 38 00:03:35,179.9975 --> 00:03:43,79.9975 And, you know, it's not perfect, but it's giving me an idea of what direction to go into, and I feel a little more confident going into some of these conversations. 39 00:03:43,130.0975 --> 00:03:50,510.0975 And also, think of things that maybe you wouldn't have thought of when you're going into those conversations, but now you can just a little bit more prepared. 40 00:03:50,710.0975 --> 00:03:55,170.0975 there's just so many directions that this AI world is taking us. 41 00:03:55,190.0975 --> 00:04:05,510.0965 And I think one of the ones that I'm mostly curious about is how are things like operations and analytics and marketing going to be handled? one of the things that I find really. 42 00:04:05,710.0965 --> 00:04:13,120.0965 Fun and exhilarating about this AI journey we're on is it's demystifying a lot of areas that. 43 00:04:13,120.0965 --> 00:04:17,260.0965 before, like I just couldn't find the answers to, or it was really hard for me to get the answers to. 44 00:04:17,615.0965 --> 00:04:20,725.0965 And one of them is marketing ops and analytics. 45 00:04:20,925.0965 --> 00:04:26,885.0955 I think that generative AI gives us a chance to create data literacy across our teams. 46 00:04:27,365.0965 --> 00:04:42,205.096 There are a lot of folks that struggle with Analytics or what should they be looking at? How should they be interpreting things? And today we have a really special guest, Grant Gregorian, who I've known for many years He is a mops genius. 47 00:04:42,235.096 --> 00:04:46,385.0955 And has helped me to learn marketing operations and get excited about data. 48 00:04:46,715.0945 --> 00:04:53,435.0945 I think the conversation today is going to help demystify some of the marketing operations and analytics, around what's next. 49 00:04:53,435.0945 --> 00:05:20,655.193 And I think that's one thing we're all trying to uncover and really understanding where can we get better insights and how do we interpret them? It's one thing to have all the data, which, we have so many data sources now and so much data, but really, what do you take from that data to action it? I'm particularly interested to talk to him because as you mentioned MOPS is not my area of expertise, but you Engagio and helped build some of their key functionality. 50 00:05:20,880.193 --> 00:05:34,120.1925 And for me, that was one of the first times that I actually could get into reporting because it was so easy to view an account from a full perspective and understand how we were able to engage and warm up that account. 51 00:05:34,380.1935 --> 00:05:43,560.1935 So I'm super curious and excited to see what he's got to say about how AI is going to help us be better marketers and we can dive into that right after this break. 52 00:05:43,760.1935 --> 00:05:47,880.1935 Hey, Grant welcome to the FutureCraft marketing podcast. 53 00:05:47,890.1935 --> 00:05:49,500.1925 We're so excited to have you here. 54 00:05:49,760.1935 --> 00:05:52,270.1935 we're back with FutureCraft Marketing Podcast. 55 00:05:52,660.1935 --> 00:05:57,960.1935 Many of you in the mops world are very familiar with our next guest, Grant Gregorian. 56 00:05:58,210.1935 --> 00:06:03,240.1925 Grant is the epitome of a startup enthusiast and a mops wizard. 57 00:06:03,480.1925 --> 00:06:07,390.1925 He's currently the co founder and CEO of Moji Technologies. 58 00:06:07,800.1925 --> 00:06:17,770.1925 a tool that's simplifying marketing data analysis that automatically delivers insights and recommendations, making complex data accessible and actionable for marketing teams. 59 00:06:17,970.1925 --> 00:06:37,650.1895 Prior to Moji, Grant had really impactful tenures at Engagio, where he was the director of product management and path to scale for many of you early days of attribution, a company that he co founded and later saw acquired by Engagio, where he developed tools to help marketers measure the ROI of their campaigns. 60 00:06:38,65.1895 --> 00:06:40,445.1895 Through multi touch attribution modeling. 61 00:06:40,525.1895 --> 00:06:42,655.1895 Grant, it is so great to see you. 62 00:06:42,985.1895 --> 00:06:45,455.1895 Thanks for coming on the FutureCraft marketing podcast. 63 00:06:45,505.1895 --> 00:06:46,425.1895 We're happy to have you. 64 00:06:46,625.1895 --> 00:06:47,235.1895 I'm excited. 65 00:06:47,435.1895 --> 00:06:47,695.1895 All right. 66 00:06:47,725.1895 --> 00:06:48,715.1895 Let's get into it. 67 00:06:48,955.1895 --> 00:06:58,65.1905 Can you give us an overview of how generative AI is currently being used in marketing operations and what are the primary functions that serve? Yes. 68 00:06:58,285.1905 --> 00:06:58,815.1905 Awesome. 69 00:06:58,865.1905 --> 00:07:01,605.1905 I would say we just can't stop talking about it. 70 00:07:01,855.1905 --> 00:07:05,735.1905 Every conference that you go to that has to do with mops. 71 00:07:06,130.1905 --> 00:07:06,570.0905 It's AI. 72 00:07:06,570.1905 --> 00:07:11,839.9895 And when we meet each other in the hallway, or when we Slack with each other online, it's pretty much all AI. 73 00:07:11,840.0895 --> 00:07:18,180.0895 And it's because I think we're both terrified and excited about the prospect of what's to come. 74 00:07:18,400.0895 --> 00:07:20,80.0905 There's just so many applications. 75 00:07:20,90.0905 --> 00:07:23,510.0905 It's a mind boggling in marketing operations specifically. 76 00:07:23,510.0905 --> 00:07:24,530.0905 And so as. 77 00:07:24,730.0905 --> 00:07:27,770.0905 Stewards of marketing capabilities and organizations. 78 00:07:27,970.0905 --> 00:07:37,610.0905 We are very much attuned to what, what should we doing today to enable our teams to use this incredible technology and not fall behind. 79 00:07:37,810.0905 --> 00:07:42,730.0915 And also not to become obsolete and also to power up our own careers along the way. 80 00:07:42,980.0915 --> 00:07:44,620.0915 And so everyone's paying attention. 81 00:07:45,110.0915 --> 00:07:47,80.0915 I don't think it's overhyped. 82 00:07:47,280.0915 --> 00:07:48,780.0915 Personally don't think it is. 83 00:07:49,30.0915 --> 00:07:55,790.0915 it is really impressive for anyone who's, talk to these chatbots and the pace at which they're getting better is also incredible. 84 00:07:56,90.0915 --> 00:07:59,270.0915 it's a scramble, first of all, there's a vendor arms race. 85 00:07:59,700.0915 --> 00:08:02,150.0915 and then we're all paying attention to what are the vendors doing. 86 00:08:02,650.0915 --> 00:08:18,140.0905 And then internally in our own organizations, we're thinking about how do we harness this? How do we enable it? How do we make it safe? How do we not expose personalized data through some LLM system that, God knows what will happen to it. 87 00:08:18,320.0905 --> 00:08:21,780.0905 So it's all top of mind and we're all discussing it. 88 00:08:22,270.0905 --> 00:08:27,540.0905 I understand that feeling of being very excited about it, but also a little scared. 89 00:08:27,690.0905 --> 00:08:29,580.0905 I think we're all in that same space. 