Ginkgo Bioworks Holdings Inc

Good evening. I'm Megan LeDuc, Manager of Investor Relations at Ginkgo Bioworks. I'm joined by Jason Kelly, our co-founder and CEO, and Mark Dimitrick, our CFO. Thanks as always for joining us. We're looking forward to updating you on our progress. As a reminder, during the presentation today, we will be making forward-looking statements which involve risks and uncertainties. Please refer to our filings with the SEC to learn more about these risks and uncertainties. Today, in addition to updating you on the quarter, we are going to provide updates on our path towards adjusted EBITDA breakeven, including a deeper dive on how we're executing against our cost reduction targets, as well as what we're doing to drive revenue. As usual, we'll end with a Q&A session, and I'll take questions from analysts, investors, and the public. You can submit those questions to us in advance via X at hashtag GinkgoResults or email investors at GinkgoBioworks.com. All right. Over to you, Jason.

Thanks, Megan. And thanks, everyone, for joining us. We always start with our mission of making biology easier to engineer. And in order to do so, particularly in this quarter and the quarters to come, we're focused on three objectives. First, reaching adjusted EBITDA breakeven while maintaining a cash margin of safety. We ended this quarter with $730 million in cash and no bank debt. We also made aggressive moves in headcount reduction and other reductions that will be reflected in reducing our cash OpEx spending in the coming quarters. Second, while we're cutting these costs, we need to keep serving our current customers well. I'm happy with our revenue number this quarter, which are indicative of continuing to serve our current customers. This was a big lift for the team alongside a riff and a change in how we are organized, but it's an early indication that these changes were effective in improving the efficacy of our delivery. Finally, we want to grow our revenue in solutions and expand into selling tools. So I'm going to cover this more in the strategic session, but we're excited to open our platform directly to our customer scientists. Previously, it's been something we've just had available to ourselves here at Ginkgo, and we're getting that out there in a more democratized way. Our technology assets in bioengineering are world-leading, so I'm excited to find more ways to sell them and drive growth. As a reminder, in our Q1 call, we noted our annualized off-ex of about $500 million was simply too high relative to near-term revenues. To address this, we announced a plan to cut this back by $200 million on an annualized basis by mid-25, including consolidating of our footprint and reduction of our labor expenses across both G&A and R&D. And I think, again, with our strong cash position, we're well positioned to continue executing on these restructuring efforts. I will be providing a detailed update on the cost reduction plan later in this presentation, but for now, I'd like to give you a summary of the actions we took in the second quarter relating to headcount reduction. At present, we have notified approximately 450 employees, or roughly 35% of the business, that they will be impacted by our reduction in force. Approximately 300 positions were impacted as of the end of Q2, and an additional 100 positions are anticipated to be impacted by the end of this year, and the remaining 50 by mid-25. These cuts, although difficult to make, are estimated to save Ginkgo over $85 million in annualized cost savings once they're fully implemented. And because of these reductions, we're very much on track to hit our goal of reducing our annualized costs by $100 million by the end of year. I know this is not much consolation, but I do want to take a minute and again thank our employees who were let go as part of this reduction in force. They contributed enormously to building Ginkgo, and we're deeply grateful for it. Much of what we're doing now at Ginkgo with these changes is to establish a firm base for the company that will allow us to then grow and meet the mission that they all helped us build. Now, I'm going to get into cost-cutting details in the strategic section, but before I do, let me hand it over to Mark to go over the financials.

Thanks, Jason. I'll start with the cell engineering business. Cell engineering revenue was $36 million in the quarter, down 20% compared to the second quarter of 2023. Similar to Q1 of this year, this decline was driven primarily by a decrease in revenue from early stage customers, partially offset by growth in revenue from larger customers. We continue to believe the shift to larger cash-based customers to be an overall positive shift. In the quarter, we supported a total of 140 active programs across 82 customers on the cell engineering platform. This represents a 33% increase in active programs year over year with solid growth across most verticals. As we discussed in our prior earnings call, we anticipate the nature of programs that we take on with our customers to evolve in the future following our recent adjustments to commercial terms and offering. While still very early days, this slide gives you some detail on how the nature of programs is changing. We added a total of 18 new programs and contracts in Q2 2024, of which 10 were generally comparable in size and scope to historically reported new programs. Importantly, you'll note that of those 10 deals, five included downstream value share potential. In addition, we commenced eight other customer contracts in the quarter that represent a variety of small deal archetypes. These are generally much smaller in scope and shorter in duration and include two lab data as a service deals in the protein characterization space that we signed with a large cap tech company, which itself is an entirely new customer segment. The current sales pipeline for both categories of deals is solid, and while it's still in its early days, we encourage that during the course of a major restructuring, we have been able to execute on existing customer programs while converting new opportunities. Jason will speak further about our approach to driving revenue later in the presentation. Now turning to biosecurity, our biosecurity business generated $20 million of revenue in the second quarter of 2024 at a gross margin of 41%. Revenue and gross margin were up significantly in Q2 due to the timing of a customer contract. And now I'll provide more commentary on the rest of the P&L. Where noted, these figures exclude stock-based compensation expense, which is shown separately. And we are also breaking out M&A and restructuring-related expenses to provide you with additional comparability. Starting with OPEX, R&D expense excluding stock-based comp and M&A and restructuring costs increased from $99 million in the second quarter of 2023 to $110 million in the second quarter of 2024. This increase was mainly driven by an increase in rent expense and AI-related spend. gna expense excluding stock based comp and m a and restructuring costs decreased from 59 million dollars in the second quarter of 2023 to $45 million in the second quarter of 2024, which reflects cost reductions we completed in 2023. Importantly, Q2 does not reflect the benefits of our headcount reduction, since those only commence at the end of June. And so we would expect both R&D and G&A expenses to decrease meaningfully by Q4 of this year. The M&A and restructuring related costs this quarter includes a goodwill impairment charge of $48 million, other costs relating to restructuring such as severance and costs relating to smaller M&A transactions that we close early in the quarter. A full reconciliation of this line item can be found in the appendix to this presentation. Stock-based comp, you'll again notice a significant drop in stock-based comp this quarter, similar to what we have seen over the past year as we complete the roll-off of the original catch-up accounting adjustment related to the modification of restricted stock units when we went public. Additional details are provided in the appendix. Net loss, it is important to note that our net loss includes a number of non-cash income and or expenses as detailed more fully in our financial statements. Because of these non-cash and other non-recurring items, we believe adjusted EBITDA is a more indicative measure of our profitability. We've also included a reconciliation of adjusted EBITDA to net loss in the appendix. Adjusted EBITDA in the corridor was negative $99 million, which was down from negative $80 million in Q2, 2023. This decline was driven by a decrease in total revenue partially offset by a decrease in certain operating expenses. In addition, I would like to note that we are now reporting adjusted EBITDA inclusive of non-cash, in-process R&D charges relating to acquisitions, and so we have separately itemized that amount for you here. And finally, capex in the second quarter of 2024 was $13 million net of tenant improvement allowance as we continue to build out the BioFab 1 facility. In terms of outlook for the full year, we are reaffirming our guidance for 2024 with cell engineering revenue expected to be $120 to $140 million and biosecurity revenue expected to be at least $50 million, totaling $170 to $190 million. In conclusion, we're pleased with our overall execution of the restructuring thus far as we navigate substantial cost reductions and commercial changes, which we see as foundational to our path to adjusted EBITDA breakeven.

