Saif Hameed [00:00:00]:
I was speaking with a large pub group. Chicken is a big item that they sell. The consulting partner that they were working with had used one emissions factor for chicken. Whatever, the characteristics of that chicken all wash out, basically because there's only one emissions factor for generalized generic chicken that is going to be used. If that pub chain wants to do something to reduce the emissions of chicken, the only thing that they can do that will show up in the numbers is to buy less chicken and sell less chicken. None of the other variations of what they could be doing, none of that really shows up because there's one economic intensity based coefficient which says 100 pounds or €100 of chicken is x in emissions.
Isobel Wild [00:00:47]:
I'm Isobel Wild. Welcome to the State of Sustainability podcast, a show for professionals transforming corporate sustainability strategies, brought to you by Altruistiq. The majority of businesses calculating baselines use less than 10,000 emission factors. What this means your emission calculations are not accurate, and your decarbonization efforts don't always show up in your inventories. So just to kick things off, I'm going to give a quick explainer of what an emission factor actually is. In its simplest form, it is a magic number that gives you an average of how much greenhouse gas is emitted by a product or activity such as manufacturing or transportation. So, for example, it's how much CO2 is emitted for every kilogram of tomato grown. But Saif, I'd love to actually just dig into what is happening in the GHG calculation world today, and how are professionals and organizations using emission factors?
Saif Hameed [00:01:53]:
Thanks, Izzy. Maybe just to get going with, I found that one of the easiest ways to explain what an emissions factor is to a non sustainability audience is that it's an environmental coefficient. It is something that you are going to multiply against a unit of what you make, what you do, what you buy, etcetera. It is a coefficient in that calculation. I think that emissions factors have been a really arcane area of the space for most businesses. If you flash back, let's say six, seven, eight years ago, I would suspect that most teams in companies were outsourcing, thinking about emissions factors to consultancies. They were saying, look, if we're engaging someone to do our carbon baseline, we are going to give them a bunch of data from our side, and they're going to work out the numbers and methodologies and emissions factors. This is all their problem.
Saif Hameed [00:02:45]:
Incidentally, at Altruistiq, we actually often see this in the woodwork, basically like when we're inheriting an emissions calculation from another vendor or another partner. So we're coming into a company. That company's been working with a consultancy before. We are taking over the calculations that the consultancy has done and using that as a historic reference point. We often see that the consultancy was not prodded or poked or supervised on what methodology or what emissions factors they were using. And they just kind of assumed that no one really cares about this stuff enough. The main thing is completeness and on time completeness. The way in which that manifests is, let's take an example that I came across recently, chicken.
Saif Hameed [00:03:27]:
I was speaking with a large pub group, so they have hundreds of pubs around the country. Chicken is a big item that they sell. It's on the menu in many different forms. The consulting partner that they were working with had used one emissions factor for chicken. What that means is that whether this pub chain is using chicken thighs, chicken breast, chicken leg, whether it's organic, whether it's free range, whatever, the characteristics of that chicken all wash out, basically, because there's only one emissions factor for generalized generic chicken that is going to be used. And so that means that if that pub chain wants to do something to reduce the emissions of chicken, the only thing that they can do that will show up in the numbers is to buy less chicken and sell less chicken. Because actually, none of the other variations of what they could be doing from buying different types of chicken, buying chicken from different types of suppliers, cooking the chicken in a different way, using different portions of chicken, none of that really shows up, because there's one economic intensity based coefficient which says 100 pounds or €100 of chicken is x in emissions. That's the crux of the problem, as I see it.
Isobel Wild [00:04:36]:
And why aren't there more emission factors available? What is the big challenge surrounding this limited accessibility to emission factors?
Saif Hameed [00:04:46]:
So I think it's actually not just about why are there not more emissions factors available? I think the first thing is, do companies care about the emissions factors, or do companies understand what role emissions factors play in their decarbonization program? And if they understand what role those factors play, then they will create the demand for more emissions factors. And we can talk about whether there are enough or not enough, or where they come from. But the first thing is just to understand why emissions factors are important. And actually, you mentioned at the start, Izzy, that most companies are getting by with several thousand emissions factors. I would argue probably much less, probably several hundred emissions factors in many cases. And actually, in certain industries, certain factors are super important, and everything else doesn't matter. I was speaking just recently with a company in the construction space, basically. So they make construction component, I won't say which component because it'll be super easy to identify them based on that.