90 00:08:29,640.0905 --> 00:08:37,860.0905 What role do you see generative AI specifically? Playing in enhancing reporting or decision making and marketing operations. 91 00:08:37,860.0905 --> 00:08:53,450.0905 And how can marketers leverage AI better to get those insights? what we're working on at Moji is on data analysis side, The ability for generative AI to explain and verbalize data. 92 00:08:53,700.0905 --> 00:08:55,80.0905 I'm going to go on a quick rant. 93 00:08:55,460.0905 --> 00:09:10,65.091 first of all, I want to distinguish and I'm glad that you're using the phrase generative AI and not just AI because sometimes we get tripped up because as soon as you come to my house, which is Statistics and data there's more than one AI here and it's been around for a while. 94 00:09:10,455.091 --> 00:09:25,935.09 And so when I say AI, I don't necessarily mean generative AI, which is a more recent phenomenon with a breakthrough with open AI and all of those incredible models, I think basically anytime we do any math, that's more than division, we get to call it AI. 95 00:09:26,135.09 --> 00:09:26,455.09 Okay. 96 00:09:26,455.09 --> 00:09:40,195.089 So any statistical analysis, progression, causal analysis, any modeling that you used to do in statistics class, wow, you're an AI expert now, right? Because that's what we, that's the label these days. 97 00:09:40,395.089 --> 00:09:57,695.0905 And one of the challenges that we've always had in marketing analytics, and I've been doing marketing analytics for over a decade, is this last mile problem, which is after we deploy the dashboard, we, imagine deploying the dashboard is like a big, Party All the planets have to align. 98 00:09:57,695.0905 --> 00:09:59,765.0905 The data has to be cleaned up and vetted. 99 00:10:00,95.0905 --> 00:10:01,715.0905 All the systems are hooked up. 100 00:10:02,175.0905 --> 00:10:11,55.0905 The users the marketers are interviewed for requirements and what do they want to see? What kind of charts and how do they want to make decisions? You do all that work. 101 00:10:11,375.0905 --> 00:10:23,820.0905 Finally, there is this dashboard and you're like, look, here's this pie chart, and then guess what happens? I would say six out of 10 times, guess what happens? They say, thank you so much, This is an incredible pie chart. 102 00:10:23,980.0905 --> 00:10:26,220.0895 I can see that you've done a lot of work here. 103 00:10:26,270.0905 --> 00:10:27,900.0895 This is just what we requested. 104 00:10:28,350.0905 --> 00:10:32,450.0905 What do you think is the next step? What should we do? And I'm like, do your job. 105 00:10:32,500.0905 --> 00:10:33,670.0905 I thought you were a grownup. 106 00:10:33,690.0905 --> 00:10:36,820.0895 I thought you were here for work, just do the thing. 107 00:10:36,820.0895 --> 00:10:37,920.0895 And it's not that clear. 108 00:10:37,920.0895 --> 00:10:38,990.0895 It's not that simple. 109 00:10:39,240.0895 --> 00:10:46,100.089 If whenever I've tried to put myself in their shoes and say does this really tell me what to do? It's not that easy. 110 00:10:46,100.089 --> 00:10:52,330.0905 Like you have to drill in, you have to think about the areas, business decisions that are floating around. 111 00:10:52,330.0905 --> 00:10:56,550.0895 And how does this actually dear me for one decision to another? It's not simple. 112 00:10:56,820.0905 --> 00:10:59,910.0905 And everybody is at a different kind of data literacy curve. 113 00:10:59,910.0905 --> 00:11:04,684.9905 And so what I was thinking about is how can we bridge that gap? And especially regenerative AI. 114 00:11:05,125.0905 --> 00:11:20,145.0905 There's this ability to verbalize and talk about the data that the computer becomes really good at, and it can come down to your level or whatever the level you need to be at in order to articulate the point that you're trying to make with the data set. 115 00:11:20,235.0895 --> 00:11:21,345.0905 And it's really good at that. 116 00:11:21,505.0905 --> 00:11:25,205.0905 And so that's what we're trying to do in terms of reporting and analytics. 117 00:11:25,205.0905 --> 00:11:28,35.0905 So we're really using the generative AI portion of it. 118 00:11:28,235.0905 --> 00:11:37,405.0905 And less on the deep kind of causal analysis and the modeling and the predictive analytics part of it, which is also valid, but still needs even more explanation. 119 00:11:37,465.0905 --> 00:11:44,155.0905 And so it's this explanation power that we're hoping to harness and really democratize data as a result. 120 00:11:44,405.0905 --> 00:11:46,385.0905 I love the idea of democratizing data. 121 00:11:46,705.0905 --> 00:12:14,95.2915 It was really good, that we're talking about generative AI versus kind of other investments in AI or innovation, just generally speaking, are there recent innovations in AI that you've seen generally not specifically gen AI that you think are particularly interesting for people who are in mops or care about, Marketing operations and being more efficient so that's the thing that's been disorienting over the last few years is just a chain, the pace of change. 122 00:12:14,295.2915 --> 00:12:20,915.2915 and also just as you plot the changes that have already been made into the future, I just, I'm having a really tough time. 123 00:12:20,925.2915 --> 00:12:25,145.2915 Like I used to be able to squint and imagine what the future will be like. 124 00:12:25,585.2915 --> 00:12:41,895.2915 And I felt I have a handle on this, and I can develop products and I can imagine future products that will help marketers and I'm having a tougher time doing it because of how Weird the world is if you start to think about the capabilities that are possible. 125 00:12:42,95.2915 --> 00:12:44,135.2915 So let me give you an example. 126 00:12:44,185.2915 --> 00:12:46,345.2915 We want to make data more accessible. 127 00:12:46,605.2915 --> 00:12:55,565.292 first thing that we did is we started going to the data sources, using API, collecting the data, doing some analysis and then sending the results to users. 128 00:12:55,865.292 --> 00:12:59,445.392 a lot of the customer feedback was we already have a bunch of dashboards. 129 00:12:59,465.392 --> 00:13:01,515.392 We already built all of the capabilities. 130 00:13:01,515.392 --> 00:13:02,775.392 I don't want you to recreate it. 131 00:13:02,975.392 --> 00:13:07,455.392 Can you use this instead? okay, I'll just go to your dashboard instead. 132 00:13:07,505.392 --> 00:13:12,115.392 I'll try to articulate what it all means and how to interpret it and what to do about it. 133 00:13:12,365.392 --> 00:13:22,325.392 one of the capabilities that I stumbled upon is you could literally take a screenshot of the dashboard and send it to one of these AI systems. 134 00:13:22,750.392 --> 00:13:24,650.392 And that, you can do a little bit more work than that. 135 00:13:24,660.392 --> 00:13:42,700.392 You can take each component and separate it out, send it to these new systems and they do a tremendous amount of job at summarizing it, finding highlights things to pay attention to and things that I would have done as a human being, but they do it so much better and consistently. 