Back over to you, Jason. Thanks, Mark. I'm going to use the first strategic section to focus on Ginkgo's efforts to reduce costs, as that's currently my primary focus. Ginkgo is a unique player in the life science tools industry. We have more than 100 active cell engineering programs running on our platform, across biopharma, industrial, and agricultural biotechnology, and we have a unique scale and breadth in both automation and software and cell engineering. We can deliver on that services business profitably, and we're focused on demonstrating that as quickly as we can. I'll cover how we're taking out costs while maintaining delivery for our customers. Second, I want to talk about how we see the opportunity for growth at Ginkgo by opening our platform up directly to our customer scientists while focusing our existing service offerings around our areas of strength. And then finally, I'd like to spend some time highlighting a growth opportunity within biosecurity as it tackles emerging threats and modalities, specifically H5N1 or bird flu. Okay, let's get started. During our Q1 call, we announced our plans to cut spending back by a run rate of $100 million by Q4 2024, with an additional $100 million expected to come out by mid-25. Earlier on this call, I mentioned that we expect to see over $85 million in annualized cost savings from our reduction in force. As you can see from this chart, of the approximate $85 million, we expect to save by mid-25, $75 million of that is expected to be achieved on an annualized run rate basis in 24, based on actions we've already taken. In addition to the people cost savings, we've taken actions expected to result in an additional $25 million in annualized cost savings by the end of year, putting us on track to hit our goal of reducing costs by 100 million by end of 24 on a run rate basis. Because we are still in progress with the majority of the non-people cost cutting initiatives, I'd like to take a minute and dive deeper into what we're planning to reduce costs there. So we've established a number of internal work streams, I think 19 or something, focusing on major spending areas. A few examples, streamlining third-party costs. We're focusing on realizing efficiency with our vendors through strategic sourcing and renegotiations. Additionally, we're reducing our dependence on third-party technical work and consulting, as well as external legal services. We're reducing, as I mentioned previously, our real estate footprint actively and looking for sublease opportunities there as well. Equipment cost alignment, we're adjusting our equipment expenses and related service contracts to match the current utilization, and then we'll scale into demand as it comes in. We're undertaking a significant effort to rationalize our software portfolio, a lot of enterprise software. by reducing licenses and consolidating applications. Overall, on the technical side, both in the lab and with software, we have a lot of historical data now on what infrastructure really pays off across these hundreds of programs we've done at Ginkgo. So we've been able to make good decisions, decisive decisions quickly in this area. And so I'm excited to see that play out and save us money. Okay, importantly, alongside our cost-cutting initiatives, we're continuing to deliver for our customers. This is key. So recently, we delivered on a major technical milestone for a previously announced large biopharma customer we have. In the midst of all these restructuring efforts Mark mentioned, we were able to sign four new agriculture deals, the largest of which was with Syngenta, where we're optimizing a microbial strain from their biologics pipeline. This is a molecule that they've earmarked as a pioneering biological solution at Syngenta. Successful, cost-effective, and large-scale production of this metabolite would expedite their go-to-market timeline for these biological solutions. Next, as Mark mentioned, we signed our first two, we call them LDAS, Lab Data as a Service, deals with a large-cap tech company, and we're excited to execute on those in the near term. And lastly, I'd like to reiterate, we're reaffirming guidance and are confident in our abilities to continue executing for our customers while we act on these cost-cutting initiatives. I want to take a minute and thank the team at Ginkgo who handled really just a crazy lift to deliver so well on customer programs in the same quarter where we had such a large amount of organizational change. That's going to get easier going into the future, but that was no small feat, and it's a testament to the strong culture at the company and our focus on delivering for customers. OK, that's an update on where we expect our cost takeouts to come from. And next, I want to talk about how we're building the tools and solutions that are going to grow Ginkgo's revenue going forward. So like these two images, if you look how the work of engineering cells is done today, it looks like this picture on the left. It's LabBench. And first, I want to say that it's a little frustrating because this is exactly what my LabBench looked like in graduate school at MIT 20 years ago. And I promise you, if you walked into the computer science department at MIT, the tools available to researchers would be wildly different from 20 years ago. But in biotech, the tools have changed a lot less than you would expect. And honestly, I thought about this, and I think it's because This set of tools is actually pretty great for what it's intended to do. It allows bench researchers to explore hypotheses quickly and adopt new protocols that they read in papers in a matter of days. They can go in that thermal catalog and order whatever reagent they just read about in a paper, pick up a pipette, and get to work. And for many problems, this set of tools and a by-hand approach to doing lab work is the best solution, right? The big downside to this approach, though, is that since it's manual, you're doing it by hand, there are no advantages to scale. In other words, it doesn't get cheaper or higher quality as you do more of this genetic engineering research work at the lab bench. We have a very different model at Ginkgo that relies on automation. So you can see part of our Boston installation of our proprietary rack robotics hardware in the photo on the right. This is an automated approach that allows for much more data per research dollar. And it gets better as you do more of it with scale. But it comes at the cost of less flexibility than you get when you work by hand. And we've been using this approach at Ginkgo now for hundreds. I'll say it's like high end R&D projects. People tend to send us some of the hardest R&D challenges they have. That's why they're outsourcing it from our customers. And we've learned what works and what doesn't work when you apply this large data approach to genetic engineering. And I'm not going to say it's useful for every single problem in biotechnology. I don't think benches are going to go away. I think scientists will still be using those in the thermo catalog. But there is a large set of problems we see could be better solved with automation and large data generation. In particular, you've seen a surge of interest recently from AI biotech companies that really want to generate large data assets for model training. A big question for Ginkgo is if we're right about that platform, and let's say we are, it is applicable to a big set of problems. What's the best way to sell it to customers who do need that large scale data generation? And I'll walk through two ways. So the way we've been selling it to date is the approach on the left. All right, so we're really primarily selling to the decision maker and our customer is the head of R&D, or if it's a smaller company, it's the CEO. And they're deciding to outsource a whole research program. So, hey, Gingo, we want you to go off and deliver us back this scientific result. and a small team of Ginkgo scientists in the middle there are going to use our platform and automation, but ultimately meet with that customer quarterly at a joint steering committee meeting and and return scientific results to them. But we're our scientists are the one using our platform. So we might return a better manufacturing process to Nova Nordisk or fertilizer producing microbes to bear or an improved enzyme for Merck and the other 100 plus cell programs. That's the kind of that's how we've been doing that work for our customers. We're going to keep doing that business, and we see it growing. And we're very excited, however, to open our platform in the way you see on the right, by making it directly available to scientists at our customer sites. And so I'm going to call these two approaches solutions on the left, the customers coming to us to kind of solve their problem completely, or tools on the right. We're providing a set of tools to our customer scientists, and they're going to solve their problems for themselves. All right? And so I want to... For folks that are new to the life science tool space, I want to just lay out a spectrum of how I see the industry from solutions to tools. On the y-axis, the top of it represents how customized an offering is for what a customer wants. Is it something we built just specific for the problem you have? It also represents how much technical risk Ginkgo is bearing with the offering. In other words, are we taking all the risk or are we sharing some of it with the customer? And as these two things go up, you get the most extreme form of a custom high technical risk B2B solution in the market up at the top left, which is where a small biotech company develops a drug asset with no intention of ultimately becoming a standalone drug company, but really intending to ultimately sell, license that drug asset to a major biotech or biopharma. That's the most extreme form of customization and high technical risk. And you see many platform biotech companies in the industry taking this approach. Companies like Abcci, Abcelera, Recursion, they all have internal drug pipelines and they will profit handsomely if they get a good result in a clinical trial. And this is a very functional, it's a good business model. It works. uh we have not taken this approach at ginkgo uh for a variety of reasons we believe we can bring more of a direct platform services business model into the industry and so we went down that y-axis and you can find our cell engineering solutions offering there that service business where gingko is definitely bearing less technical risk than a small biotech That's doing their own drug. For example, our customers pay us fees. That's where you see our revenue in cell engineering is largely fees today that supports the technical work. So they're paying us to do a lot of that research service for them. But we're making a very customized solution for them. And the key point here is the more customized the solution is, in other words, on the left-hand side of this chart, the more likely Ginkgo is to be able to negotiate downstream value share. In other words, to get royalties or milestones. And most of our cell engineering solution deals today have royalties or milestones. And if you see us keep signing them up, like Mark mentioned, signing some of those large programs, they will continue to have royalties and milestones in them in the future. The dotted line in the in the middle shows that at some point you're offering something more off the shelf. In other words, less customized. And so then the customers aren't going to share royalties with you. Right? This is roughly where I draw the line between what I'll call solutions and tools. Okay. So, as you move to the right side of the screen, you'll see things like our lab data as a service. where we're producing data at scale for customers, but we aren't bearing a ton of technical risk. In other words, the customer is coming to us with a design and we're generating data. If the design is bad, that's their fault. And so when we do those deals, we don't expect to see downstream value share. And towards the end of this tale, you'll see that we have AI and automation listed on this chart. And at Ginkgo, these two tool pieces are in their early stages. But the idea behind these is that Ginkgo can develop modular tools, robotics, software tools that scientists and developers at our customer could put together to make their own infrastructure in-house to use. And so we'll talk more about that in the future. Okay, so what I wanted to dig in today first on this chart is your cell engineering solutions business. We love this business. We think we are very differentiated at Ginkgo in this area, both in technology, having a wide enough array to build a custom solution for a customer effectively, and even in our sales. our sales team and our approach. These are really complex deals to sell. They're really complex deals to negotiate. And we do a lot of them every quarter, I think more than anyone else. And so the change we made with the