• John Ennis

Michael Nestrud - Sensory and the Consumer Voice


Welcome to "AigoraCast", conversations with industry experts on how new technologies are transforming sensory and consumer science!


AigoraCast is available on Apple Podcasts, Stitcher, Google Podcasts, Spotify, and PodCast Republic. Remember to subscribe, and please leave a positive review if you like what you hear!


Dr. Michael Nestrud is the founder and president of MNC consulting. Michael's unique point of view stems from his student days at The Culinary Institute of America and Cornell University, where he merged the disciplines of Culinary Arts, Food Science, Computer Science and Psychology to solve problems related to food choice, consumer emotional response, and menu design. Prior to founding MNC, Michael spent the past 10+ years working at government, market research and CPG organizations to design solutions to real business issues. Among other activities, Michael has served as Scientific Chair for the Society of Sensory Professionals and the conference chair for the 2017 Pangborn Sensory Science Symposium.


LinkedIn

MNC



Transcript (Semi-automated, forgive typos!)


John: Michael, thank you for being on the show today.


Michael: Thanks, John. Thanks for having me here.


John: Okay, great. So, Michael, something I think is interesting is that you and I both founded our companies around the same time. And it really seems like a tremendous amount has changed in the world in the past few years. So I think that to start the show, it'd be really interesting to hear you say you talk about what you've seen in your own client work and how you see the field changing in response to all the new technologies that seem really ubiquitous.


Michael: Sure. You know, I'll speak to one of the broad trends that I've seen is that I've heard this in every sensory conference since I've ever been to one. But it really seems to be pressuring pressure. There seems to be pressure now of doing more with less. And I think that I see that take into more and more extremes where, you know, previously might have been the loss of one head count. Now it's like loss of entire teams or entire structures and organizations. But yet the expectation that products are still being developed with the consumer in mind is still there. And, you know, when you think about the role of technology, I think it's taking a larger and larger role in how we manage our data, manage our information. And I also think that in certain ways, it hasn't taken a big enough role yet. And I think that's where I see us going, is using technology and really smart ways to facilitate what I just said of moving faster, but not without a loss of insight.


John: Interesting. So where do you see the biggest opportunities for technology to to help sensory scientists in the near future?


Michael: I think one of the biggest is meta-analysis across surveys. This is as much a technology issue as it is a structural issue of companies, you see companies that once a piece of research has done and delivered. And whatever decision is being made by that research, a lot of times doesn't even exist anymore. And the company of course are either physically or at least philosophically. It doesn't exist anymore. And so there's been some advances with some of the software out there. I know that there's organizations like yours that are working on helping companies database, provide databases. And I think that's going to go a long way because there's a lot of information out there that companies have already collected that if they could just leverage, they might not need to run, quite so big of studies in the future.


John: Yeah, that's interesting. I've said a lot to say on that topic, but I will hold back since you're the guest. Yeah. Okay. So that's I definitely agree with that. And then what else do you see? Do you see anything in terms of methodology or the ways that data are being collected that you think can be improved or that, you know, new ways of getting information that your clients should be using or are using?


Michael: Yeah, definitely. I'll use the example of like a home use test. And I'm not old enough to remember the 70's state of research where, you know, bubbling and paper surveys was the state of the art and then having to code all those out. But even though I did do some of that for the U.S. Army when I worked at data.


John: I remember one of the surveys we collaborated on. You showed it to me, this giant Scantron form.


Michael: The Scantron form. Yeah. I painstakingly built with latex, with an R back, and then generated all the randomization design.


John: Well, that was very impressive piece of work.


Michael: It was that I used exactly once. Very well. But I used exactly once. So but the way that we field surveys, at least in terms of the structure of the questionnaire. And we're still using fundamentally the same structure other than the fact that we're using technology to make it easier to pose and respond to questions for the consumer. And I think that's something that I haven't seen evolve yet as much as I would like to see. With all the new technology that we have, is there more that we can do besides ask people the 20, 30, 40, five minute survey with all the questions in order to answer everything on the same scales? And so that's where I see some of the neatest work potentially happening, although it's still slower than I would like to see it.