136 00:13:42,900.392 --> 00:13:44,580.392 And it's just taking a picture of it. 137 00:13:44,600.392 --> 00:13:47,120.392 It did, it's not even, doing anything fancy. 138 00:13:47,120.392 --> 00:13:52,980.391 And so this ability to just replicate what we do in a whole new way is mind boggling. 139 00:13:53,30.392 --> 00:13:54,590.392 I don't want to look at dashboards anymore. 140 00:13:54,650.391 --> 00:14:01,140.392 You look at the dashboard, tell me the highlights and then tell them, and I'll dive in and figure out why something's not happening. 141 00:14:01,150.392 --> 00:14:03,10.392 And I'll be tier two analyst. 142 00:14:03,210.392 --> 00:14:07,330.392 You take care of tier one problems and I'll jump in only when I need to. 143 00:14:07,730.392 --> 00:14:08,420.391 Let's make that. 144 00:14:08,620.391 --> 00:14:11,430.3905 And that's how approached it is with every new capability. 145 00:14:11,430.3905 --> 00:14:23,800.3905 I think about what does this unlock for us? How can I help, more teams leverage this within, within the mission statement that we have, Enable marketers to better understand what's happening. 146 00:14:24,0.3905 --> 00:14:26,120.3915 I haven't tried that yet with the dashboard. 147 00:14:26,120.3915 --> 00:14:27,640.3915 So I'm definitely taking that away. 148 00:14:27,980.3915 --> 00:14:42,150.3915 one of the things I think that's interesting that you pointed out is this level one, analysis and created a bit more sophistication, I think, with marketers, in terms of data management, we get these beautiful dashboards, we maybe have some insights that we pull out. 149 00:14:42,310.3925 --> 00:14:45,790.3925 But junk data in, it's not really going to matter. 150 00:14:46,170.3925 --> 00:15:00,840.3915 Do you see a world where, there's generative AI supporting, integrating, different data sources and cleaning data and just getting better results overall? Or what's your, what are your thoughts on that? I haven't seen that in real life yet. 151 00:15:01,10.3915 --> 00:15:08,20.2925 Partially because we haven't come to a consensus as a society yet as where exactly these generative AI. 152 00:15:08,500.3925 --> 00:15:17,480.3925 Systems are going to live and it probably will vary from organization to organization, but we're very, I think, correctly guarded about our corporate data sets. 153 00:15:17,910.3925 --> 00:15:24,180.3915 And we are not, going to let, Alexa and Siri come in and just do whatever they want with our data. 154 00:15:24,630.3915 --> 00:15:36,850.3905 so the plumbing of the data sets, the integrating, the working of the data is very much still the work of data engineering teams and, the vendors That host our data. 155 00:15:37,50.3905 --> 00:15:43,140.39 And then it's we then selectively expose it to generative AI or explanation. 156 00:15:43,140.39 --> 00:15:46,830.3905 Like for example, recently have low, which is part of Salesforce. 157 00:15:47,260.3905 --> 00:15:49,630.3895 They came out with a feature and they've been. 158 00:15:49,815.3905 --> 00:16:01,915.3895 Really in a forefront of using journey of AI and interpreting data as of course they would be of this notion of there's a dashboard, but there's this little sidebar and you can chat with your dashboard. 159 00:16:02,115.3895 --> 00:16:04,485.3885 you can ask a question and it will tell you answers. 160 00:16:04,925.3885 --> 00:16:10,395.3885 and it's at that level Especially with one of the areas that I've been thinking about is in marketing. 161 00:16:10,805.3885 --> 00:16:12,505.3895 it's okay to be a little bit wrong. 162 00:16:12,755.3895 --> 00:16:15,395.3895 And so it's a very probabilistic kind of a science. 163 00:16:15,785.3895 --> 00:16:20,605.3885 A lot of times you just want to paint a broad picture that's pointing in the right direction. 164 00:16:21,5.3895 --> 00:16:24,385.3885 that invites a lot of what's called synthetic data. 165 00:16:24,465.3885 --> 00:16:32,425.3895 You could pad the data the kind of polish the outliers and say, we only have a few signals. 166 00:16:32,625.3895 --> 00:16:37,835.3895 Of real data, but it's probably like this in order to tell a story. 167 00:16:38,65.3895 --> 00:16:38,765.3895 That's true. 168 00:16:38,965.3895 --> 00:16:39,955.3895 I'm okay with that. 169 00:16:40,125.3895 --> 00:16:45,895.3895 And I think we these generative AI systems actually really good at that and generating data that kind of looks like other data. 170 00:16:46,95.3895 --> 00:16:46,965.3895 And if you think about. 171 00:16:47,165.3895 --> 00:16:57,905.3905 Companies that we work with, which is B2B companies, like when they have so many examples of successful opportunities, We just have a few a year that are big enough to pay for the whole thing. 172 00:16:58,335.3905 --> 00:17:06,288.6905 And so in those situations I think it could be used to smooth out some of those curves to say, yep, this is what we want. 173 00:17:06,498.6905 --> 00:17:11,298.6905 this is what it looks like in order to help it better understand, to make connections between the data. 174 00:17:11,458.6905 --> 00:17:14,603.6905 It's the art and science, like being a little bit more integrated. 175 00:17:14,803.6905 --> 00:17:22,23.6905 Yes, because if I can explain it and get the value in being able to, someone have that eureka moment or oh, I get it, and then have them. 176 00:17:22,223.6905 --> 00:17:28,623.6905 Age, what they're doing as a result that, Increases the performance of some campaign that would be worthwhile. 177 00:17:28,823.6905 --> 00:17:39,683.69 100 percent and I think that's one of the bigger challenges because you do have so many varying levels of folks of, understanding data and what it means and what's the next step. 178 00:17:39,783.69 --> 00:17:41,643.69 as you are working to simplify some of this. 179 00:17:41,843.69 --> 00:17:49,113.69 Marketing data analysis what are some of the strategies that folks have been using and where are you seeing that be really effective? Yes. 180 00:17:49,143.69 --> 00:17:58,583.69 So that is an area where generative AI is I think going to have a big impact because the example I like to think of is a Khan Academy. 181 00:17:59,73.69 --> 00:18:01,943.69 If you guys know Khan Academy, fantastic resource. 182 00:18:02,213.69 --> 00:18:03,963.69 I use it with my kids all the time. 183 00:18:04,163.69 --> 00:18:12,623.6895 But Khan Academy has this vision of a personalized tutor, a tutor that can, that knows where you are in any specific subject. 184 00:18:12,623.6895 --> 00:18:25,553.69 And if you think about, education, that's been the Holy grail of being able to have personalized educator who knows what you know, and has the context of all the other path conversations that it add with you and any gaps you may have in some subjects. 185 00:18:26,3.69 --> 00:18:27,203.69 This is where I'm squinting. 186 00:18:27,293.69 --> 00:18:30,443.69 In the future, you don't all get the same dashboard. 187 00:18:30,443.69 --> 00:18:35,673.689 The source of the data is the same, but you get what you get with what you want. 188 00:18:35,873.689 --> 00:18:42,343.689 And so I think that the communication density changes depending on the role and the context. 