restructuring, though, is that we will no longer be taking kind of any cell engineering work a customer comes and asks for. We're going to be limiting that work to a more narrow set of offerings in each market that Ginkgo can deliver efficiently. And so let's dig in and look at ag and then the biopharma industrial. So in agriculture, the first product we'll be offering is strain optimization for existing products. So Agrivala is a customer of ours. They're currently leveraging our strain optimization service to improve the efficacy of one of their existing biocontrol products. So things they already have out there, just improve them, give them back. Another product we developed is based on some of the work we've done with Bear, where we take an early development lead, something that's still in the lab, and take it to field trials. The assets we acquired from AgBiome earlier this year fit well into this, and so we're excited to continue to expand in that area. Third, we have bioproduction. So this is, we believe, a major growth opportunity where customers are looking to either develop or improve the production of a bioactive by fermentation. So you're putting cells in a tank and producing often a small molecule. This is very similar to the work we do in industrial biotechnology. So we get a lot of efficiency on the technical side. The same back end infrastructure we use here, we'll be reusing in the next section when I talk about our industrial work. Finally, our last offering is supporting discovery of plant traits. This is typically a customer looking for novel modes of action or protein optimization. It's a large area of research spending in ag, and so I think an important one for us in the future. Okay, so now I want to talk about pharma industrial solutions. These businesses are focused on helping customers discover, optimize, and manufacture biologically derived products in three key areas here. Our first offering is protein engineering services. So our job there is to build better proteins and enzymes for both pharma and industrial process enzymes, as well as therapeutic and diagnostic biosensors. And you can see on the bottom, you know, customer projects we have in these areas. as we already currently do work in. Next, we have protein production, which is focused on building and optimizing production strains, including creating better ingredients for foods, things like these milk proteins and so on, or finding better ways to manufacture vaccines. Lastly, we have a strong offering in small molecule bioproduction. That's what I was just mentioning when I was talking about the ag. where we're looking to build and optimize small molecule production strains, including pharma APIs, chemicals, food, flavor ingredients, and so on. We're creating new strains to create products with a wide range of applications. So in all of these cases, you can see Gingko customers where we're currently delivering programs. These, like I said, This fan of what we're willing to sell is actually much more narrow than it would have been before. But the areas we're doing in are our strongest areas and the areas we can deliver most efficiently. So again, I think this is how we move on this path to profitability in the cell engineering solutions is with this more tight focus. Okay, I want to move on to our newer offerings. But before that, to clear up some confusion I heard after our last call, we will keep signing those solution deals and we will often be getting milestones and royalties on those deals. Okay, so we're not getting out of getting any milestones and royalties. It just depends on the type of work we're doing for a customer. So now as we move to the tool side of this chart, we have our new lab data as a service offering that I announced at Ginkgo Ferment back in April. We believe we have a major opportunity for LDAS with the drug discovery market. In particular, AI and ML is increasingly being used in drug discovery, and the need for large data sets to train models is growing. So people have mined a lot of what's out there in these public data sets, things like AlphaFold and so on. We're trained on the public structure database, the public genome database. GitGo's proprietary automation ability to deploy it allows customers to generate large new data sets that they can use to train proprietary models or to create data in areas that aren't structure, which is really what the big existing data sets in the public are. So things like protein activities and so forth. uh in this case our customers are designing the experiments themselves taking on the majority of the biological risk for example they'll design you know a ton of antibody sequences send them to ginkgo we'd synthesize express test those sequences for binding developability assays and since we're mainly providing data and not providing custom solutions again we don't expect ip rights so our customers own all the ip nor royalties or milestones so really clean straightforward interaction. These deals can sign a lot faster. It's very straightforward. In fact, you know, for larger biopharma companies, I think this could just run through procurement rather than needing to be really a sort of a BD negotiation like our solutions. Okay. So I know we end up with a lot of customers tuning into these calls, so I want to give a little more detail, because this is really a new offering at Ginkgo, of what our LDAS offerings look like for drug discovery in particular. So our customers start, they're going to give us a scope for a specific data set, what they're trying to accomplish, and they'll either give us a genetic library or ask Ginkgo to build it. We're world-class at building DNA constructs and so forth. This is where our proprietary platform comes into play. We'll use the foundry to generate large multimodal, in other words, different types of data, all assayed on the same cell line, for example. And then we have software, proprietary software, that can curate and annotate that data and make sure it goes back to your AI ML team in a form they can use for model training. That's a real, I think, unique strength we have coupled to the lab data generation. The first areas we're offering these services are functional genomics and antibody developability. So, in functional genomics, we can provide lots of data for AI model development and target discovery. Common use cases would be target discovery and validation. And then in antibody developability, robust data packages with key developability metrics for lead optimization or AI ML training that can predict biophysical performance of antibodies based on their amino acid sequences. And again, A lot more coming here. We have our first sort of customers running now and we're just at the beginning of this, but we are seeing good traction. So please reach out to us if you have a big data set you're planning to generate. Again, this is a new way to access our platform. We hadn't made it available like this directly to customers before. So far, people have been really excited to get access to it. So maybe we should have done this sooner, but we're doing it now. Okay, so that illustrates the shifts we're making with our cell engineering business, but I'm now going to turn to what we're seeing within our biosecurity business, especially with the recent emergence of bird flu, H5N1. Okay, so on the left here, you can see some recent articles about the federal funding as well as state and countrywide plans to help curtail the spread of H5N1. But I want to focus more on the right and the timeline about why this is coming to light today. So H5N1 is not new. It was first identified in 1996. There have been various bouts of it over the years, but the first major step change occurred in 2020 when HPAI, or highly pathogenic avian influenza, was detected in Europe, which then traveled over to North America in 21. And since January 22, 48 out of the 50 states have seen outbreaks of H5N1 among poultry impacting over 100 million birds. You might have heard about this as a big deal in the poultry industry. Another step change occurred at the beginning of this year when the virus unfortunately mutated again and became transmissible to mammals, specifically cows. Obviously, it's not great. Mammals are closer to us. It's not the end of the world for humans because the virus can be pasteurized out of milk. There's been a lot of news about that lately. Don't drink raw milk. And it can be cooked out of beef. But there is nothing stopping this virus from mutating yet again into something that could be transmissible to humans. So I'm not trying to scare you with this, but this is another example of how important persistent pervasive monitoring is. We want to catch and crush something like this before hundreds of people are showing up in a hospital with symptoms. Ideally, we detect it much closer to the animal source that it could jump out of. So now let's go forward a few months to April of this year when the USDA announced an action plan to protect livestock from this particular variant of H5N1. They announced over $800 million in new funding to combat this virus and mandatory testing of dairy cattle that are moved across state lines. So from our previous work surrounding bio threat monitoring, we know that there are three keys to a successful plan to detect and combat a biological threat, particularly around livestock. So the first is we need to find a way to gather information pervasively. Second, we need to collect genomic information regarding the virus without adding much cost or time to those information gathering plans already there. And then third, we need to find a way to work with the communities that are impacted while respecting their privacy and concerns. And let me tell you, we learned an extreme form of this when we did millions of COVID monitoring tests during the COVID outbreak in K-12 schools. Okay, the privacy and parental concerns there were huge. And we have a ton of learnings from scaling that just gigantic business at the peak of it so well there. Now, in response to these needs, I'm excited to announce Ginkgo's proposed genomic analysis program, GAP, for H5N1. Ginkgo will use the existing practice of pooling and sampling milk for food safety and add the capability to generate genomic analysis of the H5N1 virus. This provides critical data for the science needed to respond to the virus without adding any extra burden to farmers or the systems they depend on. The process is non-invasive, requires no additional time or logistics from the farm. Importantly, the program does not record or transmit the source of the milk. In the GAP program, the only information captured is the genomic data of the H5N1 virus itself when it's detected. Okay, so this can be done in a way that's not disruptive to the existing industry. Now, if Ginkgo's pilot plan is successful, we will begin sequencing H5N1. If we are successful at sequencing the virus, our sequences could potentially be used by pharma companies to develop drugs or vaccines to combat the spread, give you extra time to get started on those things. And lastly, through our sequencing efforts, we're also looking to detect harmful variants, specifically ones that could be transmissible to humans. If this does occur, we're working on developing partnerships to enable rapid scale-up of testing similar to what we did during the COVID pandemic to help get resources to the communities that need them most. You'd like to, again, test in those areas where things are happening. Now, the spread of H5N1 may never evolve into a human transmissible disease. Let us hope so. But H5N1 shows us how vulnerable we still are as people, as a society. We wanted to detect anomalies, in other words, where things differ from the norm, a sequence that you hadn't been seeing, as soon as we can so the industry can protect their herds. uh and way of life and we can all be safer uh and if when h5n1 does become a risk to humans ginkgo and its partners stand at the ready to monitor detect and intervene if that time comes i've said this before but we should monitor for viruses like we monitor the weather like we like we watch for hurricanes right uh you know We're watching all the time. We have a system for evaluating the risk of a storm when it's brewing. You know, what category is it? H5N1 is a small storm at the moment, but it has the potential to be a category five, and we should have our radar running all the time. And so hopefully this pilot work is the start of that. In conclusion, although the second quarter was a difficult one here at Ginkgo, as we had to say goodbye to hundreds of friends and coworkers, I'm proud of what the team has accomplished truly, continuing delivery for our customers and opening new avenues for growth in both our tools offerings and H1 offerings. We remain laser focused on our goal to reach profitability while leading the development of the technology that makes biology easier to engineer. All right, now I'll hand it back to Megan for the Q&A.