Saif Hameed [00:05:46]:
But a particular component that is used in buildings and steel is understandably a huge part of that. Now, when it comes to steel, they have really two emissions factors that they're playing around with. They have a european factor and a chinese factor. And those are the two places where this component is manufactured. Now, there's huge variation between Europe and China, but actually there's also a fair bit of variation beyond that. Like where is the steel for the component being purchased from? Is it russian, is it indian? Is it chinese? Is it from somewhere in Latin America? Actually, there's maybe a three or four x variation just based on sourcing location and understanding that for this component, that is probably 40% of your emissions footprint, if not more. That variation is really important. So the first thing I would just anchor on is understanding importance and need before we go into are there enough emissions factors out there? In this case, making the jump from two factors to 20 or to five or to seven will probably already deliver very significant gains.
Saif Hameed [00:06:47]:
And those factors are available. They're out there.
Isobel Wild [00:06:49]:
I know that within, there's a big shift going on from spend based data to activity data. For instance, within emission factors, is there a similar trend in trying to get more specific data around that? And if so, are there certain types of emission factors that are available?
Saif Hameed [00:07:10]:
The emissions factor or the coefficient needs to match the methodology that you're going to be using. So if you are running a spend based emissions calculation, let's say scope three emissions calculation, that is spend based. That means you're saying, hey, I'm buying €100,000 worth of steel or tomatoes or whatever it is, and I need an emissions factor that is represented in the same unit. I need an emissions factor that is represented in dollars or euros or pounds. And actually, the more abstracted I get from that, the more I am exposed to variation. What I mean by that is, let's say the emissions factor is natively in euros and your spend is natively in dollars. Well, now you actually have to do a forex conversion of that euro based emissions factor to dollars to be able to match that. And that's just one example of how even within a methodology, a spend based methodology, the lack of matching at the unit level can create some variation.
Saif Hameed [00:08:09]:
The next thing is to say, okay, actually I'm not using spend based. I'm going for a weight based approach. And I'm saying I bought 100,000 kilos of tomatoes or 100,000 kilos of steel or 100 tons of steel. And then you're going to say, well, now I need an emissions factor that is represented in mass terms, not in dollar or euro terms. I need what is a kilo of tomatoes or what is a ton of steel in emissions terms, so I can multiply it against the data that I have from a math perspective that's logical and straightforward. I think that carbon calculation often masks a lot of simplicity behind this wall of arcane expertise, often just to try and make itself seem more specialized. Everyone studied this sort of stuff in. In math class many years ago.
Isobel Wild [00:08:56]:
What does it actually take then to get that more specific emission factors?
Saif Hameed [00:09:01]:
Yeah. So I think there's probably a few steps to this. The first thing is just to understand which areas of my calculation do I care about getting granularity and accuracy? And we talk about this often, there will be certain materials in that construction related business, it's the steel. In another business it might be tomatoes, in another business it might be cotton. Just where is there a lot of concentration of emissions within your inventory? And therefore, actually, the difference between one emissions factor being applied and 100 factors being applied is likely to be important in other areas. It's likely that you can outsource this problem safely and don't have to query it too much. Now, most commonly, a sustainability professional is either working with a third party vendor or partner on this, a consultancy or a software company, or they're trying to do this internally. If they are working with a third party, I would say ask questions.
Saif Hameed [00:09:56]:
Ask which methodology will be applied. Ask which emissions factor databases will be used, which actual database providers will be used. Ask maybe what is the variation in emissions factors and the granularity. A good question to ask is, do you have regional specificity on emissions factors? I am buying tomatoes from 15 different countries. All those 15 different countries will look different. Do you have emissions factors represented at the country level for those 15 different countries? I am intending to do a mass based calculation rather than a spend based calculation. Do you have those emissions factors on a mass basis at regional level for tomatoes? Because that's what's important to me. And I would say that the more that you dig into this, the more you'll learn and the more you can shape the direction of where that calculation goes.
Isobel Wild [00:10:45]:
On the variation point, it seems that if somebody, or if a professional did want to invest, you know, we've spoken about chicken, so they want to invest in developing more supplier specific efs and get a greater accuracy there. How do you ensure that you have consistent emission factors that are comparable across the board to enable that whilst also keeping to the methodology and having a holistic overview of everything being the same.