189 00:18:42,543.689 --> 00:18:47,703.689 So some people may want a 10 slides because they're going to go present to the board. 190 00:18:48,13.689 --> 00:18:51,478.689 Another person just wants, you A quick email with bullet points. 191 00:18:51,488.689 --> 00:18:53,148.689 Another person wants. 192 00:18:53,613.689 --> 00:18:58,863.689 Automatically generated podcast they can listen to on the way to work with the same content. 193 00:18:59,163.689 --> 00:19:08,893.69 Why not? And so this malleability of the form factor and malleability of the content makes it perfect for personalizing this communication. 194 00:19:08,943.69 --> 00:19:19,803.69 that means that we can track who's at what level of data literacy or data comfort and tailor the message accordingly, make it more verbose or less verbose. 195 00:19:19,803.69 --> 00:19:20,673.69 Just get to the point. 196 00:19:20,733.689 --> 00:19:21,473.589 I know what 0. 197 00:19:21,473.689 --> 00:19:22,423.689 2 coefficient means. 198 00:19:22,423.689 --> 00:19:23,783.688 Just tell me what the number is. 199 00:19:24,143.688 --> 00:19:25,953.7385 And other people would say, can you remind me It's 0. 200 00:19:25,953.7385 --> 00:19:26,643.6885 2 good. 201 00:19:26,643.6885 --> 00:19:31,773.6885 Is it bad? What is it supposed to mean? And that's, and without judgment, because it's a robot, you can go deeper. 202 00:19:32,203.6885 --> 00:19:33,913.6885 And I am excited about that. 203 00:19:34,113.6885 --> 00:19:39,23.5875 Okay, Grant, Erin and I are both huge Engagio stans. 204 00:19:39,23.6875 --> 00:19:41,923.6885 Like we both have had a lot of success using it. 205 00:19:42,53.6875 --> 00:19:44,273.6885 So I want to dive into some of your experience there. 206 00:19:44,513.6885 --> 00:19:58,348.6885 You worked on defining customer journeys and tracking account stages, right? How do you see AI enhancing that process of customer journey mapping and what benefits could it bring or insights could it bring around customer behavior? Yes. 207 00:19:58,348.6885 --> 00:19:58,858.6885 Thank you. 208 00:19:59,38.6885 --> 00:20:03,568.6885 And you know what I miss about Engage because it was sold in Incorporated is the whale. 209 00:20:03,668.6885 --> 00:20:04,88.6885 love the way. 210 00:20:04,138.6885 --> 00:20:07,948.6885 I think his name was Geo and it was one of my favorite parts of Engagio. 211 00:20:07,978.6885 --> 00:20:08,128.6885 Yeah. 212 00:20:08,328.6885 --> 00:20:11,628.6885 So here, I think this is gonna be a combination of the. 213 00:20:11,828.6885 --> 00:20:22,448.6875 The generative AI and non generative AI and the non generative AI parts are going to play a deeper role here. 214 00:20:22,468.6885 --> 00:20:26,988.688 And they already do both, demand based and Sixth Sense and companies like that. 215 00:20:26,988.688 --> 00:20:37,408.6885 They've been around for years and for years had these predictive models that try to take different factors of success for a customer journey and weight them correctly. 216 00:20:37,408.6885 --> 00:20:39,348.6885 And try to estimate what it all means. 217 00:20:39,728.6885 --> 00:20:43,868.6885 And therefore what a prospective best case customer looks like. 218 00:20:43,878.6885 --> 00:20:45,138.6885 Those are models that are. 219 00:20:45,338.6885 --> 00:20:47,158.6885 We can call them AI, that's AI. 220 00:20:47,388.6885 --> 00:20:49,358.6885 It's not generative AI. 221 00:20:49,818.6885 --> 00:20:50,858.6885 But we use that today. 222 00:20:50,868.6885 --> 00:20:57,828.6885 Those vendors are actively using AI and we are actively as customers using AI to tell those stories. 223 00:20:58,118.6885 --> 00:21:03,408.6875 Accounts go up and down and sometimes our customers, sometimes they're not, sometimes they're in pipeline, sometimes they're not. 224 00:21:03,828.6875 --> 00:21:07,498.6885 And it's a, knowing that on a dashboard is impossible. 225 00:21:07,868.6895 --> 00:21:15,298.6895 And so I am looking forward to being able to automatically, it's show me a movie of how my pipeline has been changing. 226 00:21:15,498.6895 --> 00:21:15,958.6895 Okay. 227 00:21:15,978.6895 --> 00:21:17,958.6895 So It's something that happens over time. 228 00:21:18,68.6895 --> 00:21:19,938.6895 So there should be like a little movie. 229 00:21:19,958.6895 --> 00:21:37,728.6885 Why is it always a dashboard? And so those things will be possible and the ability to explain the data would be possible because that's the thing that's been so frustrating is that there is all this like modeling and all this data and we're just not comprehending it correctly. 230 00:21:37,948.6885 --> 00:21:41,403.5885 How do you explain it? And then tell it to me across multiple channels. 231 00:21:41,683.6885 --> 00:21:44,483.6885 Deals and then give me something that they have in common. 232 00:21:44,693.6885 --> 00:21:47,653.6885 There's no way that you can do that easily. 233 00:21:48,63.6885 --> 00:21:49,353.6895 we have the data for it. 234 00:21:49,603.6895 --> 00:21:52,393.6895 we almost lack the vocabulary to explain. 235 00:21:52,678.6895 --> 00:22:03,178.69 it's almost like there's an abyss between so much data, which I think is almost overwhelming how much access we have to data versus, years ago it was I have this lead and they filled out the form. 236 00:22:03,188.689 --> 00:22:03,518.689 Great. 237 00:22:03,528.69 --> 00:22:07,338.692 Now, all the, research that folks are doing before they're even coming to you. 238 00:22:07,553.692 --> 00:22:14,973.693 How did I digest that? I'm curious, once folks solved some of the data cleanliness problems and, have some of these tools in place. 239 00:22:15,433.693 --> 00:22:44,843.694 What would you say are some KPIs that would be important to show, from a marketing operations perspective to the business? If we're talking about in terms of generative AI, then I would say To me, the metrics that are important to the two that come, that we've been thinking about at Moji too, is what is engagement? I would want my team to be engaged with the systems and to use them and to be clicking on stuff and conversating with them. 240 00:22:45,223.694 --> 00:22:46,673.694 And so I would be thinking about that. 241 00:22:46,998.694 --> 00:22:50,323.694 Because if nobody's using it, it means something's not working. 242 00:22:50,353.694 --> 00:22:51,553.694 They don't find it useful. 243 00:22:51,553.694 --> 00:23:04,933.692 And so we need to do some additional work to make it useful so that they get embedded in the day to day work, because we know that they have capabilities that will improve the team's efficiency, that all the things, right? So I want the team using it. 244 00:23:05,303.692 --> 00:23:10,683.693 And I want metrics that show that, it's up and to the right in terms of usage. 245 00:23:11,43.693 --> 00:23:30,613.694 The thing that it's been more elusive is is it any good? Is it like if it gives you insights or advice or tells you to change something, was that like demonstrably good for the business in terms of increasing performance in some way? That would be amazing if we can get there. 