Impact Pricing Podcast

Ep123: How Price Elasticity Affects Consumer Behavior with Anshu Jalora

 

Anshu Jalora is the Founder and Managing director of Sciative Solutions Private Ltd., used to be a Director of Pricing Strategy at Starbucks, Pricing Scientist at PROS, and Director of Pricing at Overstock. 

In this episode, Anshu talks about how true price elasticity brings about change in customers behavior and creates great impact in business. He shares about what his company, Sciative does at optimizing pricing, simplifying solutions from massive data to generate insights in lightning-fast speed necessary for decision making needed to make a maximum impact to the business. 

Why you have to check out today’s podcast:

  • Find out what Sciative do to help businesses maximize value in the massive data available to them 
  • Find out the two crucial things that make a huge impact in a customer’s traffic pattern 
  • Learn how to achieve bringing in customers from low-end products to high-end products with the understanding of price elasticity and consumer behavior 

         

“Focus on three V’s – Value, Variety, and Velocity. As you’re looking at your pricing processes, focus on these three V’s.” 

– Anshu Jalora 

      

Topics Covered:

02:17 – An interesting story related to his Ph.D. story 

03:30 – How he got into Pricing 

04:07 – What made him like Pricing 

06:09 – Success stories in B2C in terms of applying scientific models for pricing 

08:52 – Capturing value through individual customer level or collection of customers together 

09:32 – Talking about the coined word Sciative 

10:39 – How can Sciative help businesses truly find value in the massive data that they have 

13:36 – Price elasticity and consumer behavior 

15:26 – Two things that basically make a huge impact on a customer’s traffic pattern 

21:20 – His thoughts on the criticism against price elasticity 

23:55 – Playing with different types of mathematical models in coming up with true price elasticity 

24:58 – What are Walmart parking lot products 

28:26 – A pricing advice he shares that would have a great impact to one’s business 

     

Key Takeaways: 

“Pricing is a very powerful tool. It’s educational for customers. Sometimes we guide customer behavior, we help customers figure out what is right for them. We make a direct impact on the business.” – Anshu Jalora 

“What we found used to work best was that you don’t have to take a blanket pricing approach everywhere, there will be some products where you will have more opportunities of increasing pricing without affecting customer sentiments and in some others you will not increase customer prices. You may even decrease prices to make the entire portfolio look competitive enough.” – Anshu Jalora 

“I fundamentally believe that if you change prices, customer behavior will change and that is elasticity for me.” – Anshu Jalora 

“Some very misleading information also get picked up by simplistic models, if we do not get into the nitty gritties and details and separate out seasonality effects, separate out cannibalization effects, cross price elasticity effect, we won’t be able to compute the true price elasticity. And this is where elasticity gets criticized a lot. And that’s what we try to avoid when you’re educating customers.” – Anshu Jalora 

      

People/Resources Mentioned: 

        

Connect with Anshu Jalora: 

        

Connect with Mark Stiving:   

   

Full Interview Transcript

(Note: This transcript was created with an AI transcription service. Please forgive any transcription or grammatical errors. We probably sounded better in real life.)

Anshu Jalora   

Focus on three V’s. And these three V’s, I would say, are Value, Variety, and Velocity. As you’re looking at your pricing processes, focus on these three V’s. 

[Intro] 

Mark Stiving   

Today’s podcast is sponsored by Jennings Executive Search. I had a great conversation with John Jennings about the skills needed in different pricing roles. He and I think a lot alike. If you’re looking for a new pricing role, or if you’re trying to hire just the right pricing person, I strongly suggest you reach out to Jennings Executive Search. They specialize in placing pricing people; say that three times fast.  

Mark Stiving 

Welcome to Impact Pricing, the podcast where we discuss Pricing, value, and the possibly elastic relationship between them. I’m Mark Stiving. Today, our guest is Anshu Jalora. Here are three things you want to know about Anshu before we start. He has a Ph.D. in Pricing and Revenue Management. He and I are going to talk about that for a second. He’s had many awesome Pricing jobs, including a Pricing Scientist of PROS, a Director of Pricing at Overstock, a Director of Pricing Strategy at Starbucks, and the Founder and Managing Director of Sciative Solutions. Welcome, Anshu. 

