← 返回

2026.09.28 | TB | Industrials | PepsiCo Water Treatment

原标题：Water Treatment Industry & PepsiCo – Former Senior Executive, Supply Chain at PepsiCo Inc

#industrials #consumer

Third Bridge · 2026-09-28

Client-led • 19 Aug 2026

Companies: PEP US

Expert Former Senior Executive, Supply Chain at PepsiCo Inc

Agenda

Questions

[00:08:55] We were trying to learn more about the water treatment industry. What was PepsiCo using water treatment solutions for? What went into the decision-making for when you needed to use it?

Expert [00:09:32]: We were using water treatment solutions in two broad areas. Actually, I would say three, but one and two are kind of similar. One was the incoming water that we received either from the city or from bore wells. In a couple of locations, we had our own bore wells. Out of the 65 facilities I mentioned, for the most part, we received water from the city, a little different from internationally. Internationally, you either take water from a river. Sometimes it's from the city, but oftentimes, it's not directly from the city. You either take it from a river or you have your own bore well. Now the water that the city provides you, even though it meets drinking water standards, according to the WHO, it does not meet the ingredient water standards. I'm differentiating between the two, drinking water and ingredient water, it does not meet ingredient water standards for a beverage company. The reason for that is that the water might contain some chemicals. It might contain, for example, a little bit of iron, a little bit of sodium, a little bit of potassium, some sulphate, some nitrate, which are all reasonably okay for health. They fall within the WHO mandate, within limits, of course, but if they are those contaminants within those regulated limits, they are actually perfectly fine for health. You won't get sick drinking that water.

However, those contaminants react with our formulas. When I'm making Pepsi-Cola, my Pepsi contains certain concentrate. If I have nitrate or sulphate or potassium or magnesium or iron, it affects that concentrate and therefore, affects the taste. The water that comes out of the city to us, even though it is good for drinking water, it is not good for ingredient water. We need to take out whatever those little chemicals are in order for it to perform properly with our concentrate and produce the product that we need to sell. That's one application. The second application comes in process and utility. We have boilers, compressors, cooling towers, stuff like that. Those are mostly for the most part. We are getting away from it now as part of the sustainability programme, but for the most part, those machines are water cooled. Water cooling means you're recirculating water. Obviously, it's not a once-through process. If it's once through, you'll be wasting a lot of water. You're recirculating water. As you're recirculating water, some of the water because of the heat generated by these machines, some of the water is getting evaporated, so the concentration of chemicals that is getting left behind, chemicals don't get evaporated, so whatever is in the water, over time, as you keep recirculating and the water part gets evaporated, the chemicals left behind, they start forming a scale.

When they form a scale and they deposit in your machine and in your pipeline and your cooling towers and so on, the effectiveness of everything drops. You consume more energy for less output. You need to treat the water to prevent the scale formation. The third area is in wastewater treatment. We discharge a lot of wastewater as part of our ongoing operations. Now waste stream water contains sugar because our product is sugary. It contains acids because our product contains some acid. It contains cleaning chemicals because we clean our equipment from time to time. It contains lubricants. We have machinery and then we put oil and grease inside the machinery. Sometimes there's a little sloppy operation, some of that gets in. It contains other contaminants like lubricants for our conveyors, definitely from the facilities, the locker rooms and the kitchens from our factories. Sometimes it contains solid materials like an errant cap, an errant bottle, an errant can, all of that goes in.

All that water with all this stuff mixed in is not acceptable to the local municipal authorities, so we need to treat that water to meet certain standards before we can discharge it to them. That is the second application. The applications are ingredient water, process and utility water and wastewater. I mentioned that one and two are kind of similar. What we are trying to do is instead of having a water stream for ingredient and having a separate water stream for process and utilities, we have been examining and implementing that why can't the ingredient water go into the process and utilities. It's a little bit of an overkill, but there are benefits to doing that because ingredient water is very clean, so there's very little chemicals in that. You can actually do many more turnarounds in the process and utilities before you need to blow down that water, whether it's a boiler or the compressor or something. There are economics around it, but there is definitely a business case where you overtreat the water, bring it to ingredient grade and even then use it for process and utilities. One and two are kind of similar, even though in many places, we do have a separate system for ingredient, separate system for process and then a separate system for wastewater.