Great, thanks Jason. As usual, I'll start with a question from the public and remind the analysts on the line that if they'd like to ask a question to please raise their hands on Zoom and I'll call on you and open up your line. Thanks all. All right, welcome back everyone. As usual, we'll start with retail question and then we will go down our list of analysts. So Tejas, you'll be up first after our retail question. The first one comes from our investor inbox and it's for you, Jason. What is the current outlook on some of Ginkgo's tech? Any signs of breakthroughs in the next one to two years?

Yeah, so we talked a little bit about opening up our platform directly to customer scientists, sort of like democratizing access to our infrastructure. I think the great thing about that is it doesn't require a big tech breakthrough. Gingo scientists have had access to all this infrastructure. We've been able to use it very successfully across customer programs. We have that evidence, but it just hasn't been made directly available, again, to scientists that are customers. And so we have a number of interesting things there, like we shared last time around how effective the rack automation has been in-house, both compared to traditional automation and certainly compared to doing work by hand. Getting that out into customers' hands would be amazing. Lab data as a service, we're seeing pickup on that now, even in interesting new areas. That is leveraging a whole suite of Ginkgo infrastructure, like a flow through both our assay technology, our build technology, a whole number of different things packaged together. Again, that's not something that a customer scientist could just shoot off a bunch of designs and get that kind of data back before. So I'm pretty excited about that, building large piece of DNA, computational design tools. We have a ton of in-house infrastructure that really are already tech breakthroughs, and we just need to make them available to folks. So that's sort of what I'm most excited about on the tool side.