Anshu Jalora   

Thanks, Mark, for the warm welcome. It’s great to be here. 

Mark Stiving   

It’s going to be a lot of fun. Now, first off, you and I are together because a gentleman named Fabio sent me an email and asked to talk about price elasticity in a retail setting. And I’m not a fan of price elasticity. And I’ll explain why later. So, I reached out to the network of pricing champions to find someone who is, and there’s no doubt you are. So, this will be a great conversation. But before we jump into price elasticity, a couple of things I wanted to ask you. First off, is your Ph.D. truly in Pricing and Revenue Management, because mine is in Marketing, but I like to call it Pricing. 

Anshu Jalora   

I have a very interesting story relating to my PhD. So, when I was searching for a PhD topic, I happen to attend a seminar talk by Dr. Andy Boyd, who was the Chief Scientist at PROS that time. And he spoke about the mathematical models that are applied in airline revenue management. And that one, our seminar talk, completely changed my life forever. That day, I decided that this is an area I want to specialize in. And while my department of education was operations research, within operations research, I specialize in applying operations research techniques for Pricing and revenue management. So, my thesis was also focused on applying revenue management in industry settings. Nice, nice. 

Mark Stiving   

Nice. So, it sounds like a relatively similar story in that the school didn’t have a pricing and revenue management Ph.D., but that’s what your thesis was, and your research was and where all your work was. And so, is that how you got into Pricing? 

Anshu Jalora   

Absolutely. That’s how I got into Pricing. And I was very fortunate that after I completed my Ph.D., I got a chance to work with Dr. Andy Boyd himself. So, I went in, started my career at PROS, where he was the chief scientist. And along with him, then I designed the next generation of scientific algorithms for Pricing in various industries, including retail and many others. 

Mark Stiving   

So, when you got into Pricing, did you do it because you liked the mathematical models or liked Pricing.  

Anshu Jalora   

The reason I like Pricing was, this is an area where we can actually impact customer behavior. Pricing is a very powerful tool. It’s educational for customers. Sometimes we guide customer behavior; we help customers figure out what is right for them. We make a direct impact on the business. So, the richness of Pricing, both in terms of mathematical sophistication that needs to go into it. And the impact it has on the business and customer behavior is really powerful. And that was the reason I jumped into the area of Pricing, never looked back. And that’s how I spent the last 20 years of my career in the area of pricing and revenue management, and it’s changing really fast, the kind of scientific models that we use to develop some 15 years back and what we’re doing now is very different. And I see that this field will continue to evolve. As business complexities are increasing as customer behaviors are changing, as companies are collecting more and more data, the scientific richness in the models will keep on increasing. That is going to keep people like me and you occupied very well. 

Mark Stiving   

Yeah. When I was getting my Ph.D., I was doing the statistical approach and using logistic regression and creating some pretty funky models. It was a lot of fun. Absolutely. As I’ve been in Pricing, though, I’ve moved away from models. And I think the key reason, mathematical models, and I think the key reason is in the B2B world where I play most often, the bigger problem is that companies don’t understand what value means. And so it isn’t that we’re not using the data; it’s that we don’t understand what it tells us. We don’t understand how it gets there. But I can see completely if I’m in the retail space; the data is going to drive this a ton. 