246 00:23:30,953.694 --> 00:23:38,663.693 And that would be just KPI like anything else that we would track an ROI on investment on a system like this. 247 00:23:38,743.694 --> 00:23:58,708.694 Does it bend the curve somehow and make better decisions for us? Or maybe because of that increased usage, we're just interacting with the data more and we're just faster to create things and iterate and so it, does it just grease the wheels of an existing team already? And so those are the things that I'm thinking about. 248 00:23:58,958.694 --> 00:24:21,453.694 What advice would you give to CMOs or MOPS leaders that are looking to integrate, whether it's generative AI or AI, into their strategies? And are there any real skills or knowledge areas that they should focus on? Yes, I would say these days, the thing to do is to Establish a safe space where your employees to play with this stuff. 249 00:24:21,703.694 --> 00:24:32,623.694 I would encourage CMOs and other Data leaders to create a place where you explicitly give permission for your employees to play with generative AI. 250 00:24:33,13.694 --> 00:24:45,863.694 So it means they have some, maybe some internal data and the model in the same room where you can start to mesh them together, or at least a place where they can copy paste things from, they go to Salesforce. 251 00:24:46,173.694 --> 00:24:53,123.694 And so that is very scary for many organizations because a, we're doing something that we don't know about. 252 00:24:53,153.695 --> 00:24:59,363.695 We don't know exactly what the outcome will be or what the capabilities that it might unlock, and we're playing, we're exploring together. 253 00:24:59,563.695 --> 00:25:06,453.695 And so the skill there is like creativity, experimentation, culture that I would encourage. 254 00:25:06,903.695 --> 00:25:18,563.695 Having done that, I would check back in and say what does this unlock for us? What capabilities did you discover and try to then maybe find a more dedicated way to double down on that capability. 255 00:25:18,593.695 --> 00:25:25,543.6955 if the team comes back and says, Hey, I'm finding myself, pasting an email from this vendor and summarizing it into the slides. 256 00:25:25,553.6955 --> 00:25:31,893.6955 Can we find a way to do that? I would leave it up to your team to experiment and come up with use cases on their own. 257 00:25:32,93.6955 --> 00:25:38,703.6955 Yeah, I think that's really interesting for the reason that I think people are still nervous on where to get started. 258 00:25:38,943.6965 --> 00:25:40,513.6955 If we could go a little bit deeper in that. 259 00:25:40,803.6955 --> 00:25:42,233.5955 Could you maybe give us like a. 260 00:25:42,693.6955 --> 00:25:53,183.6955 Good, better, best of what mops should be thinking about right now, considering what's out there for AI, whether it's generative or another category. 261 00:25:53,383.6955 --> 00:25:53,733.6955 Yeah. 262 00:25:53,783.6965 --> 00:25:57,273.6965 The worst is to just ignore it and hope that it goes away. 263 00:25:57,643.6955 --> 00:25:59,563.6955 I don't think this one is going away. 264 00:25:59,663.6965 --> 00:26:01,923.6955 This one is everyone else is excited about it. 265 00:26:01,983.6955 --> 00:26:03,263.6955 You're going to be forced to use it. 266 00:26:03,463.6955 --> 00:26:07,403.6955 Good as if you're doing it on your own at home and you're playing with it. 267 00:26:07,403.6955 --> 00:26:11,173.6955 And you're like, Oh gosh, like this could be really powerful for me at work. 268 00:26:11,563.6955 --> 00:26:18,513.6975 But you are yourself upscaling, paying attention and getting a sense of the capabilities of the models. 269 00:26:18,523.6975 --> 00:26:22,823.698 So I would say step one is get a sense of the capabilities of the models. 270 00:26:22,963.698 --> 00:26:27,653.698 Play with them, talk to them about work, ask them questions that you know, the answers to. 271 00:26:27,993.698 --> 00:26:30,133.697 That's always the best one, and see if they'll. 272 00:26:30,333.697 --> 00:26:35,503.697 Surprised and delight you because they have surprised and delighted me a lot but also disappointed. 273 00:26:35,743.697 --> 00:26:45,703.697 and level three is, can you actually use it at work? And do you have a safe space where it's being used at work? Where the lines start to get blurred and you get to really. 274 00:26:45,903.697 --> 00:26:50,273.697 Experience the future in the context of all your coworkers and have discussions about it. 275 00:26:50,373.697 --> 00:26:51,943.697 That's actually what I did in a previous role. 276 00:26:51,943.697 --> 00:26:53,183.697 I was leading a marketing function. 277 00:26:53,183.697 --> 00:26:59,633.697 I was like, guys, just go find stuff and yes, don't use any proprietary information, but play with it. 278 00:26:59,873.697 --> 00:27:03,453.697 What do you like? What works? What could fit? It's not going to be perfect. 279 00:27:03,553.697 --> 00:27:09,443.697 What could we leverage? And how could it help us? And I just think that curiosity is really important right now. 280 00:27:09,443.697 --> 00:27:10,673.697 And it is a little scary. 281 00:27:10,973.697 --> 00:27:12,983.598 how do you think generative AI. 282 00:27:13,353.698 --> 00:27:16,83.698 Is going to change the MarTech landscape. 283 00:27:16,123.698 --> 00:27:19,493.6975 Where do you think it's going? Yeah, it's going to destroy it completely. 284 00:27:19,493.6975 --> 00:27:21,213.698 It's going to be so confusing. 285 00:27:21,443.697 --> 00:27:22,333.698 I have no idea. 286 00:27:22,463.698 --> 00:27:26,153.6975 Already had that monster ocean giant tech. 287 00:27:26,183.6985 --> 00:27:28,233.6985 What is that called? The MarTech landscape. 288 00:27:28,533.6985 --> 00:27:28,853.6985 Oh yeah. 289 00:27:28,853.7985 --> 00:27:31,343.7985 The thousands and thousands of vendors. 290 00:27:31,543.7985 --> 00:27:34,563.7985 We're going to just explode that even further. 291 00:27:34,763.7985 --> 00:27:37,813.7975 and so I, again, this goes back to this. 292 00:27:37,843.7985 --> 00:27:41,673.7985 You're making me squint even harder by asking the futuristic question. 293 00:27:41,873.7985 --> 00:27:47,683.7985 Try to play this game where I say, okay let's say that we have the generative AI. 294 00:27:47,713.7985 --> 00:27:55,93.7975 It has improved to the point where we trust it with data analysis, but let's say that we've got GPT five or whatever. 295 00:27:55,93.7985 --> 00:27:57,913.7975 And I do imagine these models getting better and better. 296 00:27:58,383.7975 --> 00:28:00,523.7975 And so it's able to do data analysis. 297 00:28:00,523.7985 --> 00:28:03,243.7985 It's able to do basically what we do. 298 00:28:03,443.7985 --> 00:28:07,53.7995 What's it like at work as a marketer, so what am I, okay. 299 00:28:07,53.7995 --> 00:28:10,563.7995 So if I open my computer, what does it say? Is it like. 300 00:28:10,763.7995 --> 00:28:14,503.7995 Hey marketer, here's all the stuff I've been doing while you've been sleeping. 