[00:15:23] What are PepsiCo's most important selection criteria when evaluating water treatment vendors? Is it reliability, cost, water recovery, energy usage or something else?

Expert [00:15:45]: All of that. I think the most important one is technology because what's happened over time. If you look at what's happened to the beverage business, there was a time when there were maybe three or four products. If it was a Coke or a Pepsi, then there was a Diet Coke or a Diet Pepsi, there was a Mountain Dew or a Sprite. There were like five or six. You could count the products on the fingers of your hand, five or six products. All these products were very robust. When I say robust, they had strong flavours, strong taste, strong aroma, etc, etc. Your ingredient water quality was not that critical. For example, let's say, at that time, my specification for water was 250 parts per million of external contaminants. The 250 parts per million played quite nice with Pepsi concentrate or Coca-Cola concentrate or Mountain Dew concentrate, it played quite nice. There was no problem. Over the years, what has happened is our products have moved away or the consumer taste has moved away from these robust products, so they still drink Pepsi and Diet Pepsi and Mountain Dew and Coke, but they also drink tea. They drink flavoured waters. They drink table water. They drink coffee, they drink juices.

Now all of these products, particularly not the juices so much because juices are also quite robust, but all the others, the flavoured water, the energy drinks, the coffee, the tea, the table water, all of these are extremely sensitive to external contamination. What has happened is when once upon a time, 250 parts per million of contaminants was acceptable because your product was very robust. Today, if you look at table water, for example, if you look at Aquafina, which is the Pepsi's table water brand, the limit is 10 parts per million. Over time, what's happened is the quality requirement and the cleanliness, if I use that terminology, the cleanliness of the water required for making beverages has become more and more stringent. Now obviously, that involves technology. There was a time when we were doing lime coagulation and then we moved to carbon towers and then we move to filtration, nano-filtration. Now we are into RO. We've got ozonation. We've got all kinds of steps in the process. One of the most important things that we require from a vendor is an understanding of the technology and understanding of the water quality that we desire and an understanding of what contaminants are coming in from the city or from the bore well and how to get it out to meet our requirement on a long-term consistent basis.

I don't want a production run that lasts for two weeks and then now I got to do something about it. I'm looking for a technology that I can put in and I can run for years, maybe with some planned stoppages in between to do some cleaning or disinfection or something like that. That's the first criteria, the most important criteria in a vendor that they must have the process know-how to be able to treat our water. The second thing is they need to have a track record. If it's a vendor that came, he may have the know-how, but if he was only born yesterday and doesn't have a track record with similar companies. When I say similar companies, I'm talking of the likes of other beverages companies, Schweppes and Dr Pepper, Coca-Cola, even the beer companies like Anheuser-Busch or Molson Coors or somebody like that. They need to have a track record of having worked with large companies and delivered results over a period of time. That's the second criteria. The third criteria is, the way Pepsi looks at their vendors, we look at relationships. We look at a player who we want to work with, and we try to build a relationship, so they are the go-to. You can almost say it's almost a no-bid contract among three or four players.

If I need water treatment, for example, I'll have three or four companies that I have shortlisted that have done work with me that we have vetted in the past from a financial point of view, from a management point of view, from a legal point of view, regulatory, no child labour and that kind of stuff, we vetted these companies in the past. They have a track record. They have the right technology, so they have a good R&D centre, etc. They have scale. I don't want a company that has a revenue of USD 5m per year because my projects are much larger than that, so they have scale. They've been around for a while. Then they form into the pool, and they come into a pool of vendors that we normally deal with. When I go out to bid for a contract, I generally bid within that pool, four or five vendors. I give them all my requirements, my RFP, they come back with proposals then I pick among them. That's the second one. A track record, financial stability, scale, ability to service big companies, meeting regulatory obligations and stuff like that. The third thing I look for is customer service. This is the relationship that I was talking about. I want you to do the project for me again and again and again and again.