Anshu Jalora   

That’s true. In B2C, there have been many more success stories in terms of applying scientific models for Pricing. More recently, we have started working in a very interesting industry, which is the television advertisements industry. And what we realized was that this is an area where rather than using scientific models, the company that we’re working for is the largest television network in India. They have 80 channels, 80 national level television channels, and this company is also owned by Disney, worldwide US. So, this is a subsidiary of Disney in India. And they were following a very rudimentary approach of contractual Pricing with their customers; there was no intelligence from changing customer behavior patterns and demand patterns in supply. And we noticed multiple gaps through the scientific model. And we have taken them on a journey. I won’t say that on day one, all your scientific models will easily get adopted here. But as gradually, you keep showing improvements with scientific models; they do start making sense here. So, when we started showing them that there are lookalike customers that have very similar demand patterns, but there is 40% variations in their Pricing. And they were like, we never knew this. And we also told them how to bridge this gap of Pricing between lookalike customers with the exact same demand pattern; it was a lot of value for them, being able to forecast demand many months in advance, so that they don’t do excessive discounting with the fear that they may not have enough demand in future that gives them the power, the PSI can go with confidence to my customers and demand price, which is right for my business. So yeah, that’s how it’s been. 

Mark Stiving   

I think those are great explanations. And I hadn’t thought of this. It’s just so obvious; I hadn’t thought of this before. Until I heard you speaking of this, you and I are, in truth, doing the exact same thing. We just do it differently, right? Because I spend my time trying to understand how buyers think and how they perceive value. And what you’re doing is trying to say these are the results of how buyers think. And when you put a whole bunch of buyers together at once, we can figure out what the segments are and what behaviors are going to look like. But we do it statistically as opposed to individually. And so, in truth, we’re really working towards the same goal or the same issues. 

Anshu Jalora   

Absolutely, there are many similarities here. The idea is to be able to capture the value that we’re delivering to our customers. And this value is all relative. So, if you can understand how much value my cohort of customers is attaching with my product, sometimes we need to go to the individual customer level. Sometimes we look at the collection of customers together, try to make sense of items in business, and apply some mathematics to understand things better. 

Mark Stiving   

Yeah, nice. Okay, so, before we started, you told me what Sciative stood for. But I want to make sure that the audience knows, so tell everybody else. 

Anshu Jalora   

So, what we realized was that there are many people who are scientists and this community of people are really disconnected from business. They will do an amazing job in terms of coming up with scientific research. And then we realize that successful businesses are run by highly creative people who have creativity in their business models with creativity in their approach of solving the same problems but in different ways. So, we said we will create this next generation of scientific product, which will be backed on some very creative new ideas, looking at scientific models in a whole lot of different ways. And we combine the two words scientific and creative to create Sciative. So, Sciative is a combination of two words and it truly reflect the ethos of our working approach, where we combine scientific rigor with business creativity. And that’s how we create Sciative solutions. 

Mark Stiving   

Nice. Take 30 seconds and give me an advertisement for Sciative. What do you guys do? 

Anshu Jalora   

So, we know that Pricing is very complex. And with more data becoming available to businesses, where they spend huge number of hours trying to decipher data, we feel that in order for businesses to truly value from all this data, we need solutions that apply artificial intelligence at a massive scale at lightning-fast speeds, so that businesses can get maximum value out of their data in the shortest amount of efforts and times. And that’s what we do with Sciative. We create artificial-intelligence-powered automated solutions to simplify this complexity in Pricing and give the decision-makers insights and relevant insights at lightning-fast speeds to make maximum impact in their businesses. That’s what we do with Sciative. There are multiple industries where we have created such types of automated solutions. So, we realized that like I gave you this example of the television industry, we also created one of its own kind of solution in the passenger bus industry. So many countries, intercity bus transport is the major source of transport in those countries. If you look at Europe, there is a huge usage of buses for intercity transport Asia, Southeast Asia, Latin America, Africa. There is a very big market for buses. And we realize that this is a very complex problem. The variability in demand is very high; customer behavior is very dynamic. Demand for travel can change very fast. And there was no commercial-grade solution available in this particular domain. So, we created one of its first kinds of solutions in the bus travel industry where every 15 minutes, our solution recalibrates itself on demand, demand forecasts, it updates itself on the elasticity of demand that exists in the market, at every 15 minutes, it re-computes optimal prices. So, it sounds very simple. But in the bus industry, you’re doing close to 3 billion price optimizations every day. 