301 00:28:14,973.7995 --> 00:28:35,603.7975 Here's, I guess it will have to tell us how it's going at least here's the leads that came in that I responded to that is the lead also a computer that is inquiring, because I said, Hey, I'm buying the software, go get me bids and go evaluate technology, and so it's just so hard to imagine for me what it's like, because we've just got these robots running around. 302 00:28:36,83.7975 --> 00:28:47,463.796 And we're just hitting pause on Netflix, just in time to listen to what they're doing and then going back to watching, what am I doing exactly in this world? cause there'll be like I want to launch a campaign. 303 00:28:47,663.796 --> 00:28:48,33.7955 Okay. 304 00:28:48,33.7955 --> 00:28:55,363.797 We'll and so you just give it thematic ideas and then it goes and finds both the customers that, prospects that it needs to contact. 305 00:28:55,363.797 --> 00:28:58,303.797 We will send them an email, but those people don't want to read emails. 306 00:28:58,333.797 --> 00:29:01,523.797 They'll just want a daily summary of all emails that came in. 307 00:29:01,713.797 --> 00:29:10,993.796 You know what I mean? Read to them in a whispery voice on the end of the day, who knows? I have no idea, but they can do whatever they want with the content that's coming in. 308 00:29:11,368.797 --> 00:29:16,528.798 what does it look like, whether it's the end of the year, or even like a couple of weeks from now or months from now. 309 00:29:16,568.798 --> 00:29:18,888.798 we'll start to see it seep into data. 310 00:29:19,188.798 --> 00:29:29,618.797 We're sure we'll see, all the vendors, Salesforce, Tableau, Looker, all the big vendors incorporating data summaries, data visualizations. 311 00:29:30,18.797 --> 00:29:36,638.797 It's inevitable that robots will inspect, analyze, and summarize data for us. 312 00:29:36,658.797 --> 00:29:38,8.797 It's just way too much of it. 313 00:29:38,288.797 --> 00:29:40,388.797 And so they'll need a way to explain it to us. 314 00:29:40,588.797 --> 00:29:53,148.797 So that's for sure coming and that means personalized communication through our marketing automation systems through our ads and then automated summaries of performance. 315 00:29:53,348.797 --> 00:29:59,728.797 the thing that I'm looking forward to is the ability to really embed these systems into our data. 316 00:29:59,978.797 --> 00:30:08,257.1625 And in a deep way, not just in a copy paste kind of a model or even chat model, but where the model is sitting on top of a. 317 00:30:08,577.2625 --> 00:30:16,767.2625 Multi relational database, and they can see, transactions over here, leads over here, interactions over here, and accounts over here and make sense of it. 318 00:30:16,957.2625 --> 00:30:18,17.2625 That would be incredible. 319 00:30:18,57.2625 --> 00:30:19,207.2625 We don't have that yet. 320 00:30:19,517.2625 --> 00:30:23,87.262 And that's a near term possibility, I think. 321 00:30:23,87.262 --> 00:30:24,167.2615 And I'm looking forward to it. 322 00:30:24,367.2615 --> 00:30:26,817.2615 you mentioned Looker and you mentioned other technologies. 323 00:30:27,77.2615 --> 00:30:38,582.2115 What are your thoughts on that? One or two considerations that marketing leaders or people in mops should consider when looking at integrating AI into their existing tech stack. 324 00:30:38,782.2115 --> 00:30:39,632.2115 It's a good question. 325 00:30:39,662.2115 --> 00:30:44,802.2105 I would say these days, what's really good is meeting transcription. 326 00:30:45,2.2105 --> 00:30:45,792.2115 That's fantastic. 327 00:30:45,832.2115 --> 00:30:47,532.2115 Like transcribing text. 328 00:30:47,652.2115 --> 00:30:54,272.2115 I think that could unlock a lot for sales leadership and maybe that's more rev ops and not mops. 329 00:30:54,742.2105 --> 00:31:01,422.2105 Those transcripts have so many insights and good information for marketing to use. 330 00:31:01,602.2105 --> 00:31:06,822.211 I think it's a underused area of those meeting transcriptions there. 331 00:31:06,832.211 --> 00:31:29,32.212 I think they're currently being used to coach the sales rep or at best, do you at least keep track of all the conversations and where they are, but it could be such a treasure trove of common objections of tone of language and vocabulary words used by our customers for content creation for any marketing assets to resonate. 332 00:31:29,162.212 --> 00:31:37,982.212 So the, I think some of the gold being mined today from data that was unavailable previously, isn't those all transcripts. 333 00:31:37,982.212 --> 00:31:43,362.211 I would encourage folks to look into that and ask your sales labs, friend. 334 00:31:43,687.212 --> 00:31:53,217.212 Where are those call transcripts? Can we get them? Can we take a look at them? Cause I think they might have some really interesting things for marketing team to explore and that's available today. 335 00:31:53,687.212 --> 00:31:56,657.212 Those systems are already deployed and out and about. 336 00:31:56,857.212 --> 00:32:03,567.113 being able to take that data driven approach of taking what customers are saying and incorporating it into, marketing. 337 00:32:03,617.113 --> 00:32:16,877.113 what are some of the other things that you're thinking about that aren't, maybe top of mind for folks that, obviously the, there's the integration from our email systems and things of that, but I'd love to get your perspective of what are, what should we be thinking about? I can tell you what I'm thinking about. 338 00:32:16,897.113 --> 00:32:28,797.113 So I've been thinking about on the data side of if you live in a world where analysis and insight is free and abundant all of a sudden, something that used to be so expensive can be generated in a moment's notice. 339 00:32:28,997.113 --> 00:32:38,747.113 So what are we going to do? We have to think about who should get what, when, right? Which of our team members needs what information and what format when. 340 00:32:39,147.113 --> 00:32:48,467.113 what is the cadence of it, do you want to receive, a little insight every day, or do you want to receive a big digest at the end of the week? So it's unlocked all these new questions. 341 00:32:48,517.113 --> 00:32:51,457.113 And those are the questions that we're asking ourselves. 342 00:32:51,467.113 --> 00:33:00,807.1135 It's how do we enable this? And then how do we give the user the ability to tune it? It's how you said it to send it to me weekly, but actually want it every day. 343 00:33:01,7.1135 --> 00:33:03,397.1125 And I want it in this format and I want it like this. 344 00:33:03,717.1135 --> 00:33:11,227.1135 And can you also give me this other thing? as a result of generative AI, some of the new questions that we've been thinking about recently. 345 00:33:11,427.1135 --> 00:33:13,457.1135 it's really neat because it evens the playing field. 346 00:33:13,457.1135 --> 00:33:23,897.1145 I think, some of us who are real data nerds and then folks that are maybe that's not their strength that if they do other things really well, it really helps Okay, Grant, this has been an awesome conversation. 