John: Okay, so what sort of measurements do you think Michael should be being made? You're saying that you'd like to see things kind of move along a little bit more quickly when it comes to the use of new technology, collecting different types of data instead of just, you know, the old liking scale or purchase interest. What sort of measurements do you think should be made and what might be more informative in terms of predicting actual behavior?


Michael: I think one of the biggest limitations of sensory studies is the fact that they are generally treated as one off experiences with a product. And a lot of times that's the first experience with a product. In the case of like CLT, or we'll call it a standard home use test where you've got you know, maybe they're tasting the first product over three or four days and then maybe there's a different version of the same product, then test over the next three or four days and then answer some question in the end and they're done. And with some of the new technology out there, I've seen the use of communities used not for sensory research, but I've seen them used very much on the more pure consumer research side for all sorts of things, whether it's ideation, whether it's to understand specific behaviors that consumers are using or deploying in their kitchens or living rooms or with their kids or bathrooms or whatnot. But I haven't seen that actually be merged yet where you've got, you know, maybe a longitudinal community that's consistently responding to sensory questions and whether it's about current products, new products, I think something, you know, something like that would really provide the ability for companies to, one, rapidly get information. So I think that's one. But two, you know, serve the need of needing an immediate decision because that's always gonna be there. But then also provide a little bit more information through that continued touch points with those same consumers. Or maybe they're slightly different consumers. I don't know. But continue touch points that allow you to check in. Okay, now that you've been consuming those products, how many times have you actually gone out and bought it? You know, you said you would buy it without having to initiate a new study. And right now you've got to initiate new studies to do that. And that's just it's inefficient. And so I think that leads to it just not being done.


John: Yeah. And I think it's a deep point. Actually, I see technology really helping with that. Something that I'm sure you've seen that's coming as Alexa based surveys or AI powered speaker based surveys, that something we started to do in our client work. And you can see, for example, Amamar had some demo they put up of this sort of thing. It's just like, on the one hand, it's a great idea. On the other hand, it's obvious if you're paying attention that this is what has to happen is that you're going to have, you've got all these devices that people can interact with. Why not leverage them to collect data? Right? So I could definitely see what these communities that Alexa, every now that might ping in, "Hey, you might take a minute to tell us about how, you know, your experience is going with whatever". That is very interesting.


Michael: Yeah, I completely agree. I imagine that with Alexa, Google Home or Google even Siri, that people's tolerance for answering 45 questions, JAR's scales etc., is going to be pretty low. But yet the demand for the similar insights is going to be just the same that come out of the traditional surveys. And so how do I design sensory into some of these new technologies, whether it's online experiences or even offline new ways of talking to consumers? I think that's can be critically important and I think that also builds, it should be a direction of our focus of our field over the next 5 to 10 years is to understand what exactly consumers are doing when they're responding to Alexa or Siri with sensory questions.


John: Right. It's very interesting, actually. Yeah. So speaking of these sort of thing, when you and I last spoke, it was around Christmas time I recall and you were just wrapping up. You were using some sort of Google tool or some sort of Google Dashboard creation tool that I thought was very cool. And I'd like to hear your kind of talk about that a little bit because I thought that was really neat application of new technologies to our field.


Michael: Yeah, absolutely. You know, I've been somewhat dissatisfied with, although I am certainly a practitioner of the 40, 50, 60 page PowerPoint, you know, report associate with studies. And a lot of that happens because, you know, as a supplier, no matter how much I try to understand the business, I don't understand the business as well as the clients so instead of trying to make sure that they could answer every single possible business question with at least one slide in your deck. And so you end up with this proliferation of slides when in reality the client only really needs a handful of questions answered. And so I was looking for technologies and I think I described to you my ideal technology would be a portable PowerPoint like deck that's had the interactivity with the graphs and information the same way a nice dashboard does. But all the dashboards are cloud based. And so what I did for this particular client was develop a cloud based dashboard with sensory data, it was exported from one of the popular software is out there and built it paged like a PowerPoint deck. It's just every now and then they go onto a slide that had a ton of information behind it. And so they can manipulate the sliders or the drug down boxes or the graphs or whatever to tailor to the specific question. And you know, what I found through doing that was that it was satisfying to me in particular because I was able to get the entire survey into what was a understandable and smallish type deliverable. And so every single question was in there. All of the analysis that could possibly be in there were behind it. But at the same time, it was simple, like it was clean. It could answer a simple question, or if you wanted to answer a complex question, the ability was there. And I wanted something good for the client beyond. I knew the immediate question was but would lead beyond that immediate question.