Mark Stiving   

Wow, that sounds pretty impressive. And so, since you brought up the word elasticity, let’s talk about elasticity in the retail market for a second if that’s okay. Now, in the things you’ve described to me so far, AI, I’m going to change prices. If I’m going to change prices every 15 minutes, then my complaint or issue with price elasticity may have less substance behind it. But let’s start with talking about, let’s use Starbucks. It’s used to work at Starbucks for a second. Did you do price elasticity studies at Starbucks? 

Anshu Jalora   

Yes, we used to do price elasticity studies. But we need to understand customer behavior again. See, in a place like Starbucks, there are multiple products that are competing with each other. So, saying, here is an absolute price elasticity for a single product doesn’t make any sense. So, if say, for example, if I drop prices of my tall latte, will it have an effect on my, if I’m also dropping the price of my tall brewed coffee, if I’m doing both of these together, I may actually not see any substitution effects. However, if I’m dropping prices of one and not dropping the prices of others, I will start seeing substitution effects. So, in a model where there are multiple options available to customers, you would not compute the price elasticity of individual products. In that scenario, what we do is find out how my likelihood of someone purchasing product X changes when prices of a collection of products are changing together. 

Mark Stiving   

Yeah, that makes a ton of sense. So, the cross-price elasticity of the other products that we offer, not necessarily competitive products because it really is interesting. Now to me, I think the bigger question, obviously, it’s a huge question for you to know, what products are people going to buy? And how much margin revenue dollars am I making on each of those and so like want to influence the purchase selection? To me an even bigger more important question is how do I know if someone’s going to walk into Starbucks or a competitor’s coffee shop? 

Anshu Jalora   

Okay, so well, there will be seasonal patterns of customer traffic, that is something that will help you understand what kind of customer traffic will like it. And then there is a pattern of traffic which you would have observed more recently. So, this is if I’m not touching Pricing at all, even if my competition is not touching Pricing at all, and my business is fairly stable. And if I’m looking at my same store traffic year over year, the patterns will be very similar. I may see an increase of maybe 2 to 3%, or a two to 5% increase in my traffic year over year, but the seasonal patterns will remain very similar. Now, when you bring in this additional element of Pricing, now, here, the price of every individual product will not play as much role as in changing your traffic patterns, there are two things that basically make a huge impact on customer’s traffic pattern. One is, every business will have certain key products. And Walmart likes to call this parking lot survey products. And I can talk more about what they mean by that there are always certain products that play on customers’ minds. So, if there are certain products where I have very high amount of repeat purchases from customers, and if I’m changing prices specifically for those products, it will affect their visit behavior. And now if my price is changing significantly compared to my competition, so far for customers, they have multiple options together, I’m not the only one, they sometimes probably are buying products from my brand, my store, sometimes they’ll buy products from my competition stores also. Now, relative Pricing definitely plays a role. But not every individual product price is important. There are certain products that drive perceptions of customers. So, businesses need to figure out which are those traffic-driving products, and second pieces, the overall ticket value, if my overall ticket value of customers increases significantly, without much changes in their customer or in their purchases. So, for example, if someone is buying a sandwich and a glass of coffee, if that combination, ticket price increases significantly compared to my competition, that would play a huge role. So, in such scenarios, what we found used to work best was that you don’t have to take a blanket pricing approach everywhere; there will be some products where you will have more opportunities of increasing price without affecting customer sentiments and insert some other, you will not increase customer prices. You may even decrease prices to make the entire portfolio look competitive enough. 