Because you've done it the first time and the second time and you've learned from it, you know how I want it done, so the third time, I don't need to teach you. I just need to tell you, "Hey, I need a water treatment plant. This is my water quality, incoming. This is my water quality that I desire. This is the capacity of the plant. Just go and give me a quotation. You know how I want it done. You know the individual process steps that I require, so just go and design a plant, give me a proposal and then we'll go from there." I'm looking at consistency and repeat kind of business. That's customer service, is doing it again and again and developing a relationship that makes me trust them and want to go to them without really casting a wider net and asking for more providers. The fourth one, of course, is price and delivery. Delivery, not so important because if I'm doing a major project, let's say, I'm building a new factory, the delivery time for construction of the building and purchasing my manufacturing equipment and so on is generally longer than the delivery time for a water treatment room, so delivery is not that important if I'm constructing a new factory. It is important if I've got a problem in an existing factory where my water treatment is acting up and I need it modernised.

Particularly if it's acting up and it's causing me quality issues, then sometimes I'm down or sometimes I'm running at less than full utilisation and then delivery becomes important. I need to fix the problem as quickly as possible. Price definitely is a factor. We look at price from a capital investment point of view, and we also have a metric which we call the total cost of ownership. Total cost of ownership, we look at it over a 10-year timeframe. I say in a 10-year timeframe, what is the system going to cost me? There's initial investment, of course, initial investment, installation, training, start-up, etc. Then there's ongoing cost, chemical consumption, energy consumption, consumables like membranes and filters and stuff like that, that kind of thing, downtime, which involves maybe labour. Definitely, all of these things involve labour, maintenance involves labour. We look at how much is going to be the maintenance requirement on an annual basis? How many parts and how much labour? We price that. We do all of that and then we look at a 10-year window.

We say in 10 years, this system is going to cost me X to buy and operate. In 10 years, this system is going to deliver to me W gallons of water. What is the cost per 1,000gal or per one million gallons over a 10-year timeframe. That is one of the metrics we look at. You mentioned some of the items, pricing and energy and so on, that all gets captured into that metric, chemical consumption, energy consumption, consumables, maintenance, initial capital investment, lubricants. Lubricants, of course, part of maintenance. The number of the frequency and the number of cleanings that I require, disinfection. There's obviously chemicals involved and time involved in those factors. All of that gets captured in the metric we call the TCO, the total cost of ownership. That's one of the parameters we use to compare vendors.

[00:24:23] Do you have experience interacting with Gradiant? How does it compare with water treatment vendors such as Veolia, Aquatech and Xylem?

Expert [00:24:55]: As our water quality has become more and more stringent, we have moved away from the older processes, which was basically lime coagulation. We moved from lime coagulation to simple filtration to nano-filtration to reverse osmosis. Our technology has also evolved over time as the water requirement has become more stringent. Now over time, sustainability has also become a very, very important factor. Let me set a baseline. When you come to reverse osmosis, if you go back, let's say, 15 years and you did a reverse osmosis membrane, for every 100gal of water you put into the system, you got about 85-87gal out as product ingredient water. The remaining 13% to 15% to 17% went to the drain as brine. Basically, for every 100gal of water you were taking in, you were wasting about 13-15gal of water. This obviously was an ongoing cost, but more importantly than the cost, it was a sustainability problem because everyone was complaining that, "Hey, you guys are taking water and you're wasting water." If you just look at the ratio, 100% going in, 87% going forward and 15% going to the brine, that's a 15% wastage.

Now over time, the water treatment companies that we were dealing with, the Veolia, the Xylem, the PureChem and so on, these companies started improving their membrane technology. They started putting recovery membranes and double membranes and stuff like that. We were able to improve the technology to almost, in the best case, 5-7% wastage, in the worst case, 8-9% wastage. Obviously, it was a function of the quality of water coming in because if your water quality coming in is not the best, obviously, when you're extracting good water, the ingredient water, there is a concentration that's building up in the brine and there's a limit at which you can build up the concentration. After that, you have to discharge it. Otherwise, your membrane's performance starts getting affected. Let's say, 7-8% wastage. We were able to drop wastage from 15% to 7-8%. Now that was where it stagnated. That was the sweet spot between number of cleanings that were required, the membrane contamination, which required replacement of membranes, which is not cheap, downtime requirement and the quality of water and sustainability. If you take all of those things together, that was the sweet spot, about 7% wastage, improved significantly from 15% to 16% wastage. Then along comes Gradiant. Gradiant came with a technology, and they said they have a technology. It's a series of different kind of membrane, different operating pressures and so on. They can get me into the 3-4% wastage.

