WEBVTT - SignalFlare.ai’s Lukianoff on Data Analytics, AI 0:00:13.360 --> 0:00:16.040 Welcome to Chopping It Up. I'm your host, Mike Hanlon, 0:00:16.040 --> 0:00:19.040 the senior restaurant and food service analysts at Bloomberg Intelligence. 0:00:19.239 --> 0:00:22.239 Our research and that of bi's five hundred analysts around 0:00:22.280 --> 0:00:25.239 the globe can be found exclusively on the Bloomberg terminal. 0:00:25.760 --> 0:00:27.240 If you enjoy the pod, I'd love it if you 0:00:27.240 --> 0:00:29.720 could leave us a review on Apple or Spotify. Today 0:00:29.720 --> 0:00:33.200 we're joined by Mike Luciano, founder and CEO of signal 0:00:33.280 --> 0:00:37.760 Flare dot ai. Signal Flare is a decision intelligence platform 0:00:37.840 --> 0:00:41.440 for restaurants and other industries that translates data into decision 0:00:41.520 --> 0:00:46.159 frameworks using AI and machine learning models. The company was 0:00:46.200 --> 0:00:50.000 the twenty twenty four Snowflake Startup Challenge winner. Welcome back 0:00:50.040 --> 0:00:51.640 to the podcast. 0:00:51.159 --> 0:00:54.000 Mike, Oh, it's my pleasure of Michael. Good to see you, 0:00:54.040 --> 0:00:54.600 good to hear you. 0:00:55.160 --> 0:00:56.880 Yeah, I was going through my notes yesterday. I can't 0:00:56.880 --> 0:00:58.240 believe it was three years ago? 0:00:58.640 --> 0:00:59.240 Is it really? 0:00:59.680 --> 0:01:02.600 Yea three years ago? You were you were episode number four? 0:01:03.040 --> 0:01:06.120 All right, well I was. I was honored then honored 0:01:06.160 --> 0:01:08.560 to be back, so thanks for thanks for having me again. 0:01:08.959 --> 0:01:11.319 Well, I always love talking about their industry with you, 0:01:11.520 --> 0:01:15.639 so it's great to have you back on episode seventy six, 0:01:16.480 --> 0:01:19.600 UH signal flair. Like I said, three years now, time flies? 0:01:19.640 --> 0:01:21.320 What inspired you to start the company? 0:01:21.680 --> 0:01:24.399 So, you know, i'd been as as you know, you 0:01:24.440 --> 0:01:28.920 know this, I'm on middle of my my third decade 0:01:29.360 --> 0:01:34.480 doing UH data and analytics for restaurants. And when I 0:01:34.520 --> 0:01:37.240 started in it, there weren't that many people who cared 0:01:37.240 --> 0:01:41.560 that much about data for restaurants, especially pulling it out 0:01:41.560 --> 0:01:43.800 of pulling out of the point of sale and you know, 0:01:43.840 --> 0:01:47.920 figuring out what it all means and building building models, UH, 0:01:48.120 --> 0:01:51.520 you know, can predict things. UH. And the world's changed 0:01:51.520 --> 0:01:53.400 a lot since then. So and this is you know, 0:01:53.400 --> 0:01:56.000 that was a few a few companies ago, and you 0:01:56.040 --> 0:02:00.120 know the last one I uh, I sold and exited. 0:02:00.360 --> 0:02:05.360 And then you know, I took some time off, you know, 0:02:05.400 --> 0:02:08.560 around twenty twenty, and turned out that some other people 0:02:08.600 --> 0:02:12.280 took some time off around them too, And as I 0:02:12.320 --> 0:02:17.120 started to analyze data, because people were still coming to 0:02:17.160 --> 0:02:18.960 me to say, you know, can you analyze this? And 0:02:19.000 --> 0:02:20.880 so forth, and a lot of the old methods, you know, 0:02:20.960 --> 0:02:27.360 when demand patterns just completely change like they did during 0:02:27.400 --> 0:02:32.440 that time, and even trade areas completely changed as people 0:02:32.960 --> 0:02:37.720 changed working from home and you know, and so you know, 0:02:37.800 --> 0:02:40.040 I had to come at a lot of these problems 0:02:40.080 --> 0:02:44.360 from a completely different in a completely different way. You know. 0:02:44.440 --> 0:02:46.720 So instead of using a lot of these linear models 0:02:46.760 --> 0:02:49.520 that look at trends over time, say, okay, well, how 0:02:49.520 --> 0:02:53.799 do I change both the statistical approaches to more of 0:02:53.840 --> 0:02:56.520 a machine learning approach but also the data