Saif Hameed [00:11:13]:
This a lot on back here, Izzy. So I would say the first thing is consistency in methodology is going to be key because that means that whatever the emissions factor that you're using consistent with the methodology that you're applying, and there has to be some level of scrutiny on where that emissions factor came from. Did it come from a trusted source? Does it come from a source that has some basis in academia, for example, and is known to be conformant or is verifiably conformant with the methodology that you're using? And so to some extent you're outsourcing the problem to whatever that standard is that you trust. And so maybe you say, look, this is a World Food LCA database emissions factor. And I trust the World Food LCA database because most of my peers are using it. And so this is fine for me. Maybe there are some other databases and sources and we should talk about that as well at some point, Izzy, because there's a lot of new sources coming up. So you sort of say, okay, I trust this source and this source corresponds to the methodology that I'm using.
Saif Hameed [00:12:16]:
So I'm not going to worry about this. That allows you to maybe not be too hung up on whether you're using different sources for different regions because they're still governed by the same methodology progressively. You'll probably want to start getting into that as well, but I don't think most companies are yet at that level of frontier, if that makes sense.
Isobel Wild [00:12:36]:
Okay. And on your sources point is this in terms of, because I guess, secondary sources. So you know, your, your databases that you go to are verified, but your primary sources might not be third party verified. What is the golden rule here?
Saif Hameed [00:12:53]:
What I've learned is that primary means very different things to very different people. So back when I was at school, Izzy, or at college, primary data was data that was the original source of information, was factually as close as possible to the truth. What I find in this space is that companies tend to think of, let's say supplier data as primary data and a modeled emissions factor as secondary data. So they say if I'm buying 100 kilos of tomatoes from you or 100 kilos of chicken from you and you're giving me an emissions factor for this 100 kilos of chicken or this 100 kilos of tomatoes, that is primary data. And if I'm using, let's say an axio base or an eco invent or a database to source what chicken looks like in emissions terms, and multiply that against the volumes that I'm buying from you. That is a secondary data point that I'm using. And most companies are drawing that line. This is factually not always correct.
Saif Hameed [00:13:59]:
And the reason for that is that what I'm finding, and what we're finding at Altruistiq, from running supply chain engagement programs across hundreds of large companies, is that very often the supplier is giving you the same secondary emissions factor packaged up as their own data output, basically. So they are themselves tapping into those same databases that you're looking at, or you would reference, and they're basically just giving you 100 kilos, times one kilo's worth of emissions from this database, and they're giving you that number. And that's the same secondary data. There's no difference, really, between that data and the data that you could get yourself. That is, to me, secondary data. In this space, primary data is basically an LCA done using the actual processes that you're running, the actual attribution of those processes to the product that you're selling, and taking into account the locations that all the ingredients and materials came in for. So at Altruistiq, when we generate a product carbon footprint for an individual product or an individual stock keeping unit, that product carbon footprint takes all of the ingredients that you purchased from the places you purchased them, in the combinations that you combined them into using the processes that you applied, incorporating even the factory floor space that was used, and the energy intensity or consumption in that floor space for making that product layered on at the level of that individual product. And that product carbon footprint can then be used as an emissions factor on your customer side in their inventory.
Saif Hameed [00:15:33]:
That is more like primary data in my world.
Isobel Wild [00:15:36]:
So it's essentially a bit of a win win, because if you're investing in getting more, as you've just explained, their primary data for PCF, it helps on both sides, because then you can actually use that for your emission factor as well.
Saif Hameed [00:15:51]:
One of the big problems here is if you think about LCAs, LCAs are cumbersome. They are good, but they are cumbersome.
Isobel Wild [00:16:00]:
Can you first make the distinction between an LCA and a PCF?
Saif Hameed [00:16:04]:
Yes, for sure. If you think about an LCA. An LCA, as I would describe it, is an academic piece of research. It will be typically conducted by an academic, either one that you've employed or one that you've outsourced to. And so in our organization, for example, like peers peers could conduct an LCA, but peers has also studied and got degrees in this space, and therefore can do this well. And a proper full LCA will usually take months of work. It will require quite a lot of data. It might require even some interviews and conversations around assumptions.