I had already gone from 15% wastage to 7%, so I had improved my waste to drain by a factor of 50%. They came and said, "They can improve it by yet another 50%," which was awesome. It helped a lot with the sustainability. When they came to me, they gave me a quotation for a complete water treatment system. I found that that was a little bit expensive compared to the companies that I had been dealing with. They were a relatively new player. Again, what I mentioned in the beginning of the call. I need somebody with a track record, somebody with whom I've done business and so on. That was another concern that they are a new player. If I give them a complete water treatment system, which sets me back USD 6m-7m, are they going to be able to perform or not? If they don't perform, there's a big problem for us because I've set my planning on that basis.

For them, they had a technology which they believed it worked. They had done some lab-scale trials and some small-scale trials with some other companies, so they had a technology which they believe that worked. They had not tried it out on a massive scale in the beverage industry, and they had not tried it out over an extended period of time because some of the things they might work at the upfront, but like I mentioned to you, membrane degradation. If you put more and more concentrated brine into your membrane, the membrane degrades and membrane replacement is not cheap. There are ways around it. You can put chemicals to slow down the degradation, but now you're consuming chemical cost and energy and so on.

[00:36:15] How sticky are these vendor relationships? Once a provider successfully delivers a project, how likely is PepsiCo to reuse that provider for the same site, additional phases or across other facilities globally?

Expert [00:36:38]: Fairly sticky, I would say, because like I mentioned earlier, and it was not one provider. We have a choice of two or three providers. We obviously need two or three for strategic reasons because we don't want to have one in case something goes sour with our relationship. We also need to keep them on their toes. We don't want one provider to feel that he's got the business so he can charge us anything. Strategically, we always had two, three, sometimes four providers, depending on the technology we were talking about. How sticky? I think they were very sticky because like I mentioned, we spent a lot of time and effort bringing a provider into our family. We used to do a lot of due diligence, not just technical due diligence, market research, technical, of course, we used to ask them for references and talk to their other customers and performance with other customers, look at their track record. Even our legal team used to look at their business structure, look at if they were having any legal issues, that kind of thing. Our finance team used to go in and look at how solvent they were as a business, if they had had any issues in the past, our legal team looked at if they were having any issues with the regulators which could affect their performance and therefore, indirectly affect our performance. We used to spend a lot of time and effort.

From the moment, let's say, a vendor approached us and said, "Hey, I want to be your vendor." The time where we even agreed to give them a small trial could be six months. Once they got there and then, of course, after they got the first job, which typically would be a smaller job, it will be a pilot. After they got the first job, which would be a pilot. I might give them a USD 0.5m job and then I might give them a USD 2m job and then one day, I might give them a full-scale water treatment, which is USD 5m-6m. There would be learning along the way. They would do it and then we tell them, "No, no, no, that's not what we wanted to do. We wanted you to do it this way." There is learning on both sides. We understand how they operate. We understand their capacity, capability, technology, the quality of their people. The quality of their science, process engineering. They understand what we want, what things are dear to us and what things we compromise on or willing to compromise on, what things we are absolutely not willing to compromise on, how we want the system set up in terms of what are our testing parameters, where we want instrumentation, what type of operator screens we want?

There's a lot of learning that goes along the way. Once you get a vendor in and you do the pilot and then you do a bigger scale and then you do a still bigger scale, they are learning as they go and that takes time. That takes multiple projects, and that makes them sticky. Unless they do something ridiculous, they do something completely stupid, they botch up a job completely and they don't fix it or they're not willing to fix it or unless they are exiting the business, they say, "You know what, Pepsi, you guys are too tough. We don't want to work with you." That's a different thing. If they work with us and they do a few projects and they do them well and they start to develop a track record, they become very, very sticky. Not one vendor, like I said, so there are three or four vendors who are very sticky, who have done it with us a numerous times. Whenever there is a new job coming up, I will bid those and only those three or four vendors. Among them, I will decide who I want to award to.