sets. Right, 0:02:56.720 --> 0:02:59.160 how can I instead of just starting with the point 0:02:59.200 --> 0:03:04.640 of sale data, how do I pull data from mobile 0:03:04.880 --> 0:03:09.639 you know, mobile digital platforms, right, Uh, you know, cell 0:03:09.680 --> 0:03:12.760 phone devices to understand you know where people are shopping 0:03:12.800 --> 0:03:18.239 when they're shopping there, and you know who they are 0:03:18.520 --> 0:03:21.800 right going into every restaurant and every retailer, and you know, 0:03:21.880 --> 0:03:26.600 and then you know, bringing in local economic data and uh, 0:03:26.760 --> 0:03:30.640 credit card panels, just masses of external data you that 0:03:30.720 --> 0:03:35.040 tells us more things about restaurant consumers before I start 0:03:35.120 --> 0:03:38.560 layering and the point of sale data. Uh. And what 0:03:38.640 --> 0:03:41.640 I found was two things, right, One is that it 0:03:41.760 --> 0:03:48.720 builds an infinitely more robust model. And in two, instead 0:03:48.720 --> 0:03:51.880 of always looking in the rear view mirror, you can 0:03:51.920 --> 0:03:56.400 start to layer in things that are forward looking. So 0:03:56.480 --> 0:03:59.880 for instance, when you know when gas prices started to spike, 0:04:00.120 --> 0:04:03.840 right before you know the inflation like really went nuts. 0:04:04.960 --> 0:04:07.600 We were able to take that information because we know, 0:04:08.440 --> 0:04:11.680 you know, which restaurants are more dependent on people who 0:04:11.960 --> 0:04:15.400 drive more. We can get all of that from from 0:04:15.440 --> 0:04:18.200 the data, and we know and we can calculate what 0:04:18.200 --> 0:04:21.560 people's disposable income is, not just uh you know, not 0:04:21.720 --> 0:04:25.040 just you know what the areas, uh you know income 0:04:25.120 --> 0:04:28.760 levels are. Uh. So when we can factor that in 0:04:29.040 --> 0:04:32.520 and we know that some of these things are leading indicators, uh, 0:04:32.560 --> 0:04:34.479 then we can use that data to say, you know 0:04:34.520 --> 0:04:37.039 what these restaurants are the ones they are going to 0:04:37.040 --> 0:04:39.680 be having trouble in you know, three to five months. 0:04:39.720 --> 0:04:42.279 So what kind of strategies to you employee? Right? Do 0:04:42.320 --> 0:04:44.880 you do you do discounting? Right? Or do you send 0:04:44.880 --> 0:04:47.440 different messaging? But when you start to use it in 0:04:47.480 --> 0:04:50.640 that framework, then you know, then it becomes uh, you know, 0:04:50.839 --> 0:04:56.680 planning tools, not just you know, backward looking analysis. So 0:04:56.680 --> 0:04:58.920 so so long, long story short. I did that for 0:04:59.000 --> 0:05:01.440 a while and really moved it out with with some 0:05:01.560 --> 0:05:05.440 chains more on a consulting basis, and about two years 0:05:05.440 --> 0:05:08.120 ago I said, you know what this is. This this 0:05:08.160 --> 0:05:11.240 is a platform, right, this is this is a much 0:05:11.360 --> 0:05:14.000 different thing than what I've done in the past in 0:05:14.080 --> 0:05:16.200 terms of being able to make it, you know, scalable 0:05:16.240 --> 0:05:19.360 and accessible to like really any size restaurant. Yeah, it's 0:05:19.400 --> 0:05:19.840 really cool. 0:05:19.880 --> 0:05:23.160 What's how many units does your smallest customer have? 0:05:23.720 --> 0:05:27.279 I think our smallest is a unit. You know, we're 0:05:27.480 --> 0:05:29.479 in the process now of taking all of that data 0:05:29.480 --> 0:05:32.839 and putting it into a platform that's going to be 0:05:33.480 --> 0:05:37.760 a lot easier to access, you know, even I think 0:05:37.800 --> 0:05:43.000 to independent restaurants because the way that the way that 0:05:43.040 --> 0:05:45.520 you had to operate the data in the past, you know, 0:05:45.640 --> 0:05:47.359 to do the kind of work that we're doing, and 0:05:47.400 --> 0:05:49.480 we had to pull in all of their clean point 0:05:49.480 --> 0:05:51.440 of sale