Thanks, Jason. Like I said, Tejas, you're up first. Your line is now open.

Great. Can you guys hear me okay? Yeah. Hey, Tejas. Hey, so Jason, maybe I'll kick it off here with one on the simplified tech and the automated system stuff that you've been working on. How has the progress been with accelerating the startup time for new projects using the racks you currently have in your Boston facility? And are there any metrics you can share to help us sort of benchmark progress? And similar sort of question on lab data as a service, just any color on that. traction there relative to expectations and what type of customers do you see showing the strongest demand in terms of the order funnel for the lab data offering?

Yeah, so let me, I'll go in reverse order. So on lab data as a service, yeah, we are seeing traction. I think the biggest, so the biggest category for us, I think is going to be companies that are like, you can call them like tech bio companies or AI in bio companies, but you have a whole bunch of either new startups or occasionally like a group within a large biopharma that's really trying to build an AI model that's sort of proprietary to them to help them discover drugs. And they're often starting with some of the public models that are out there. Maybe it's some of the protein design models, the ESMs or the alpha folds or things like that. But then what they want to do is fine tune those models with a bunch of additional proprietary data. And so they want to generate large amounts of this data. And that drives them quickly. I showed that chart of the lab bench and the automation. There's no way to generate that amount of data without deploying automation. And so our usual fight with our whole approach, even if we're selling a solutions deal, is sort of to say, hey, for this type of biotech problem, we think generating a huge amount of data is really the way to solve the problem. And what's great about the AI bio folks is they believe that implicitly. Right, like that is foundational to model training is having large amounts of data. So that's a real boon for us. And I would say that's really gonna be the big demand set. Mark mentioned, we have even got a deal with like a large tech company, right? Like there's even new entrants into this space that are kind of interesting, but in general, that category will be first. And then I think you'll see the AI efforts intra the biopharma companies as sort of the next layer of demand for lab data as a service. Does that make sense? Yeah, that makes sense. Then on the automation side, onboarding things and racks. I think what's great about the rack system is we've had our group out in Emeryville, California. This is the previous Zymergen site and team that has been basically developing and improving on the rack hardware since we acquired Zymogen a few years ago. They made it much more manufacturable, much more scalable, cheaper to build it out, improve the software. And they've sort of, Ginkgo has almost been, you know, Ginkgo in Boston, like an external customer to that group. And so we've gotten a lot of experience sort of deploying racks quickly. So I'd say the thing I'm most excited about is our ability to add Ginkgo, say, hey, we need this new set of equipment added to our rack setup. getting that quickly assembled into a rack, shipped to Boston, and then just plugged into our existing automation setup by attaching a rack to it is really exciting to see in practice. Because that's just so different than a traditional automation setup, right? If you go to a traditional automation vendor, your ability to expand the existing inbuilt infrastructure you have is basically zero. If you want to add to your system, you got to go through a whole new process. You got to sit down with them, plan it all out. Whereas we are now just being able to add new systems. We've now expanded the system in Boston twice very easily. And so that I think is probably the bigger proof point. When it comes to doing our solutions business, we leverage the racks. We leverage the automation there. We also leverage other infrastructure at Ginkgo. So it really depends on the project, how much the racks are really driving change. But as an example of how quickly you can bring in new equipment into an automation setup, I think Ginkgo is a proof point. And I'm hopeful that we can actually bring that to other customers. We want to have those racks ultimately land in the hands of our customers' labs. And Ginkgo is a great proof point for that.