John: Yeah, that's fascinating. That's a great use of the mew technology, because I think communication is one of the biggest challenges we have as sensory scientists trying to communicate the value of what we've done to people who may not appreciate the science side that much, right? Yeah. I mean, to create essentially a living report where someone is presenting, but they can ask questions of the presentation. I think that's really exciting. So, yeah, I was invigorated by you.


Michael: Yeah, it's using Google Cloud platform in piecing together some of their technologies there. You know, Google's approach, it's all based on, you know, ad analytics like their entire cloud platforms based around helping people without ad analytics, like, not surprisingly, because that's where they get their income from. So tailoring it towards something like this took a lot of work and there's some funky limitations there. But, you know, with any of these technologies are all new and we all have limitations.


John: Seem to hack it. You can had to hack the interface. You kind of had to hack their interface. We have to hack the Alexa interface.


Michael: I wrote a lot of code to do really simple tasks, put it that way. Still there are things that would take you a few seconds in excel that due to the way the technologies are pieced together. So, you know, maybe a couple hours, but then once it solved, it solved forever, right? So there's some limitations. But again, it shows promise and it's just trying different things to see what might be helpful to clients going forward. And might that kind of guide me towards what technologies people actually want?


John: Yeah. So what was the name of the actual? Think it was Google Cloud? Is there something in particular?


Michael: GCP is Google's cloud platform, and that's the umbrella for everything that they have. And I use two elements of it, cloud data proc, which is their pre-processor. And so that brought in raw data from the sensory software and then allow me to transform it, do some calculations similar to the power BI. It works similarly to power BI if anyone is familiar with power BI, and then once that was done, push it into data studio and Google data studio is there visualization platform. So that was the core of the deliverable was the data studio aspect.


John: That's fascinating because I can imagine having an almost like a bot that you could interact with where you do your survey, you load everything and then start asking questions and, you know, can eventually start to give you answers in real time about, well, so what happened with these people? What about this? And have a conversation almost with the deck.


Michael: Yeah. Right. And, you know, I see a path to it to doing this and it's not there yet, but I see a path to connecting this to what would essentially be live or maybe it's hourly outputs from live surveys. And, you know, with the infrastructure in place, being able to see your final deliverable get kind of piece, you know, get built as the survey is being fielded and, you know, the technology is not there yet. So I don't want to get too excited for it because they're still data cleaning and some different steps that need to be done. But, you know, it's close. And I think within the next few years, we should be able to see that where you can have if the survey is being fielded in the cloud through, I'm personally a big fan of Compusense, but I use most other software out there and it's mostly all cloud based. If that's in the cloud and your reporting's in the cloud, then you should be able to connect it to, right? And that's where I think it will go. I'd like to see some of the software providers provide more access to the data they're collecting the API. I think that would make huge difference in the ecosystem that we have allowing that, I'm sure in your own work to having API access to survey some of the survey data would be extremely helpful.


John: Yeah, definitely. I can see that for sure. Okay, well, that's great. So then so, Mike, what are some of the other things that you're saying like? That sounds like an interesting project that you're working on. What are some other things you've worked on that have excited you in the past few years?