Saif Hameed [00:16:38]:
It is usually done at the level of, let's say, one product, but not always one stock keeping unit, because it usually doesn't make sense for one stock keeping unit. Typically have a number of data points covered beyond just carbon. It will usually look at a range of environmental impacts. Often you have lcas also covering social impacts as well. In fact, I think actually Piers has in his academic life looked at that a lot as well. I think, unless I'm misremembering, and that LCA takes a long time, is very expensive, is very cumbersome, and the output is like a PDF, maybe a spreadsheet, but like usually a PDF or a document of some sort of research paper or publication. And that is very hard for a business to use at scale on the go. Basically a PCF, as we generate it at Altruistiq, a product carbon footprint is a data set snapshot of the product that you've made.
Saif Hameed [00:17:34]:
It says, here is a product that you've made, and we can do this a thousand times over, 100,000 times over, a million times over. And it is the unique combination of materials that went into this product, the manufacturing process that was transforming those ingredients into the finished product, the energy consumption, the logistics, all the different elements that went into this. And it is a snapshot of what all of those things look like at the level of this individual product. And you can define that product as a stock keeping unit. So a six pack of baked beans, or a product, baked beans in general, or whatever level you want, and it is dynamic and it can just be run as many times as you want. There are some trade offs that that has versus an LCA. One trade off is that we, for example, do this just on carpet, we will progressively do this on water and the other environmental categories under PEF legislation or requirements, EU pef. But it is not, for example, covering all the social aspects.
Saif Hameed [00:18:34]:
It will often lack a lot of the qualitative aspects that you might have in an LCA. An LCA might have a narrative quality to it, or a research quality around the processes, etcetera. Whereas a PCF is just data, it is just numbers in a snapshot. And that those numbers are usually in some tabular form, broken down. There isn't a lot of context embedded in that PCF. If that makes sense. There is some metadata, which means you have facility associated with it. You have supplier name tags associated with it.
Saif Hameed [00:19:03]:
That is the metadata. But there isn't, let's say, the narrative quality that you can often get in a good research paper.
Isobel Wild [00:19:10]:
Thank you for making that distinction. Some of the challenges we've therefore spoken about is variation, accuracy, scalability. And one that's just popped my head then is around frequency and frequency of emission factors, or coefficients, as you call them. How do you solve for frequency, especially when you're making regular updates or having lots of decarbonization initiatives going on at the same time, and you want to ensure that that is being shown in your numbers?
Saif Hameed [00:19:44]:
I think that this is fundamentally also a question of what is important to you in your business and how frequently does the dynamic of that thing change. Let me illustrate. Let's take this steel based component that I was talking about. Steel emissions don't change super frequently relative to some other things. And that's because steel is the manufacturing process requires a lot of capital investment. Companies are not going to make that capital investment very frequently. You have a certain manufacturing facility thats going to run for ten years. Yes, theres this talk about hydrogen based steel and green steel, et cetera, et cetera.
Saif Hameed [00:20:23]:
But these are slow moving initiatives that will take years to mature. So arguably, if I have an emissions factor, let's say, for steel in Russia or an emissions factor for steel in India, there may not be that much variation across this. Certainly within the year that's very unlikely. And even across a few years, there might not be that much variation. It is good practice still to have an annually refreshed emissions factor, and most databases should be able to provide you with this. Really, there's not likely to be that much difference. Let's take something else as an example. Let's take cotton.
Saif Hameed [00:20:56]:
In the case of cotton, I would actually argue that the emissions might be very different every year. And so we've used this example before, but if you think about, let's say, cotton grown in Pakistan, I know being from Pakistan, that cotton yield varies by about 20% to 40% in some years versus other years. And that's a function of many factors, climatic factors, how many farmers are growing cotton, maybe how much cotton they're growing. Like, there's just a lot of things that go into that. And so if you're assuming that the yield on the same amount of input is going to vary by 20% to 40% in a given year, how is it possible that the emissions intensity of that output will not vary similarly. So the emissions factor, if you're using it for the wrong year, it won't be accurate.
Isobel Wild [00:21:40]:
Yeah, and I think on that point, I was actually looking at our emission factor database at Altruistiq, which is a mixture of secondary emission factors as well as supply specific emission factors. And there was one which really stood out to me, showing the regional variance, and that was vanilla. And about 80% of vanilla comes from Madagascar, but there are three main sourcing areas. So Madagascar being one of them, Indonesia and China. And I was looking at the emission factors, and for the Madagascar vanilla bean, it was 31.86. And then looking at Indonesia, it was 127.43, which is like a huge variation. And if you're using a global ef to account for your Madagascan or to account for your vanilla bean, you're going to be either really, really overcompensating or undercompensating. And I think those kind of variations just show the importance of getting more specific emission factors.