data and so forth. But now we know 0:05:51.480 --> 0:05:54.560 what's actually going on, you know, in the surrounding community, 0:05:54.680 --> 0:05:57.080 right we know, you know how people are spending and 0:05:57.160 --> 0:05:59.479 what the trends are, and you know what segments they're 0:05:59.480 --> 0:06:02.240 spending and how much they're spending and all of that. 0:06:02.560 --> 0:06:06.400 So uh and we're putting uh, you know, an AI 0:06:06.960 --> 0:06:10.000 interface over it that also leverages all of the major 0:06:10.320 --> 0:06:13.320 major models. So you know, by doing that, we're able 0:06:13.400 --> 0:06:17.200 to cross section our machine learning models, you know, with 0:06:17.400 --> 0:06:21.080 the broader market understanding. And you know, you know, before 0:06:21.120 --> 0:06:23.840 I was a in restaurant data and tech. I was. 0:06:23.960 --> 0:06:27.440 I was restaurant operator. So the idea that we can 0:06:28.040 --> 0:06:31.000 you know, go from you know, early days working only 0:06:31.040 --> 0:06:33.679 with the very largest mega chains to now being able 0:06:33.720 --> 0:06:37.400 to provide similar things all the way down to small 0:06:37.480 --> 0:06:40.359 chains and independence and that that excites me, right, not 0:06:40.560 --> 0:06:43.200 just for you know, for the technology, but for what 0:06:43.240 --> 0:06:44.920 we were able to do for the industry. 0:06:45.160 --> 0:06:49.320 Yeah, it's really cool. So you know, as I mentioned, 0:06:49.360 --> 0:06:52.800 it's a decision intelligence platform. Did you always envision it 0:06:52.800 --> 0:06:55.960 to be that or was it? Did it start you 0:06:56.000 --> 0:06:59.160 know with menu pricing? You know, was there one problem 0:06:59.200 --> 0:07:02.279 that you were most about solving back when you started it? 0:07:02.520 --> 0:07:05.919 Yeah, So you know, I think you know, as you know, 0:07:06.040 --> 0:07:07.720 and you know a lot of folks who I've worked 0:07:07.760 --> 0:07:10.400 with over the over the years, you know, I've built 0:07:10.440 --> 0:07:13.360 a lot of pricing models, right, so you know a 0:07:13.400 --> 0:07:16.200 lot of you know, my my last company you know, 0:07:16.400 --> 0:07:19.200 still is the pricing engine for for a lot of 0:07:19.480 --> 0:07:24.640 national chains out there. So you know the methodology of 0:07:25.000 --> 0:07:30.200 how you apply sophisticated pricing algorithms to restaurants. It was 0:07:30.240 --> 0:07:36.280 the starting point. But the way that I was building it, 0:07:38.000 --> 0:07:41.440 you know, in order to build a really robust pricing model. 0:07:42.000 --> 0:07:44.880 You need to not just understand, you know, what people 0:07:44.920 --> 0:07:46.880 are paying for the price, and you know, it's not 0:07:46.920 --> 0:07:49.920 just about competitive benchmarking. That's that's a part of it, 0:07:50.760 --> 0:07:53.800 but you really need to understand what the fundamental demand 0:07:53.920 --> 0:07:56.400 is around each restaurant so that you're not just saying, 0:07:57.000 --> 0:07:58.720 you know, okay, how is this trending or how do 0:07:58.760 --> 0:08:01.720 I compare against them? But how do you layer everything 0:08:01.800 --> 0:08:07.560 from you know, sentiment to macroeconomic, microeconomic, you know, competitor 0:08:07.600 --> 0:08:10.680 performance data. How do you take all of those things? 0:08:11.320 --> 0:08:14.640 And once you bring all those things in, then the 0:08:14.720 --> 0:08:18.880 kinds of questions that you can answer go way beyond pricing, right, 0:08:19.000 --> 0:08:21.480 it goes to everything about you know, how do I 0:08:21.520 --> 0:08:24.640 optimize my market? Right? How do I look into new markets? 