[00:40:38] Did water treatment ever become a larger or smaller portion of PepsiCo's budget? How do you expect that to either grow or shrink?

Expert [00:40:56]: Yes, water treatment became a larger portion of our budget because it was driven by our product quality requirements. Like I said, once upon a time, if we were not making Pepsi, we were making Diet Pepsi. That's all we were making. The water requirement for that was not as stringent as the water requirement for the products we make today. That is one. The second thing is the plants that we have, some of my factories are 50 and 60 years old. Some of my water treatment systems are 35 and 40 years old. Water treatment is not a very corrosive environment, so to say, there are only some portions of the water treatment where you're handling chemicals and they could be corrosive. Then mostly, it is all stainless steel. The water treatment systems don't degrade so rapidly. Our equipment degrades in about 15 years. Water treatment systems tend to degrade in about 25-30 years, but we were getting to the point. The technology was changing. The quality of water required was changing. We basically put up a programme, which said, "Okay, in the next 10 years, this is what our product requirement is going to be.

This factory today produces only Pepsi, tomorrow is going to produce Pepsi and flavoured water, for example." We had a 10-year timeframe. We were looking ahead in time, and we were saying, "Okay, what does the map look like?" Obviously, that was driven by where our consumers were, which products we were talking about and stuff like that. We had a map of this is our footprint today in terms of what gets manufactured where and what quantity. This is what the footprint looks like three years from now, five years from now, 10 years from now. Of course, 10 years was very far to predict, but 3-5 years, we were able to predict. This is what it is going to look like. These are the products that are going to be made in these factories. These are the quantities of products that are going to be made because that defines the water requirement. Something had to be done about them to get their water quality up. Then we did a mechanical assessment of what was the condition of the existing water system. We prioritised that, five years from now, these are the 12 factories that I need to address. Out of the 12 factories that I need to address, currently, these three factories are in really bad shape, so I'm going to address them first.

Then next year, I'm going to do the next two. The next year, I'm going to do the following three and so on. We set up a programme which was based on what quality and quantity of water was going to be required, overlaid that with what was the condition of the existing infrastructure and then put a programme together. This is an ongoing thing because if you think about it, we don't have unlimited capital. It's not like the company says, "Okay, go fix all the 65." I told you it's about USD 6m-7m per treatment system. Now if you think about it, I've got 65 plants. I'm doing 2-2.5 plants per year. If I do my first plant today, by the time I get around to it again, it's going to be 30 years from now. It's an ongoing business. 30 years from now, it's going to be degraded because I did it today. I'm doing 2.5 per year. I've got 65 plants to go through, 30 years, 25 years. It's about time to do it again.

I see that as an ongoing business. Water treatment, remediation of degraded assets and modernisation to include newer products, I see that as an ongoing process. Then in addition to that, from time to time, we build factories. We find out that the total capacity we have nationwide does not meet our total sales requirement nationwide, so I got to build a new factory or I got to expand a factory. Every time I do that, I need more water. Those peaks and valleys continue to happen. I might build a new factory every two years or three years and then I'm talking large water treatment investments. I'm not talking USD 4m-5m, which is addressing the degradation or modernisation. I'm now talking USD 15m-20m each time that happens, a new factory. I see the business as being there to stay. The other area where there is a lot of emphasis on water, like I mentioned, was sustainability, so putting in systems where you're not putting water in the drain and there's a lot of water recovery processes. The sustainability is a huge impetus to corporations to improve the sloppiness around their water consumption, but that's a growing area. There's a huge market out there for that even today.

[00:46:01] You mentioned that you typically use a few different vendors. Did any vendor do a better or worse job than the other? What were they delivering on? Where did you feel more favourably?