Isobel Wild [00:22:43]:
And as you said, Saif, for those material areas in your supply chain, in terms of what you anticipate, the key trends and development in the use of Es in the coming years to be, do you have any sense of where it's going?
Saif Hameed [00:22:56]:
I do. By the way, this is just an interesting area where I think there is not a lot of consensus. And I know that different members of our team also have slightly divergent views to me. So let me maybe talk about what I think are the common areas and where then there might be some variation of opinion. I think there are likely to progressively be more emissions factors out in the world. And I think that this is going to rise by a massive factor. If you're looking at, let's say, the World Food LCA database, which has about 2308 emissions factors, last I checked, I could be a little off on that, but I think that's about the number you're going to get, many times that in the larger emissions factor databases. And so if you look at our database, for example, Gaia DB, we have around 200,000 factors.
Saif Hameed [00:23:45]:
Sometimes that rises, sometimes that falls, because we're constantly curating this and trying to deduplicate and make sure that were ending up with emissions factors that meet our bar on quality. We are optimizing now no longer for volume or breadth. We were optimizing for volume at some point and having the most emissions factors in the world. We have learned that actually you need to have enough, but then beyond enough, you need to have quality, and quality starts to matter more than volume. At some point round about 200,000 right now is the speed where we're sort of hovering. That is for largely secondary emissions factors, or what would be termed as secondary emissions factors by my definition. What I mean by that is that these are modeled factors. They've been, in some cases, modeled by us.
Saif Hameed [00:24:30]:
There are thousands of factors we've developed ourselves with our own modeling capabilities that don't exist in the world, and that we are the only ones to possess. And there are many others that we have sourced from third party databases that number that 200,000 number. There are probably going to be more databases being added to the market. For example, I think it's Pulse Canada, which has developed a database for the canadian pulse industry to have their own emissions factors. And I expect many more industry backed databases to emerge in the secondary world. So that's going to be one innovation that will change the number of emissions factors available. From a secondary perspective, however, the big game changing move is going to be the movement towards PCs. Because if you think of, let's say, a large ingredient supplier, for example, like an archer, Daniel Midlands, or a Cargill, or an olam, and if you imagine a world where each of their stock keeping units might have an emissions factor generated, at least partially with primary data around their actual processes, their actual ingredients, etcetera, we're talking about millions of pcfs coming onto the market.
Saif Hameed [00:25:38]:
And those pcfs will be operated and dealt with in a very different way. They won't be publicly available in datasets, they will be available in closed data sets, effectively closed databases, accessed through API permissioning on all sides. Whether they're charged for, whether they're free, this is a question of debate. And actually I think there are very different views on this. Whether the Olams and the Cargills and the ADM's can extract value in exchange for giving PCs, that's where the divergence comes in. But there will, in my view, undoubtedly be millions of PCs in the ecosystem, and that is just a massive order of magnitude greater than what we have today.
Isobel Wild [00:26:18]:
And in terms of financing this, because I know that access to secondary EF databases actually can be quite expensive, how would you use your budget to create a concrete emission factor database that's available to you? Would you invest more in those supply specific emission factors? Or would you try and go for breadth and actually get a range by the secondary databases?
Saif Hameed [00:26:46]:
I actually had at some point a dual role at McKinsey, where I was leading sustainability, but also working extensively on advanced analytics for some large companies. What I found is that the data side of things is often commoditized. And actually six years ago, what was interesting was the algorithms that you would build to use that data. And that was actually where the IP or the value is shifting progressively. I think that it is going to be the user experience and the application layer or the way in which those algorithms are deployed to deliver value that is going to be valuable for companies. And the reason I'm saying this is that I actually think emissions factors are going to become radically cheap. I think they're already reasonably cheap today. I think that they feel expensive because companies are now having to pay for them.