347 00:33:23,957.1145 --> 00:33:26,287.1145 I've learned a lot here at FutureCraft marketing. 348 00:33:26,327.1145 --> 00:33:30,827.1145 We're all about giving our listeners kind of practical tricks and advice. 349 00:33:31,47.1145 --> 00:33:35,507.1165 So I'm going to give you four questions and just give me your quick hits for our listeners. 350 00:33:35,917.1165 --> 00:33:39,737.1165 First of all, what's your best quick AI tip? Yes. 351 00:33:39,837.1165 --> 00:33:43,477.1165 I would say quick, best day to talk to it. 352 00:33:43,677.1165 --> 00:33:47,267.1165 Try the talking capability on the open AI app. 353 00:33:47,697.1165 --> 00:33:51,87.1165 It is so much better than Alexa and my kids use it. 354 00:33:51,287.1165 --> 00:33:55,97.1165 It was asking me questions and I'm like, you know what it's really good at is just making stuff up. 355 00:33:55,97.1165 --> 00:34:01,597.1165 It's a make belief machine, and it just like unlocked so much for her because she's also a make belief machine. 356 00:34:01,797.1165 --> 00:34:04,127.1165 she was like, it, I ghost real. 357 00:34:04,127.1165 --> 00:34:06,877.1165 And it gives a boring answers of no ghost, sir, not real. 358 00:34:07,97.1165 --> 00:34:10,637.1165 Tell it to tell, create a ghost story with it. 359 00:34:10,687.1165 --> 00:34:12,852.1165 Can you create a ghost story about this and it does. 360 00:34:13,62.1165 --> 00:34:13,692.1165 And then it can. 361 00:34:14,7.1165 --> 00:34:17,807.1165 And then it's can you make it like this? And it just keeps on, and then she's just sitting there talking to it. 362 00:34:17,967.1165 --> 00:34:19,87.1165 And I felt good about it. 363 00:34:19,97.1165 --> 00:34:22,887.1165 Cause it's not a screen time, but it's like a creative ideation. 364 00:34:23,247.1155 --> 00:34:26,837.1165 And my, my quick tip is I talking to it too. 365 00:34:27,107.1155 --> 00:34:28,287.1155 Now don't just chat with it. 366 00:34:28,627.1155 --> 00:34:29,977.1165 Yeah, that's great. 367 00:34:30,37.1155 --> 00:34:38,767.2155 What's your best prompt or workflow that you recommend? Go long, that's a, you're tempted to just do short, like texting. 368 00:34:38,817.2155 --> 00:34:40,87.2155 Hey, tell me this, tell me that. 369 00:34:40,327.2155 --> 00:34:44,117.2145 We found that the best prompts are actually multi pages long. 370 00:34:44,597.2155 --> 00:34:46,917.216 These things can ingest many tokens. 371 00:34:47,157.216 --> 00:34:52,557.216 And so what we do is we craft a long essay, essentially with many sections. 372 00:34:52,757.216 --> 00:34:54,807.216 And then you just put the whole thing in. 373 00:34:55,77.216 --> 00:35:05,167.214 It feels overwhelming, but it reads it right away and it starts giving you much better responses than if you were just constantly kind of short form texting with it. 374 00:35:05,397.215 --> 00:35:12,757.215 What's your best tip for keeping up with all of these latest AI trends and marketing? Use it, man. 375 00:35:12,907.215 --> 00:35:15,567.215 You got to get in there, get your hands dirty. 376 00:35:15,877.215 --> 00:35:18,777.215 I'll tell you for me personally, it's been amazing. 377 00:35:18,817.215 --> 00:35:21,517.215 I've always wanted to be a more technical person. 378 00:35:21,667.215 --> 00:35:28,277.214 Like I took coding classes in college and I've always been very impatient with all the syntax. 379 00:35:28,387.215 --> 00:35:32,927.215 And now it's been unlocked because I just tell it exactly what I want. 380 00:35:32,937.215 --> 00:35:34,167.215 And I say, can you make a. 381 00:35:34,367.215 --> 00:35:40,367.215 Python script that does this to this table and takes this and, and it does, and I just paste it in and it works. 382 00:35:40,677.215 --> 00:35:43,107.215 And I'm like, I'm a coder now, this is incredible. 383 00:35:43,547.214 --> 00:35:51,887.214 And so I would encourage you to whatever your career goals are to try to Use it as a coach to coach you along to level up in that way. 384 00:35:52,157.214 --> 00:36:05,857.214 Or if you want to be technical, it is incredible for programmers, for coders it truly transformed how I work from that point of view, because suddenly I'm coding APIs and I'm getting Data results back and I have headphones on and dark screen. 385 00:36:05,857.214 --> 00:36:10,607.214 And I'm one of those people now, all of a sudden, instead of being just a business user. 386 00:36:10,887.214 --> 00:36:11,237.214 Yeah. 387 00:36:11,437.214 --> 00:36:18,327.213 what technology should our listeners check out that they may not know about? So this goes back to the original question of you got to be curious and exploring. 388 00:36:18,327.213 --> 00:36:19,987.213 There's so many coming out all the time. 389 00:36:20,177.213 --> 00:36:25,777.214 The ones that I've been playing with I mentioned earlier is the image recognition capabilities. 390 00:36:25,997.213 --> 00:36:26,967.214 I would try that. 391 00:36:27,147.214 --> 00:36:39,767.212 I would try sending it things other than text to play with it because it's becoming surprisingly good at it and recognizing, recognizing pie charts and what they mean and what they say. 392 00:36:40,187.212 --> 00:36:42,577.214 And lay with the imagery, with the sound. 393 00:36:42,967.214 --> 00:36:46,107.214 Don't just limit yourself to text when it comes to the system. 394 00:36:46,167.214 --> 00:36:53,117.311 I think that's an area of active research and the systems are getting better and better at it every week, it seems this has been awesome. 395 00:36:53,467.311 --> 00:37:02,337.312 I have learned so much about the democratization of data and how AI can maybe help us get more clarity into what we're doing and I think our listeners have learned a lot as well. 396 00:37:02,337.312 --> 00:37:14,797.312 So thank you so much for joining us and we'll be back I think it's really crazy how, even the experts are trying to squint to see what's next and have a hard time figuring that out beyond, six months or a year. 397 00:37:14,847.312 --> 00:37:19,877.313 And I think that's what's cool about generative AI is it's, moving really fast and everyone's trying to figure it out. 398 00:37:20,77.313 --> 00:37:21,697.313 Yeah, I agree with you. 399 00:37:21,717.313 --> 00:37:30,807.312 It really excited me to hear about some of Grant's views on how AI can help. 400 00:37:31,117.313 --> 00:37:36,597.313 remove some of the barriers for people like me who aren't as comfortable with analytics. 401 00:37:36,807.313 --> 00:37:41,937.313 you can get data sets packaged in a way or insights package in a way based off your level of expertise. 402 00:37:41,937.313 --> 00:37:46,287.313 So if you're, you know Heavy data analyst with a lot of expertise in that area. 403 00:37:46,527.313 --> 00:37:48,897.313 You can get an output that meets you where you are. 404 00:37:48,897.313 --> 00:37:52,297.312 And if you're in the earlier stages, you can also get that. 405 00:37:52,317.313 --> 00:37:55,757.313 And I think that's really going to help people be more comfortable. 