Got it. Super helpful. And then one, maybe Mark can chime in here as well. Jason, curious to get your perspective on, you know, decent quarter on both sides of the house, biosecurity as well as, you know, the cell engineering piece. You kept the guide sort of intact despite that. Is that just you basically de-risking the back half of the year? And the context of the question is, you know, recently, you know, earlier this week, one of the largest sort of preclinical CROs talked about a very rapid deterioration of the demand environment on the global larger pharma side of things. And the biotech improvement that's coming through is coming through slower than anticipated as well. So I'm just curious as to what you guys are seeing from those two customer constituencies. And then as you sort of contemplated the back half, was that really sort of playing in the back of your minds as you thought about the guide?

Yeah, I'd be happy, Jason, to start on the question. And Tejas, I'll start with biosecurity first, just because I think that's a little bit easier than we can get into cell engineering. So in the case of biosecurity, the revenue really popped in the quarter, but you should not interpret that as sort of a new run rate. for biosecurity right now. What happened there was we in effect had some revenue booked in Q2 that we thought originally would have been scheduled or was originally scheduled for Q1. And so you had a little bit of sort of, I would just call it lumpiness tied to a customer contract in Q2. And so you sort of want to average Q1 and Q2 on the biosecurity side and and not think of Q2 as kind of a run rate. And so that's why we kept the biosecurity guide intact. On the cell engineering side, first of all, we are going through a restructuring. And so we considered that when we brought the guidance down back in May. And we are pleased with the revenue number this quarter, but And there's still a lot of work for us to do in terms of both the restructuring as well as the changes we've made in terms of the commercial terms, etc. And so we just felt it was appropriate to keep it where it was right now. Got it.

And then last one for me, Jason, a bit of a fuzzy question for you, but organizationally in light of all the headcount cuts, I'm sure there's roles or there's people in the organization whose roles have evolved, expanded, changed over the last few months and probably more to come in the next few months as well. Where are you in terms of that process? Are you already in steady state? And if not, Mark, I mean, to your comment there, Can you just help us think through the degree of sort of disruption from that that you anticipate or have already contemplated in the guide? Thank you.

Yeah, I can speak to that. So yeah, we made a lot of those changes right out of the gate. We sort of restructured a little bit, created some business units within the company that are smaller scale than the entire organization to drive a little more P&L accountability and had leaders in charge of those. And all those changes were made alongside, like during the restructuring process so that those leaders could make choices about who would be on their teams and how to do that work. And this was really important to us because we felt if we were going to deliver well for our customers while spending less doing it, you know, wasn't just as simple as, you know, keeping things the same and just having less people. We were going to have to do it in a better way. And so I think, again, I'm along with Mark. I'm pleased. with our revenue this quarter and how that process has been going. I think those major changes have occurred. There'll be more bits and pieces here. It's not like you're going to get it right exactly perfect on the first shot, but the major changes have been made. Got it. Thanks, guys. Appreciate the color. Yeah, thanks.

Thanks, Tejas. Next up, we have Matt Sykes at Goldman Sachs. Matt, your line is now open.

Great. Can you hear me? Yeah. Hey, Matt. Great. Hey, Jason. Hey, Mark. Hey, Megan, thanks for taking my question. Maybe first question for me, just how do you make sure your commercial capabilities, so I'm talking about like footprint, knowledge base, skill set, match with your shift into the tools market, particularly in light of some of the cost reductions you're making?

Yeah, so I mean, I can speak to that a little bit. So I think one of the things I like about us entering the tool space is I think we can kind of ramp into it nicely, right? Like I mentioned this with regard to like the automation, but I think this is effectively as true when it comes to like the lab data generation. The scientific team, so if you looked on that, I showed that slide where I said, hey, we're selling our customer with the head of R&D and we were an outsourced research team delivering them a scientific outcome. Well, the way that was structured at Ginkgo was we had Ginkgo scientists who were basically internal customers of our platform. That interaction, I mean, it's a little unfair because they're all under the same roof, but we had so many of those teams. We have 100 active R&D projects, all with a leader for each one, all ordering lab work from the foundry at Ginkgo. And that's not too different from receiving lab data as a service request from an external scientist. So the nature of how Ginkgo managed to do our work, the way we're able to scale to so many projects across diverse markets, ag, industrial, biopharma, all with the same infrastructure. And I know we were spending too much, but it's still, I think, a quite impressive amount of lab research to be doing all in one place efficiently was because we were already operating with a certain tools like model internally. Right. And so, so I am, you know, again, we'll see how it goes, but I'm optimistic that by pointing that externally, there is going to be some changes, but you know, maybe, maybe less than you would think just the nature of how Ginkgo did our work.