0:08:24.720 --> 0:08:28.120 You know, how do I start planning budget planning? And 0:08:28.160 --> 0:08:31.120 how do I do scenario analysis, you know, to to 0:08:31.200 --> 0:08:35.400 be able to react or to prepare for contingencies you know, 0:08:35.600 --> 0:08:40.920 in the competitive market or in the macroeconomic landscape. I think, 0:08:40.960 --> 0:08:43.680 you know, what I've been doing my whole career really 0:08:43.960 --> 0:08:47.840 has been what I call that last mile of turning 0:08:47.880 --> 0:08:51.520 the data into the decision. But in the past, you 0:08:51.559 --> 0:08:56.400 know that had to really happen by bringing insights and 0:08:56.520 --> 0:09:00.760 visuals and summary data to a really small our analysts, 0:09:02.320 --> 0:09:05.360 and all of that is is changing now because you know, 0:09:05.480 --> 0:09:07.679 when you can bring that kind of summary data in 0:09:07.760 --> 0:09:10.440 and have it read by an l M or or 0:09:10.760 --> 0:09:14.280 or an analyst agent, it need to be trained, right. 0:09:14.280 --> 0:09:17.600 It's still like having you know, an intern, you need 0:09:17.640 --> 0:09:23.400 to you need to bring along and teach. But it's 0:09:23.520 --> 0:09:26.360 moving so quickly that I'm just every day I get 0:09:26.400 --> 0:09:29.800 more surprised and impressed and excited about what those you know, 0:09:29.840 --> 0:09:31.760 what those possibilities are. Yeah, that's cool. 0:09:31.760 --> 0:09:33.520 And you were kind enough to invite me to your 0:09:33.520 --> 0:09:37.760 AI Restaurants summit recently. It was a great event and 0:09:37.800 --> 0:09:41.880 I learned a lot, but a couple of things stood out. One, 0:09:42.000 --> 0:09:44.679 can you talk a little bit about why data quality 0:09:44.760 --> 0:09:48.280 is so important for AI and machine learning projects? I 0:09:48.320 --> 0:09:52.160 was kind of floored about the number of product projects 0:09:52.160 --> 0:09:54.680 that fail due to data quality issues. 0:09:55.200 --> 0:09:59.160 Yes, yes, no, really really important. And it's funny because 0:09:59.480 --> 0:10:01.920 you know, maybe a year ago I was at another 0:10:02.040 --> 0:10:06.320 conference and you know, I heard somebody talking about you 0:10:06.360 --> 0:10:08.920 know how oh well, now in the age of AI, 0:10:09.520 --> 0:10:11.280 you know, you don't need to worry about the data 0:10:11.320 --> 0:10:13.360 anymore because a I will just clean it up and 0:10:13.600 --> 0:10:15.760 it's like nothing could be farther from the truth, right 0:10:16.440 --> 0:10:18.319 there are you know, there are ways that you can 0:10:18.440 --> 0:10:21.800 use AI to help you, you know, clean it up 0:10:21.840 --> 0:10:26.200 and so forth. But you really the standard in the 0:10:26.240 --> 0:10:32.320 bar for data, you know, cleanliness and context and and 0:10:32.440 --> 0:10:37.719 usability is higher, not lower, right, because what an ll 0:10:37.880 --> 0:10:40.200 M will do, right, what these you know, you know, 0:10:40.240 --> 0:10:44.559 the chat GPTs and the clauds and all these they 0:10:44.559 --> 0:10:48.640 will give you an answer that sounds wonderful, right, even 0:10:48.679 --> 0:10:51.679 if there's no data to support it. They they you know, 0:10:51.920 --> 0:10:55.079 what's amazing about them, right, is that they can get creative, right, 0:10:55.120 --> 0:10:58.000 as creative as a person. So you know, the data 0:10:58.160 --> 0:11:00.559 might not be there right, or it might interpret it 0:11:00.600 --> 0:11:04.640 completely wrong, but the pros sounds wonderful, right. And if 0:11:04.679 --> 0:11:08.120 you've got you know, crappy data that's not structured, and 0:11:08.280 --> 0:11:10.960 you know, in fact, really what they're best at is 0:11:11.000 --> 0:11:14.480 reading text documents. You know, if all of your data 0:11:14.520 --> 0:11:17.080 is in a data warehouse, you need another layer over 0:11:17.160 --> 0:11:20.680 that to say, okay, well how do I summarize this 0:11:20.840 --> 0:11:22.200 and how do I chunk it out? And how do 0:11:22.240 --> 0:11:25.080 I make sure that it's in usable nuggets, so that 0:11:25.080 --> 0:11:29.480 then the machine can can can interpret it. You know, 0:11:29.520 --> 0:11:33.640 people are using you know, these for calculations, but they're 0:11:33.679 --> 0:11:37.080 not good at calculating. You know. Maybe it's a unpopular 0:11:37.520 --> 0:11:40.679 you know fact, right, I don't know popular opinion. It's 0:11:40.720 --> 0:11:44.680 a fact. They can invoke tools, right that can help 0:11:44.720 --> 0:11:48.959 to do a calculation. But when it comes down to it, right, 0:11:49.040 --> 0:11:51.000 I think you know the stats that I put out there. 