406 00:37:55,957.313 --> 00:38:04,227.313 Not only inquire about data, but also ask questions because if you can do it with a bot, you don't have to worry about feeling stupid or like you should know this already. 407 00:38:04,237.313 --> 00:38:18,727.314 It removes all of those What about you, Erin? What was your big takeaway Yeah, what I found, really interesting, but also really empowering from what Grant had to say about the role AI is going to play specifically in analytics is the accessibility. 408 00:38:19,17.314 --> 00:38:27,677.314 You know I'm comfortable with data, but if it's an area in reporting that I'm not familiar with, doesn't always come natural to me. 409 00:38:27,677.314 --> 00:38:28,7.314 Right. 410 00:38:28,47.314 --> 00:38:32,897.314 And so him saying that the chat bot or AI could produce. 411 00:38:33,192.314 --> 00:39:01,2.314 A report that was made for me at my level of understanding of that topic, where I could also ask questions and not feel judged really is empowering to make me feel like, okay, I can dive in at my level, but also someone who's maybe more advanced or maybe someone who's a little bit less experienced in that area could also dive in and get the information they need and are looking for without some of those Psychological barriers we put up around not feeling confident or not wanting to look stupid. 412 00:39:01,202.314 --> 00:39:03,852.313 Yeah, I really think about it as being a leader. 413 00:39:03,892.313 --> 00:39:16,847.313 And, when you're bringing new folks on or folks that are new to the industry or new to marketing and really being able to take a lot of the things that maybe come, second nature to those of us have been doing it like a hundred years. 414 00:39:17,137.313 --> 00:39:23,147.313 And really be able to help translate what that data means to them and to provide a little bit more self service. 415 00:39:23,147.313 --> 00:39:24,627.313 Cause I think it can be so overwhelming. 416 00:39:24,647.313 --> 00:39:33,282.213 a dice lot of times, especially as you're more junior in your career, or as you're trying to help people get a handle on what's the action they should take from it. 417 00:39:33,452.213 --> 00:39:34,772.213 It's not always that intuitive. 418 00:39:34,772.213 --> 00:40:03,992.213 And so I feel like with generative AI being able to query what is You know, what's the next thing? Or what should I be thinking about? Or, what does this even mean to me? Can be really powerful to your point, not having to go to somebody and ask Sometimes people feel a little oh god, I should know this or There's something that maybe just should be a little bit more intuitive and it's just not necessarily So I think that's what excites me the most I think it's definitely worth checking out, and we can certainly talk about it on a later date. 419 00:40:04,252.213 --> 00:40:08,172.212 But I do think there's a lot of tools out there on the analytics side. 420 00:40:08,422.212 --> 00:40:12,122.113 And as generative AI is continuing to evolve in that direction. 421 00:40:12,362.213 --> 00:40:20,402.212 In that realm, we can dive a little bit deeper, but I'm super excited to, really understand the future of data analytics and marketing. 422 00:40:20,402.212 --> 00:40:27,717.213 And I think one of the other powerful things that Grant said, and one of my key takeaways is really about, it doesn't have to be perfect. 423 00:40:27,817.213 --> 00:40:41,897.2125 I feel like oftentimes we're trying to nail what is that one thing that, had a prospect turn into an opportunity? And I think it's really about analyzing across accounts and getting a better sense of directionally what's right. 424 00:40:42,97.2125 --> 00:40:54,207.2125 Yeah, it was making me think of some previous roles that you and I had working together where we had some kind of flagship metrics that we lived and died by as a team. 425 00:40:54,487.2125 --> 00:41:14,947.212 And when we were onboarding new people, imagine having some sort of, chat bot that could give you the ability to say, here are dashboards, and now you can ask questions about why do we measure it that way, or what are some reasons that a customer might not move to the next stage that we want them to so that we don't actually have to be there to answer their questions that they can actually dive in more at their comfort level. 426 00:41:15,147.212 --> 00:41:16,67.212 Even at our level. 427 00:41:16,67.212 --> 00:41:20,447.212 I think that the thing is, I'm still trying to learn and trying to understand what's important. 428 00:41:20,467.212 --> 00:41:30,287.212 And I think what Gen AI does for us is it really also helps with the segmentation because not all accounts look the same as we know, AI's existed for a while now and certainly has helped us. 429 00:41:30,307.211 --> 00:41:46,597.212 But the generative AI piece, just being able to like bounce ideas off of it or, tell me a little bit more about, this conversion rate compared to this getting those insights, I think will be super cool and really being able to then incorporate those into our broader programs is something I'm super excited for. 430 00:41:46,647.212 --> 00:41:51,437.213 So more to come, certainly, but a huge thank you to Grant Gregorian. 431 00:41:51,687.213 --> 00:41:54,887.214 If you guys have not interacted with Grant before he's just lovely. 432 00:41:54,927.214 --> 00:41:59,697.214 And certainly check him out on LinkedIn, as well as Moji Technologies. 433 00:41:59,747.214 --> 00:42:00,877.214 Didn't really suck. 434 00:42:01,162.214 --> 00:42:02,382.214 Yeah, it sucked pretty bad. 435 00:42:02,382.314 --> 00:42:05,212.214 I think I need to angle my microphone down. 436 00:42:05,412.214 --> 00:42:09,292.214 Because I'm like looking at how your sound looks, and it looks a lot louder. 437 00:42:09,492.214 --> 00:42:11,202.214 I'm Naturally louder. 438 00:42:11,252.214 --> 00:42:11,992.214 Sorry, Jake. 439 00:42:12,192.214 --> 00:42:12,442.214 right. 440 00:42:12,642.214 --> 00:42:21,482.213 So more to come on the analytics front, which I could not be more excited about, and I think it's going to be very cool to get better insights. 441 00:42:21,682.213 --> 00:42:25,62.213 To really be able to incorporate into all of our marketing programs. 442 00:42:25,72.213 --> 00:42:32,292.213 So a massive thank you to Grant and thanks to Grant over the years for educating me and others in the industry as well. 443 00:42:32,602.213 --> 00:42:34,42.212 And thanks to everybody listening. 444 00:42:34,52.213 --> 00:42:38,432.213 Hopefully you took away some interesting takeaways on our marketing ops special here. 445 00:42:38,462.213 --> 00:42:38,772.213 Yeah. 446 00:42:38,972.213 --> 00:42:40,702.213 Thank you to Grant Gregorian. 447 00:42:40,712.213 --> 00:42:47,452.213 If you haven't checked him out, Check them out on LinkedIn, check out Moji Technologies and really, dive into the metrics. 448 00:42:47,452.213 --> 00:42:49,572.212 I think that's the takeaway I have for today. 449 00:42:49,992.213 --> 00:42:52,112.2115 And thank you to everybody that's been listening. 450 00:42:52,162.3115 --> 00:42:54,432.3115 let's keep crafting the future of marketing together.