Michael: You know, this isn't gonna be technology related, so to speak, but I'm still gonna say it because it's still excited. You know, I got involved with quality work from my previous corporate role. And, you know, I've been through, you know, the one quality class at Cornell, I went through and you know, I've been through a couple of quality workshops at different conferences, but I didn't really understand it. And, you know, getting out there and working on a quality program to redesign it. And, you know, it really got me focused on the question of why does this, you know, quality control still matter to the consumer? And I thought, you know, it was a different point of view than working in innovation. But it was to me, in a lot of ways, it was the most important point of view, because in manufacturing, you're creating the consumer experiences life or the workers, right? And so when you look at the impact of the work, it was a direct line between, you know, creating a better program and creating better consumer experiences, right? And reducing complaints and kind of all those other things that go along with shoring up a quality control program. And, you know, I think one of the challenges with it is in traditional quality control, you know, kind of taught that like, well, every plant needs a descriptive panel. And I know that you've had some other DraughtLab was one of your previous speakers, Lindsay. And, you know, I think that they're taking an approach which I'm generally aligned with that you don't need a full blown descriptive panel trained in every single plan to do quality well. There's other ways to go about it. And it's like simple and impactful is the best way to go. And so I like that aspect of it. And so I'm continuing some quality as part of my consulting business. It's not the technology. It's not the fancy visualizations always, but it's some of the most important sensory work we can do and it gets neglected.


John: Interesting. But it is true that it does power technology. For example, you mentioned DraughtLab where you can have technology facilitates quick measurements, you can have dashboards if you have real-time updating of results potentially into a dashboard I mean it. It can still be decorated by technology.


Michael: Absolutely. I saw this, I was never able to get it all connected. But a lot of the you know, as the quality control systems themselves become more connected, you know, I was sitting with an engineer who was logged in to the plant manufacturing system in a different country and watching live, what was going on with the processing steps and including certain quality checks that weren't sensory related. And while it was an IT project to incorporate sensory into those, I can see that happening, right? So then you're starting to understand rather than guessing what's causing or affecting or is sensory doing its job in this particular place? When you've got it live and you can see it happen, like that's to me, one of the most exciting aspects of it. I haven't seen it happen with sensory data, I'm sure there's companies out there doing it. I just haven't personally seen it.


John: Well, I'm working on a graph database to support that really. Actually, I don't know if you've learned much about graph databases, but if a graph database can help you in real time to monitor complaints and I've got some stuff on my board here, but I can't see it. But yeah, the project I'm working right now is this sort of thing where you've got real time call center complaints, reviews, and you want to try to figure out what's wrong with the product and at the networks really help support that. So I think you've got the right vision that those IT. Yeah. This tool in IT are going to come and help us in sensory science.


Michael: Yeah, absolutely. And I mean, I can piggybacking on that a little bit. You know, one of the other areas that I got experience with that I think has been under influenced by sensory is this area of customer experience and customer service specifically. And so I had the opportunity to lead customer service for just under a couple of years and work with the front line agents that were answering the phones.


John: Oh, really? Wow!


Michael: Yes. You know, again, like I had one in the other manufacturing side with my quality experience and at the other end, you've got consumers calling in with complaints and I was surprised. I shouldn't have been, but I was surprised at both how articulate and generally how correct consumers were.


John: Consumer said to be under appreciated or at least underestimated a lot.


Michael: I agree. Well, I think we elevate. I mean, with any consumer customer service, there's going to be consumers that might be acting unreasonably. But I think we, in a fleet kind of the size of that in our minds, because they can tend to be kind of negative or frustrating experiences. In reality, time and time again, I would see I would see patterns and then I would go look for maybe a sensory checkpoint and quality or something like that, that could have prevented that. And lo and behold, there was a link between either a positive comments and something that happened or negative comments, something that happened. And so that taught me a lot about the consumer. And then on the data side, this was kind of fun, but it was, you know, trying to help quality and the customer service team translate consumer language in the way they would describe products into a way that R&D could understand because they had to do the diagnostics.


John: That's fascinating. That's a really interesting idea. Okay, I can use that idea almost immediately.


Michael: If you need a consultant I'm eager to help you.


John: I may just call you.