0:11:51.120 --> 0:11:55.600 You know, look, you know eighty percent of of AI 0:11:55.720 --> 0:11:58.720 projects are still failing, and seventy five percent of those 0:11:58.760 --> 0:12:00.800 are failing because they didn't get the data, right. 0:12:01.080 --> 0:12:05.400 Yeah, staggering numbers. Yeah, and you mentioned the fact that 0:12:05.880 --> 0:12:09.720 ll ms are terrible at math. We also discuss some 0:12:09.760 --> 0:12:14.079 other concerning issues, including hallucinations and excessive praise. 0:12:15.040 --> 0:12:18.280 Yeah, it is interesting how you know they can be 0:12:18.360 --> 0:12:21.680 tuned right, and you can tell it, you know, don't 0:12:21.720 --> 0:12:23.840 don't praise me, or you can tell it, you know, 0:12:26.120 --> 0:12:29.760 you know, you can try and give it a personality, 0:12:29.760 --> 0:12:33.440 which is fascinating. But yeah, I mean hallucinations is just 0:12:33.679 --> 0:12:36.400 you know, a fancy way of you know, of saying 0:12:37.240 --> 0:12:39.640 you know, it just it's just b s's right, their 0:12:39.720 --> 0:12:44.320 language models, right, they read you know everything that's ever 0:12:44.360 --> 0:12:48.559 been written, and then they figure out how to you know, 0:12:48.720 --> 0:12:51.319 you ask a question and then it's you know, searching 0:12:51.440 --> 0:12:54.280 this repository to try and say it in the way 0:12:54.400 --> 0:12:59.839 or the style that you specified. So you know, you know, 0:13:00.080 --> 0:13:02.800 and you know an example of like you know, bad 0:13:02.880 --> 0:13:06.240 mathroom an l M. Right, it's like you know, we 0:13:06.440 --> 0:13:12.720 we we're constantly doing you know, tests and validations, but 0:13:13.400 --> 0:13:15.880 you you don't know what the answer is that you're 0:13:15.920 --> 0:13:19.320 going to get because it's all probabilistic, right. What you know, 0:13:19.400 --> 0:13:23.160 what we do in analytics is deterministic. Right. You need 0:13:23.240 --> 0:13:25.760 a number, right that you've got some level of certainty 0:13:25.800 --> 0:13:28.040 and there's this and there's a clear calculation to it. 0:13:28.720 --> 0:13:30.760 You know. L l M s are all you know, Hey, 0:13:30.960 --> 0:13:34.160 if you ask this question, then the probability of this, this, 0:13:34.160 --> 0:13:36.360 this and that right ur X and y and that's 0:13:36.440 --> 0:13:40.480 what all these sort of computational nodes are doing as 0:13:40.520 --> 0:13:46.120 it's doing the calculation. So, uh, you know, so if 0:13:46.160 --> 0:13:48.679 you telling it, you know, an l l M without 0:13:48.720 --> 0:13:51.400 any more instructions, you know, hey, tell me what the 0:13:52.080 --> 0:13:54.640 year of a year sales are for you know, X, 0:13:54.760 --> 0:13:58.280 Y and Z company. You know twenty you know this 0:13:58.400 --> 0:14:02.000 year versus last year. The most likely answer you're going 0:14:02.080 --> 0:14:06.000 to get out of that is going to be something 0:14:06.040 --> 0:14:10.240 that says, oh, well, you're down about you know, fifty percent, 0:14:10.480 --> 0:14:13.280 you know versus last year. Why because we're about halfway 0:14:13.280 --> 0:14:17.000 through the year, right, So it's comparing this year, you know, 0:14:17.080 --> 0:14:20.360 the last year without any more you know, clarity than that. 0:14:20.560 --> 0:14:23.920 So as you're either figuring out how to prompt it 0:14:23.960 --> 0:14:26.360 and ask the questions correctly, or you're trying to figure 0:14:26.400 --> 0:14:29.280 out how to do the engineering in the background, you know, 0:14:29.360 --> 0:14:32.000 the things that the ways that we talk about things 0:14:32.040 --> 0:14:34.520 and the common vernacular that we use, you know in 0:14:35.200 --> 0:14:37.160 this industry, as you know, it's like, okay, well, what's 0:14:37.160 --> 0:14:40.320 a check average? You know, as fifteen different operators, what 0:14:40.440 --> 0:14:42.800 you know what that means to them? Right? How are 0:14:42.840 --> 0:14:47.280 you calculating that check average? It's all right, Well, you 0:14:47.400 --> 0:14:50.080 kind of need a single standard, right, or a very 0:14:50.080 --> 0:14:53.160 clear definition of all of these things in order for 0:14:53.200 --> 0:14:54.720 them to work and to learn. 