Saif Hameed [00:27:36]:
And so if it's a matter of several thousand dollars, this is a new expense and you hadn't banked on it. But it is not hundreds of thousands of dollars for an emissions factor database access. It is thousands of dollars, maybe several thousand dollars, depending on how many databases you need. For a large business, it is not a huge expense. I also think a lot of that will get cheaper because I think that those emissions factors are going to be increasingly commoditized. I know that the database providers are looking to protect value and capture value. I don't really think they're going to succeed in a big way, at least on a per unit pricing perspective. They may get a volume play and so there may be just so many customers buying their factors.
Saif Hameed [00:28:18]:
But I think on a per unit basis this stuff is going to become really cheap. I think what gets interesting is how do you set up your capabilities to use those factors at scale, such that as new factors come onto the market, they are seamlessly deployed in your calculation. By seamlessly I mean with proper guardrails and controls so your numbers don't move all over the place. And as new pcfs become available from your suppliers, for example, they are accessible, let's say, through the pact API. And they can also be seamlessly incorporated in the same way into your inventory. Setting up that layer of capabilities, which is obviously what we do at Altruistiq, that I think is actually what gets really interesting. I wouldn't worry about emissions factor cost because I think that's a race to the bottom.
Isobel Wild [00:29:00]:
Yeah. And actually Dan Enzer, who's our emission factor man? Dan, the EF man at Altruistiq highlighted something quite interesting to me, which was around this quality and the frequency of emission factors, where NemeSec studies from 2002 actually bolster up the eco invent agricultural emission factors. So the ecoinvent cheese EF is based on six us mozzarella factories in 2013. So this is also to the point of actually, when you're looking at your emission factors that you're using, it's well worth having been quite, like, scrutinizing with them, seeing what the quality is, what the vintage is, and actually how much data has gone into it, which is something, Saif, I think we're trying to play for Altruistiq by actually having the visibility into each EF that you use to ensure that you know where this data is coming from.
Saif Hameed [00:29:56]:
Yeah, I think the side note is that there's a decent chance all those factories are owned by the same company, Liprino, which has a near 100%, I think, market share on at least pizza mozzarella in the US. But that's something we can chat about later. But the Nemechek report, I think, has brought a lot of light onto this, and we should also credit, frankly, it is the poor and Nemechek report, and our friend Joseph Poor is the co author, our friend and scientific advisor. That report was really interesting because it was the first of its kind to be thorough across all of the different materials and ingredients that you might find in a grocery store. And I think that that's kind of going to be the foundational piece. But there's so much room to now dig into granularity, where these factors come from, what's being used, et cetera. This space is just getting started.
Isobel Wild [00:30:45]:
Awesome. So maybe to summarize quickly what we've.
Saif Hameed [00:30:48]:
Spoken about today, we've talked a bit about what an emissions factor is. An emissions factor is an environmental coefficient for something that you've done. We've talked about the importance of being consistent in the methodology you're applying to your calculation and the emissions factors that you're looking for and making sure that those emissions factors are fit for purpose in your calculation. We've talked about the difference between primary emissions factors and secondary emissions factors, and how this isn't always as obvious as it may appear. We've also talked about PCs versus lcas, what those two are, and the trade offs, and also how they relate to emissions factors. We've also touched a little on the direction of travel in the emissions factor space, and whether this is a cost component that is likely to rise or actually a cost component that increasingly will be something companies don't need to worry about.
Isobel Wild [00:31:41]:
Amazing. And is there anything else that you want to add? Any other final nuggets of wisdom to this discussion?
Saif Hameed [00:31:49]:
Yeah, I think that there is a world where emissions factor creation presents a lot of opportunity for companies. Actually, this is an interesting under invested in space. I think it will grow somewhat slower than some might expect. So I think it'll take years for this space to properly grow into itself. I think pockets of this space will grow quite rapidly, like pcf generation. I actually think that next calendar year, you're probably going to see 100 times more pcfs created than this calendar year overall in the industry or more. I think in the secondary emissions factor database or market. I think there's going to be a steady stream of growth year on year in availability.
Saif Hameed [00:32:35]:
I think double digit percentages in the growth of emissions factors available is probably going to carry on for the next four or five years. So this is an exciting area of innovation.
Isobel Wild [00:32:45]:
We've had challenges, we've had solutions, we've had predictions, and we've had some high quality advice from sefamead. Thank you, everybody, for listening, and thanks, Saif.
Saif Hameed [00:32:56]:
My pleasure, Izzy. Thanks.
Isobel Wild [00:32:57]:
Bye.