Michael: But, you know, it is the issue of consumers, you know, during inclination is like, well, you know, I just bought this soda and it tastes bad. And it's like, okay, well, it takes off. Okay. But how do you coach a customer service agent to go back and actually peel off the layers with an eye towards the issues that, you know, can happen in manufacturing or with the ingredients to try to separate the events from a maybe the consumer just had toothpaste and forgot about it, right? Like it tastes bad for that reason. Or more realistically, maybe there's a bad batch of ingredients or maybe the carbonation unit was a little out of spec or that variability was a little too high and you got that issue going on. So, you know, but without helping without an eye towards the root causes on the manufacturing side, it's very hard to create a good customer service agent that can go help a consumer diagnose that. Right? I think that's an interesting area for sensory in a lot of way because we know consumer language and we know technical language and we're constantly translating marketing language into product development language. And so it's a nice place for us to be, it's an area that's vastly under connected to the broader organizations, at least at most companies for sure.


John: Yes, that is a great idea. Yes, definitely, Michael. I think if you if you've been listening as podcast and you heard that idea, that idea was worth the whole 30 minutes you've invested because, yes, that is a really, really good idea. You can imagine actually, that, cause a lot of customer service reps now have chat bots that help guide them through the, you know, the experience that if you're typing in or maybe even just having transcribed what the customer is saying, it's possible that chat bot could be informed by the sensory research and could prompt questions that would tend to extract information like would you say, you know, to prompt the customer service agent to say, so would you say that it was bitter? Was it too bitter? You know, or whatever you were gonna help to try to get, I mean, is that the sort of thing you're talking about, trying to take this general statement and put it through some sort of decision tree to figure out what's actually wrong with the product?


Michael: Yeah, more or less. Right. Like, I mean, I built flowcharts. And building on the bitter example, though, right? Like when a consumer says bitter, if that gets trans given directly to a product developer or a quality scientist, it says, go solve this, what you could end up happening is that they're going towards the kind of the basic taste of bitter. We're in reality, the consumer, maybe it has too much grilled notes, has nothing to do with bitterness. And maybe it's too acidic. And they just started to use the language bitterness. Or maybe English isn't their first language. And bitter is this catchall bad phrase, right? So it could be all these different things that could lead you down passed that they are a waste of time to go down. If you just translated exactly as bitter and in too much, I see the technical people try to take consumers literally. But that said, like, it's not an easy thing for a customer service agent to do because they have to diagnose the temperament of the consumer. They want to make sure that, you know, if the consumer says it's bitter and they're not challenging them and saying, well, it's not bitter, it's acidic or whatever it is. And so it's a challenging problem, but it's an interesting one where, again, like, it's not traditional sensory science, but certainly the tools and point of view we have is an area that can help out there.


John: Yeah, that's fantastic. Excellent. Alright. Now, amazingly, we're out of time, Michael, this half hour has flown by. So really appreciate you being on the call. Let's talk a little bit of how people can get touched you, suppose they want to hire MNC or they want to follow up with questions after this episode? How should they reach out?


Michael: The best way to reach out is go to my website, it's mnestc.com. Also on LinkedIn. I haven't posted an article recently, but I post articles there occasionally and I'm fairly active in the LinkedIn community. Those are the best places. Give me a call, my phone number, you're gonna laugh at this, it's 209-SEN-SORY.


John: I recall that from a long time ago.


Michael: I know, I snatched that Google Voice number up way before I had an idea to start a consulting business.


John: And you're okay with putting that number on because we've had the show notes that go out. We'll put your online contact information. Do you want your phone number going in the notes or would you rather?


Michael: Absolutely. It's all over my website.


John: Okay. Alright. There you go. Excellent. Any parting words of advice for the young sensory scientist, someone who just graduated, what should they be thinking about for the next couple of years?


Michael: What should they be thinking about for the next couple years? Stay engaged with the community. I think the hardest times that I've ever had as a sensory scientist is when I've got disconnected from the sensory community a little bit just to do business pressures, job roles or whatnot. So I would say stay connected, because that's what's going to keep you in sensory and it's going to feel you.


John: Excellent.That's great, Michael. Thank you so much. It has been really insightful.


Michael: Alright. Thanks, John. I appreciate it.


John: Okay. That's it. Hope you enjoyed this conversation. If you did, please help us grow our audience by telling a friend about AigoraCast and leaving us a positive review on iTunes. Thanks.


That's it for now. If you'd like to receive email updates from Aigora, including weekly video recaps of our blog activity, click on the button below to join our email list. Thanks for stopping by!



©2020 Aigora