0:14:55.000 --> 0:14:58.000 Yeah, I learned at the event that it's important on 0:14:58.200 --> 0:15:00.680 us to ask the right question. That's you get better 0:15:00.720 --> 0:15:05.080 at using these lll ms. One of the subjects that 0:15:05.560 --> 0:15:09.760 I brought up was just concern about you know, security 0:15:09.840 --> 0:15:13.360 of your data and what the lll ms are doing 0:15:13.400 --> 0:15:16.440 with your data. Actually saw a great tweet today. It 0:15:16.480 --> 0:15:19.800 was Stealing from one person is called theft. Stealing from 0:15:19.840 --> 0:15:22.040 everyone all at once is called chat GBT. 0:15:22.600 --> 0:15:25.360 Yes, no, I think. I think that this is a 0:15:25.400 --> 0:15:32.960 reality of where we are. It's really important because you know, people, organizations, right, 0:15:33.120 --> 0:15:36.480 first of all, it's moving so quickly, and you know, 0:15:37.240 --> 0:15:43.160 once any you know, analyst, employee, marketing, you know, administrator, 0:15:43.200 --> 0:15:46.440 whoever it is in the organization, once they start using it, 0:15:46.520 --> 0:15:49.480 they start to find all of these ways that, oh 0:15:49.720 --> 0:15:52.560 this can actually save me. You know this amount of 0:15:52.600 --> 0:15:55.440 time you know doing these three, four or five tasks. 0:15:56.160 --> 0:16:00.320 So it's addictive very quickly, right because people or like 0:16:00.360 --> 0:16:02.000 I can do a better job with it. You know, 0:16:02.080 --> 0:16:05.200 it's one of the best productivity tools you know, invented 0:16:05.280 --> 0:16:09.120 probably since you know, since Lotus one, two three, right 0:16:09.200 --> 0:16:13.160 precursor to to Excel for those of who don't know. So, 0:16:13.240 --> 0:16:20.280 if you don't have a policy for enterprise level, which 0:16:21.400 --> 0:16:26.520 typically right the the g p T, the l M 0:16:26.800 --> 0:16:29.480 will say that in their enterprise versions, you know, none 0:16:29.520 --> 0:16:33.440 of your data is used to train their models. There's 0:16:33.480 --> 0:16:35.520 another way of using it, which is, if you use 0:16:35.600 --> 0:16:38.520 the A p I S in your own environment, then 0:16:38.680 --> 0:16:41.760 you know, there's no backward propagation, right because it's it's 0:16:41.800 --> 0:16:45.280 being calculated in yours and you're using you know, you're 0:16:45.280 --> 0:16:47.280 getting the benefit of it, but there's no way that 0:16:47.320 --> 0:16:49.880 it can actually backward propagate, which is, you know, which 0:16:49.920 --> 0:16:54.160 is the better the better process. But if you have 0:16:54.240 --> 0:16:58.000 no policy, right and you're not, you know, and then 0:16:58.280 --> 0:17:02.120 if people are taking your company data and just using 0:17:02.160 --> 0:17:06.560 it on a free version or a personal version, all 0:17:06.640 --> 0:17:10.600 of that information is being you know, you're you're you're 0:17:10.640 --> 0:17:14.440 donating to the you know, to the broader knowledge base 0:17:14.520 --> 0:17:17.760 of the world right with whatever you put in there. 