Expert [00:46:34]: I think the vendors we had were pretty comparable in their capability to deliver technology. Obviously, the size of the organisations was different. I had four or five vendors. Technologically, they were similar. If they had lacked the technology in working with us, they developed it. I was not too concerned about that. What I can say is that from a financial point of view and the ability to take on multiple jobs or from an R&D point of view and the ability to take on an R&D, let's say, I'm just being facetious here, but let's say, I talked about the 250 parts per million of dissolved solids. I said our table water is 10, that's the limit. Let's say, I came out with a product and I needed three total dissolved solids. There's some research to be done in that. Let's say, I said that I want to improve my brine to waste from 5% to 2% or 1% or let's say, I want to invest in technology that gives me almost distilled water quality for whatever reason, EDR or something.

That is where some of the bigger vendors, they had the financial wherewithal to do it. Some of my smaller vendors basically would say, "No, that's not up our street. That's not something we are focused on. Frankly, we don't have the money or the talent to invest in that." That was the differences. Then there were a couple of vendors, which said not only they will provide us a system but they will put in a mechanism by which they can remotely monitor our system. Of course, there's a retainer fee you got to pay for that. They were willing to remotely monitor your system to make sure that the performance was meeting design specification over time. They would put in their PLCs and their modems and so on, and they would capture data from the system that we were operating. They were looking at pressure drops and conductivity and chemical usage and energy usage. They were looking at all those things, gathering it into a PLC, which was in our plant, but that information was being transmitted over the cloud to them, and they were tracking the information at their end. If they were finding that there was a drift, something was starting to drift, in some cases, they had the ability to remotely fix the problem.

In some cases, they would call us and tell us, "Hey, this is going wrong over here. You might want to go take a look, this pump is drawing too much power or the chemical specification on your finished water is starting to drift, so you need to go and take a look at what's going on in that." Then some companies actually offered us water as a service, which I did not take, but they at least offered us a long-term water management programme. They said, "You give us a retainer, and this is what we estimate." The TCO that I talked about, this is what we estimate your water treatment system is going to cost you YoY, year one, year two, year three, year four with membrane replacements, chemical consumption, energy consumption, etc, etc. They said, at least for the parts and maintenance portion of it. Disinfection chemicals, cleaning chemicals, membranes, they said, "You pay us X dollars a year, and we will take care of that for you." We will come whenever it's time to disinfect your membrane, we will send somebody out. He will do the disinfection. We will provide the chemicals. You, of course, provide us the water and the energy because that's fixed installation, but we'll provide the chemicals, we'll provide the manpower.

Let's say, your membrane performance used to be 100%, it had degraded to 85% or 80%. When we come back and we are done, we'll get it back to maybe not 100% because membranes do degrade over time, but we'll get it back to 97-98%. We'll do that over a three-year, five-year timeframe, and here's the cost we'll charge you per year for it. Sometimes it could be a total cost for three years divided into three equal instalments or 12 equal instalments because we paid them by quarter or it could just be that this year, we are going to charge you so much, next year, so much, next year so much, and then we had an agreement. Some companies offered us that in addition to just putting a system in and commissioning it and walking away. That's ongoing support. Then some companies offered us water as a service, so they said, "You don't even have to make the capital investment. We will make the capital investment. We will run the water treatment for you. You need to give us a space in your property where we can put our system. We'll run the water system for you. We'll give you the quantity of water you require at the quality you require with a guaranteed service level agreement. 99% of the time or 98% of the time, we will meet your requirements. You pay us X dollars per 1,000gal of water."

They offered us and we didn't go down that path because my finance people felt that the price that they wanted to charge for it, we could do better by doing it on our own. I mean water treatment is not so complicated. If you've designed your system correctly and you've installed it correctly and you operate it correctly, it's not such a complicated business. I don't need an expert to come and do water treatment for me. The only reason I would do that is if I was trying to defer or avoid capital investment. Pepsi was a AAA-rated company. We could borrow money at rates that the other companies could not. Obviously, if they invested in our facility, they were going to charge us an interest charge, that was going to be part of the financials. We always found that we could do it better by investing on our own rather than having them invest for us. The other programme, the one where they would monitor and remotely monitor, remote remedy and that kind of thing. In fact, the other programme where they said, "You pay us X dollars per year and we take care of everything," we did sign up a few factories for those programmes.