Got it. Thanks for that. And then maybe just kind of conceptualizing the tools business and, and just in the context of the kind of environment we're in, particularly with biopharma customers, It seems like a lot of what you're offering is sort of would come out of like an OpEx budget for a customer, you know, libraries, software, AI models, things like that. Then there's the CapEx side, which I can think of rack systems, but I'm curious about others. As you think about sort of maybe in a longer term at a high level, sort of the split between sort of an OpEx purchase and a CapEx purchase from your customers, what do you think the tools business will look like And what would you like it to look like going forward? Yeah, this is a very good question.

Yeah, so I think when it comes to CapEx, I think the only thing we really have in mind are the rack systems. So we don't have other sort of things in mind there. And we don't see ourselves as sort of expert, you know, specialized equipment developers or anything else like that. Where Ginkgo really had a unique need, and this is true of Zymergen too, frankly, which is why the technology was getting developed there at the same time, was for flexible automation. right and and so i think that's the thing that we can bring out on an equipment basis uh that would tap into capex budgets but even then you know a lot of how we see that sale happening is we've got our cloud software on top of it that allows you to essentially add more equipment to a setup and then download new protocols and be able to do new things right like so there is also the opportunity for sort of a a sas uh, type of business on top of it. And we, and, you know, design before we acquired them had actually already sold a few systems like that. So over the last couple of years, we've had external customers that do pay us, uh, you know, it's a small thing, you know, it's only one customer too, uh, but they pay us a SaaS, um, license for the software. So, and for, and for, uh, services. And so the, so that, that's, I think that, that model, otherwise everything would come out of OpEx, I think pretty much. um so when we offer like lab data as a service and things like that we really see it as a still a services business just no no royalties right it would really just come you know straight in uh as used um when the customer is buying it right so or when the customer is consuming it we'd be getting paid directly um so it's a bit different than the solutions model where you have these like longer time horizon things that are really going to be ultimately the source of the profit here we'd have to we'd hopefully be driving a margin on the sales directly of the services

Got it.

Thank you very much.

Thanks, Matt. Next up, we have Steve Ma at TD County. Steve, your line is now open.

Hi, Steve. Oh, great. Hey, how are you doing? Thanks for taking the question. I got a two-part question on the reduction in force. Can you talk about your confidence and the ability to service your existing customers, new partners, and then also be able to onboard the new cell engineering tools business if it takes off? And then the second part of the question, can you provide some color on the scope of the RIFT? Did it impact all sites equally, or were there any concentrations to note, or any facilities closed, or any facilities going to be consolidated? I specifically asked that because on the Syngenta collaboration announced in July, I just wondered if that's being done at Sacramento and if there was any impact to the Sacramento facility.

Okay. So I'll, maybe I'll work on, I'll speak to the first one and then Mark, if you want to talk a little bit about facilities. So, yes. So this is like, it was a number one concern for us. Obviously we are a, you know, high touch white glove services business and protecting all of our existing customer relationships was really critical. And so that's why, again, I was very happy to see, you can see it reflected in the revenue, but behind the scenes, I think that went really well without a hitch in the sense that I don't think our customer programs were impacted in a really negative way, even going through the RIF process. Now, we did restructure how we're doing things, and that meant that different groups were affected differently. The folks working directly on customer programs were affected less than folks in indirect roles and things like that for a large part to make sure we did maintain consistency of delivery for current customer projects and the ability to keep selling new solutions businesses. So our commercial team that's selling in those product areas that we're maintaining are also protected and are still out there selling and adding to that business, as you saw even in the last quarter, which I think was a a tough quarter to sell during. You also asked about how we expanded to tools given that. And so that's where you'll see us make more selective growth bets where we're really asking a team to operate almost like a startup within Ginkgo. And if they demonstrate success, if we see a ton of purchases, whether it's on the racks, the AI stuff, the lab data as a service, and it starts to run, you'll see us push lots of resources behind that. Right. And I think, you know, Geeko has a good history of this, right? Like I think our experience, for example, during the K-12 COVID testing, you know, that, that grew to a, I don't know what it was ultimately, you know, 300 million plus revenue business in a matter of 12 to 18 months, because we had a strong pull and we were able to go behind it, productize it and scale against that demand. If that happens in any of these tool areas, I don't think we'll have trouble scaling behind it. I think we're, we're those kinds of people. But we do need to, kind of turn those cards over and see if any of them is an ace, uh, that's, that's worth, uh, pouring a bunch of resource behind. Um, but we're certainly kind of, we're like running the experiment and if it hits, you'll, you'll see us lean more into it.

Mark, Mark, do you want to speak to the, uh, yeah, I'll take the second part of the question. So first of all, West Sacramento wasn't, um, and isn't going to be impacted in terms of site, uh, rationalization. So we're happy with the agriculture business. We've rolled up really good capabilities. It is centered at West Sac, as you know. And yeah, there's a little bit of excess space there. So you'll probably see us do a few small subleases on that excess space sort of opportunistically. But no, that site remains. Generally speaking, Steve, the RIF was fairly broad-based, meaning that it touched, I would say, all parts of the company when you think about operational or foundry-type departments, as well as G&A functions, just as well as various locations. So I would just describe it as pretty broad-based as opposed to focused on a particular location.

Okay, thanks for the call.

And maybe the only thing else I would add about, you know, you will see us like a big thing. We had to get out of these spaces. That reduces the kind of overhead costs of maintaining them, the HS costs and the facilities costs and everything else. That already saved us a lot of money. And then once we're out of them, we would like to sublease them. And so I think you will also see us be strategic, right? Like if someone wanted to rent all BioFab 1, Sounds good to me, right? Like we could stay right here in dry dock, right? Like we will be hungry for where we can pick up cash quickly so that I can offset our cash burn and make sure we have the stability to grow this business, right? So you'll see us be opportunistic about where we see where we see sublease opportunities with the real estate. OK, thanks for the call. Yeah.

Thanks, Steve. Next up, we have Mike Ricekin at Bank of America. Mike, your line is now open.