0:17:18.440 --> 0:17:22.560 So so yes, you need to be looking at enterprise 0:17:22.640 --> 0:17:26.000 grade tools to be able to use this, but don't 0:17:26.000 --> 0:17:29.160 fool yourself that, you know, if you're not using it, 0:17:29.400 --> 0:17:32.359 then uh, you know that then you're safe. You know, 0:17:32.520 --> 0:17:35.000 if you don't have a policy in place, then you're 0:17:35.080 --> 0:17:39.720 probably in more security danger than than if you have 0:17:39.800 --> 0:17:41.600 one and you're actually teaching people the right way to 0:17:41.680 --> 0:17:42.000 use it. 0:17:42.200 --> 0:17:45.720 Great, let's switch gears. With all the inflation we've seen 0:17:45.800 --> 0:17:49.919 since twenty twenty, pricing menus down to the unit level 0:17:50.040 --> 0:17:54.520 is more important than ever. And you know, I guess 0:17:55.160 --> 0:17:57.120 what I'm wondering is what chains are doing it right? 0:17:57.200 --> 0:17:57.640 Right now? 0:17:57.880 --> 0:18:01.679 Yeah, I think you know, without naming name specifically, but 0:18:01.760 --> 0:18:05.880 I think that the things that you see are common 0:18:05.920 --> 0:18:10.240 with success are you know, those that I think are 0:18:10.280 --> 0:18:14.080 are actually well. One thing is that, you know, what 0:18:14.200 --> 0:18:18.879 we keep on seeing that qs RUH is lagging, you know, 0:18:18.960 --> 0:18:25.080 even entry level casual dining and and fast casual and 0:18:25.200 --> 0:18:28.680 when economic times are uncertain, usually we expect that people 0:18:28.760 --> 0:18:30.680 are going to trade all the way down to QSR 0:18:30.720 --> 0:18:32.960 because it's the least expensive option. But that is not 0:18:33.080 --> 0:18:35.639 what we're seeing, right. We're seeing that it's you know, 0:18:35.720 --> 0:18:38.320 it's you know, it's a lot of the fast casuals 0:18:38.359 --> 0:18:40.960 and so forth that are doing a lot better, especially 0:18:41.040 --> 0:18:45.560 if the cuisine that they're serving up is not yet 0:18:45.840 --> 0:18:48.560 you know, at a saturation point and in a lot 0:18:48.600 --> 0:18:51.359 of their markets. Right, I think what the consumer is 0:18:51.440 --> 0:18:56.200 telling us is that, you know, they are making choices, 0:18:57.040 --> 0:18:59.680 but their choices are being guided not by what's the 0:18:59.720 --> 0:19:02.080 least expensive thing that I can get, but what actually 0:19:02.119 --> 0:19:04.600 represents the value, Right, what's going to what's going to 0:19:04.600 --> 0:19:06.560 give me better service, what's going to be the taste 0:19:06.560 --> 0:19:08.720 profile that I want, because if I'm going to dine out, 0:19:08.840 --> 0:19:11.400 you know, maybe less often, I want to make sure 0:19:11.480 --> 0:19:14.040 that those occasions that I am dining out, you know, 0:19:14.080 --> 0:19:15.320 I'm actually getting what I want. 0:19:15.600 --> 0:19:18.280 Yeah, that one has caught us by surprise over the 0:19:18.359 --> 0:19:22.080 last five months, the fact that casual dining and it's 0:19:22.160 --> 0:19:25.200 broad based. It's not just Chili's, it's broad based. Most 0:19:25.200 --> 0:19:29.119 of the casual dining chains we cover, you know, Cheesecake Factory, 0:19:29.400 --> 0:19:32.520 Texas Roadhouse still crushing it, Dartin Cracker Barrel, all these 0:19:32.600 --> 0:19:36.080 chains are now performing qs R. Part of it is 0:19:36.080 --> 0:19:39.960 they're they're lapping easier comp but they're being them by 0:19:40.000 --> 0:19:44.000 a wide margin. Man. We haven't seen that, you know, 0:19:44.119 --> 0:19:47.479 in a market, in a slower restaurant spending market. I mean, 0:19:47.520 --> 0:19:49.640 I don't think I've never seen it, you know, since 0:19:49.680 --> 0:19:50.600 I've been in the business. 0:19:51.920 --> 0:19:54.240 Well, you know, the other thing that we really need 0:19:54.280 --> 0:19:59.240 to think about in context, right, is how how the 0:19:59.320 --> 0:20:03.800 industry shrank and then grew, you know, from twenty nineteen 0:20:04.040 --> 0:20:09.400 to now and where it's going. You know, we all 0:20:09.400 --> 0:20:13.720 remember back in mid COVID eight times, right, you know, 0:20:14.200 --> 0:20:17.960 staggering numbers of closings. Right, It's like, okay, like twenty 0:20:18.000 --> 0:20:21.520 percent of the industry was closing. And but you know 0:20:21.560 --> 0:20:24.159 what if you had a drive through, you know you 0:20:24.200 --> 0:20:26.440 were living you know, you were living the good life. 0:20:26.560 --> 0:20:29.840 You know, you had you know, there was you know, 0:20:29.880 --> 0:20:32.040 there were a few businesses that were as good as 0:20:32.119 --> 0:20:34.679 you knows as having a drive through right in an 0:20:34.760 --> 0:20:38.840 underserved market, right. And what ended up happening is that 0:20:39.200 --> 0:20:41.959 unit account started to grow, but it was all limited service. 