Mark Stiving   

Yes. Okay. So, here’s what I just heard you say, and I’m going to make stuff up for a second; let’s pretend that 30% of the customers who walk into a Starbucks buy a grande de latte. And I say that because that’s what I always buy when I go to Starbucks. And so, they know the price of a grande latte. And if we were to raise the price of a grande latte, we’re at risk of losing a portion of 30% of our customers because they buy that product, they know that price. And so, they’re watching that. And likewise, if we lower the price of a grande latte, we may be able to get people to visit more often. They may tell their friends; we may get more customers. So, we end up with more business, more unit sales, because we lowered the price of a grande latte. So, is that relatively accurate or not? 

Anshu Jalora   

Maybe yeah, relatively accurate. And see, when I’m dropping prices of my grande latte, I will also need to look at how customers are changing their coffee from brewed coffee to latte. How is that also getting affected by that? So, I will see some redistribution of customer demand patterns within my portfolio also. So, I don’t want to just blanket, lose prices or lose business or lose revenues on my grande latte, but I also want to make sure that some additional customers are buying my lesser-priced items and moving to a product, which is slightly higher. So, the migration of customers from low-end products to high-end products is something that you always want to achieve. 

Mark Stiving   

Yep. Perfect. I love this conversation. So, I think what I think my issue in price elasticity is more along the lines of, if I’m looking at elasticity of a single product against its competitive alternatives, because in every conversation that we’ve had, so far, I don’t see a problem at all with using price elasticity or cross elasticity in these retail settings. I mean, I think a lot of ways those make a ton of sense, just so that everybody knows, my issue with price elasticity, as I see many, many companies calculate the price elasticity of products, where they’re selling that product against a competitive product. And so, if they run a test, and they lower the price, and they get more sales, they say, well look at the price elasticity, we should lower our prices, where in truth, the competitors had never responded. And after they lower their price in more of a steady-state situation, competitors will respond, market shares will go back to normal, and we just hurt industry profits. And so that’s usually what I don’t like about price elasticity. 

Anshu Jalora   

And my thoughts are very aligned with your thoughts also. See, I fundamentally believe that if you change prices, customer behavior will change. And that elasticity for me if customers are changing their shopping behavior, if I increase my prices by let’s say, 20%, will I lose demand for that particular product? If the answer is yes, then yes, there is elasticity. Now, if I ignore the effects of cross product, or cross competition price elasticities, I will end up with very funny situations. So, most companies will increase prices a little before the start of big season. For example, at Starbucks, also in many markets, we use to increase prices just before the festive season. Now, in this situation, if you’ll notice, prior to increasing the prices, I was having lesser number of units at lower prices, I’ve increased the price. And now I’m even having more sales. So, some very misleading information also gets picked up by simplistic models. Suppose we do not get into the nitty-gritty details, separate out seasonality effects, separate out cannibalization effects, and cross-price elasticity, in that case, we won’t be able to compute the true price elasticity. And this is where elasticity gets criticized a lot. And that’s what we try to avoid when you’re educating customers that yes, there is elasticity, but we will compute elasticity in a very different way. But we will not try just to fit a simple regression curve and say that my slope is a good indicator of elasticity with some corrections. That’s not how we would go. We will try to factor in all of these multiple factors that add together having an effect on demand, separate out all those effects in the interest, leave out the true effect of Pricing. And that’s what will help us understand the elasticity point. 

Mark Stiving   

Okay, Anshu, the words you said a minute ago, you and I can agree 100% on, and that was if you change prices, it will change buyers’ behaviors, and that is elasticity. Right? I think that was perfect. And my biggest issue with elasticity is the formula that people use blindly, which obviously you and your team don’t do. So, I think this has been awesome. 

Anshu Jalora   

Thank you so much. This is something that makes Pricing so much fun because we get to play with different types of mathematical models. Sometimes we’re using multinomial logistic regressions; sometimes we’re using choice models. Optimization is also very rich. Sometimes we are using mixed-integer linear programming, and sometimes we are using stochastic dynamic programming approaches. So yeah, it helps us use that gray cells better and make some sense out of the data that we end up with. 