0:20:42.600 --> 0:20:46.399 So the ratio of limited service to full service restaurants 0:20:46.960 --> 0:20:49.960 is completely different than it was back in twenty eighteen, 0:20:50.040 --> 0:20:57.840 twenty nineteen, big overbuilding of you know, of limited service restaurants, 0:20:58.680 --> 0:21:01.919 not not as much, you know, catching up with the 0:21:01.960 --> 0:21:06.400 full service. So you know, and yeah, and the way 0:21:06.440 --> 0:21:09.000 that the trade areas you know change, right. You know, 0:21:09.040 --> 0:21:12.040 people sort of you know move to more you know, 0:21:12.200 --> 0:21:15.040 more rural where they could because their workplaces would do it. 0:21:15.160 --> 0:21:17.119 Maybe they had to move a little bit closer, but 0:21:17.400 --> 0:21:22.280 you know, not so much into city centers. So all 0:21:22.320 --> 0:21:26.520 of those factors together, right, are showing these traditional metrics 0:21:26.560 --> 0:21:32.480 of same store sales as opposed to total industry growth right 0:21:32.560 --> 0:21:37.160 or total industry usage. They are two very different looks 0:21:37.280 --> 0:21:42.919 at the industry, right, because the the saturation of markets, 0:21:43.080 --> 0:21:46.680 especially when you look at different cuisine types and uh 0:21:46.880 --> 0:21:51.639 service styles. Because population growth is also, you know, is 0:21:51.680 --> 0:21:54.240 not near what it was five years ago either, So 0:21:54.960 --> 0:22:00.240 rest the restaurant build build out is outpacing population growth 0:22:00.320 --> 0:22:02.560 by about you know, about five x. 0:22:04.080 --> 0:22:07.000 Wow, five x is a lot. What I found interesting 0:22:07.040 --> 0:22:11.040 about twenty twenty as well, is like QSR had that 0:22:11.440 --> 0:22:14.600 year where they were able to raise prices pretty aggressively 0:22:14.680 --> 0:22:17.159 because they were the only game in town, and so 0:22:17.240 --> 0:22:20.919 I feel like their pricing got ahead of itself faster 0:22:21.520 --> 0:22:24.280 than it did for some of their full service competitors. 0:22:24.680 --> 0:22:27.240 Yeah, I think, I think you're right. And it wasn't 0:22:27.400 --> 0:22:31.919 just the menu prices. It was also that they eliminated 0:22:32.520 --> 0:22:36.880 every discount, right, So all the dollar menus disappeared, right, 0:22:36.960 --> 0:22:39.919 all of those you know, all of those you know, 0:22:40.000 --> 0:22:42.280 special things that you get, you know, if they gave 0:22:42.320 --> 0:22:46.960 you some price certainty inside those restaurants for some period 0:22:47.040 --> 0:22:50.640 of time, all of those just just vanished. Right. That's 0:22:50.680 --> 0:22:55.920 starting to come back, But they sort of retrained customers, right, 0:22:56.200 --> 0:22:58.960 They started, you know, started to find things that were 0:22:59.200 --> 0:23:01.639 you know, oh well, I can get better food you 0:23:01.680 --> 0:23:05.080 know all the time you know, at these different types 0:23:05.119 --> 0:23:11.159 of uh, you know, restaurants, you know, also limited service 0:23:12.080 --> 0:23:13.959 you know. But I but I like it better, right, 0:23:14.040 --> 0:23:19.119 So it really I think it created something that you know, 0:23:19.160 --> 0:23:22.600 I mean, we'll see if people revert. But qs R 0:23:22.680 --> 0:23:27.040 every day is getting more and more like off premise 0:23:27.119 --> 0:23:30.200 grocery you know, or you know, and even the convenience 0:23:30.200 --> 0:23:32.920