Mark Stiving 

So, Anshu, I love the last sentence you said, but most of our listeners are like, what the heck is he talking about? So, I do want to circle back real quickly on this last point, because you talked about the Walmart parking lot products. And I assume we’re talking about something like milk at the grocery store, the product that people go into the grocery store to buy because the price was advertised in the weekly flyer. Did you have to know what Walmart’s products are?  

Anshu Jalora   

I probably would not know the entire list. But what I’ve heard from many folks who have closely run this process is that many people will go and meet customers in parking lots as they’re exiting from the store and ask them to tell the prices of some of the products they’ve bought. And it usually is limited, the pricing knowledge of customers is limited to some 20, 25 products only. And like what you mentioned, milk is one of those bread, bananas, eggs, some basic items that people purchase at very high frequencies. These are the products on which the prices will be advertised in the weekly flyers also, and they’ll put big banners inside their stores also, $1 a gallon, those sorts of things you will see. It doesn’t mean that they will charge exorbitantly high prices and other products, they still want to make sure that the prices are within customers’ reach, but the margin rate will be much better than these so-called loss leaders. This seems to work very well when in the case of Walmart and we see the same concept gets repeated in multiple different industries in multiple different ways. So, this seems to be a very powerful concept, it has lived the test of time in multiple industries. And we try to identify which are these products. For example, when we were doing a project for one of the largest resort companies in India. This resort they are the largest in India; they have about 80 different properties in India. And they had a very unique and different type of problem. Not many customers were purchasing food and beverage items from their restaurants; they would either go out and eat somewhere else, they will not consume the food items within their own properties. And for this particular resort, food and beverage was a big source of revenue, and they were not growing that piece at all. And then we figured out that there were actually some products, which was driving price perceptions with customers. We figured out what those prices were, created some very nice attractive combos where they can go and talk about those combos with customers and show them that hey, guys, we have value offerings for you. And that suddenly changed the entire customer behavior. So yes, loss leaders, combos, these things really work out great in multiple different industries.  

Mark Stiving   

Nice. Anshu, we’re going to have to wrap this up. One quick comment, though, I’ve always heard that the product that brings more people into Walmart than any other product is diapers. And that would make sense to me because it’s a relatively high-priced and very frequent product. 

Anshu Jalora   

That’s true. In fact, there was another company, an e-commerce company. And there we found that men’s watches were the biggest driver for women traffic on that particular website.  

Mark Stiving   

All right, Anshu, the final question, always ends with this one. What’s one piece of pricing advice you would give our listeners that you think could have a big impact on their business? 

Anshu Jalora   

One piece of advice that I would like to give to listeners today is to focus on three V’s. And these three V’s, I would say, are value, variety, and velocity. As you’re looking at your pricing processes, focus on these three V’s. Value is try to find out areas where there is value leak happening in your business. And see how you can plug those value leaks. Variety, where it means try out something new, something different. Experiment with certain models and do control experiments to learn something from your customers that you have not already learned about them. And the third V is velocity. Have a faster learning pace so that you can fail fast and recover fast. So, these three V’s we try to educate our customers and pricing professionals in general. And that’s the advice I would like to give to your audience also today. 

Mark Stiving   

Excellent. Thank you, Anshu. Hey, thank you so much for your time today. If anybody wants to contact you, how can they do that? 

Anshu Jalora   

They can contact me on LinkedIn. Fortunately, I’m the only person with my name. They can easily find LinkedIn with my name. And they can also send me an email at anshu@sciative.com. 

Mark Stiving   

Okay, and that’ll be in the show notes, as well. All right, Episode 123 is all finished. Thank you very much for listening. If you enjoyed this, would you please leave us a review and a rating, if at all possible? And finally, if you have any questions or comments about the podcast or Pricing, feel free to email me at mark@impactpricing.com. Now, go make an impact! 

Mark Stiving   

Thanks again to Jennings Executive Search for sponsoring our podcasts. If you’re looking to hire someone in Pricing, I suggest you contact someone who knows pricing people. Contact Jennings Executive Search! 

Related Podcasts