Hey, Mike. Hey, can you guys hear me? Yeah. Yeah. Okay. Excellent. Yeah. Um, a couple of quick thought questions first on the, you know, the LDAS workflow, some of the things you highlighted, um, obviously you guys have a lot of capabilities here. You sort of take it down, break it down component by component. Um, and certainly there are, um, a lot of more traditional pharma that could probably benefit from, uh, you know, improved automation, um, you know, incorporating AI, um, and just sort of high throughput, um, screening for some of these, uh, programs, but you're not the only ones doing this right now. There are a number of companies that have come up over the last five years that are doing some sort of high throughput, functional genomics, antibody development. Um, you know, if you, if you really look at each of these capabilities, this is technology that already exists. So I'm just wondering, you know, you have the capabilities here, but you sort of are coming out a little bit later than some of the existing, um, players. So, um, how do you view, you know, catching up in that space in what I think actually is a relatively competitive space when it comes to some of these new, you know, large multimodal data sets.

Yeah, I mean, I think it depends. It kind of feathers out pretty quick, right? So I think you do see a couple of players who sort of invested early in the AI space, built up lab infrastructure, like recursion or something, or I do think they have really proprietary good tech, but again, are largely using it for their own pipeline and not really trying to do like a broad based services thing. Maybe they change that in the future, but I think that's the gist. Among traditional CROs that are out there, I think like take antibody developability. Okay. I think on the antibody binding side, you have a lot of players, but when it comes to developability, like those assays are kind of tough to do at high throughput. And honestly, with the customers that are coming to us are coming to us because they can't get hundreds or thousands of data points in developability. So I don't actually think we're behind there.

Okay. Yeah. I don't want to name any names, but I think we all know who the existing players are. So we're talking about the same ones. Okay. And then on the H5N1, that was an interesting overview and, you know, something we've been following for a couple other different reasons. But just curious, you know, that proposal, the genomic analysis program, all that makes sense. Any, you know, any timelines we should look to? Is this just sort of like, how do you get this from theoretical to, you know, actually implemented?

Yeah, so where we are today is we're currently entering into agreements with dairy farms for sort of H5N1 testing under that pilot. I think if those go well and we start to be able to show data coming out of that, then I think that's what opens the door to ultimately tapping either government or private sector money there. So we'll have to just see how the kind of data gen goes in the coming months. Awesome.

All right. Thank you. Yep.

Thanks, Mike. Next up, we have Mark Massaro at VTIG. Mark, your line is now open.

Great. Can you hear me okay? Yeah. Hey, Mark. Hi. Great. Hey, guys. So my first question is on the lab data as a service transition. How should we think about the potential or expected value or maybe the economics of these types of deals relative to some of the existing cell programs. So you have 140 active cell programs. What I'm just trying to determine is like, is one of these LDAS projects comparably similar to your existing cell engineering programs? Just any intel on expected value or economics, I think would be helpful.

So they're much smaller in size than a typical end-to-end cell engineering program. They would also be much shorter in duration. So the revenue would burn faster on those. And I think the idea though, is that a customer is not just buying one LDAS project from Ginkgo, but they, after buying it, getting data, they're buying more data and more data. And so that customers become It could very well be that the aggregate amount that's purchased would look like a larger cell engineering program. Jason?

Yeah, I think that covers it. I think we're a little bit early and it marked it to no, but I would say the gist is faster to close, less total dollars, shorter period of time. And so each one, you shouldn't think of it like it's the same as a cell engineering solutions or just early. So I'm actually hopeful we end up having lots more of these LDAS than we would be able to close of cell engineering solutions deals. And it pretty much has to be that way for this business to work. So they won't end up being apples to apples. The solutions deals are great. They just... They take a long time to close. Each one is like a research partnership, right? Whereas the LDAS would be much more transactional. And I mentioned this a bit on the call, but we'll likely engage through procurement if we're dealing with a larger company there. So it'll just be something people are buying off the rack.

Okay, great. And then my other one is just, you know, recognizing that anytime you do a riff, it's difficult. And so, you know, I completely understand the difficulty of that. I think I heard you guys say that 300, if I have this right, 300 individuals were notified or terminated by the end of Q2. I think there might be another 150 that may be notified by mid-25. I guess... How are you messaging this internally, mainly just to keep people motivated and keep people incentivized to keep doing good work for Ginkgo?

Yeah, and just for clarity, those additional 150 have already been notified. It's just folks have periods of time where if there's a program that's concluding or things like that, that they're there for a period. And so that's the biggest thing. We wanted to try to do as much of that at once as we could. And then, yeah, look, I mean... I think people believe in the mission of Ginkgo. I think we're trying to do a hard thing. And so I think there's a lot of motivation, like a pool of motivation there. But ultimately, I think what will drive people is to see continued success, right? So as we see ourselves building success in Lab Day as a service, building success in robotics, as we are actually delivering on our cell engineering solutions business, getting closer and closer to profitability there. Those are the things that are going to drive success. kind of momentum and, you know, good cultural energy. I can't go, that's it. It's simple as that. Like, you know, the only way out is through that for sure. Understood. Thank you for the clarification. Yep.

Thanks, Mark. Our last set of questions will come from Matt LaRue at William Blair. Matt, your line is now open.

Hey, good morning. Can you hear me? Yeah, I'm out. Hello. Okay. End of burning season. So, you know, you're now at a quarter since you announced sort of the plan to achieve just an EBITDA break, getting here by the end of 26. You've talked certainly about the RIF, and that's going to be a piece of it. But also the composition of revenue that you're expecting over time is different than obviously when you initially went public in terms of now contemplating tools. Obviously, LDAS, I think, being a newer opportunity. Just how are we thinking about what the revenue mix will look like at that level? And maybe to that point, are you expecting LDAS at scale and your tools entry to be things that can be

uh you know margin accretive to perhaps the the legacy's whole engineering business at scale yes uh very good questions so i think we will see on the tool side like i said earlier like the particularly in the next the second half of this year we're going to be turning over a bunch of cards there to know how well that's going right and if we see paul and you know You'll see us march behind that. And then absolutely, I would expect to have that obviously be then a big contributor. We don't want to have that be a requirement for us to get to profitability. And so if those weren't hitting at all, then I think you'd make different decisions about what you're investing in. You would pare back how much you're investing in those things. But ultimately, the solutions business, we believe we could scale up ultimately to a profitable Ginkgo on that. It's just a matter of you know, cutting and right sizing and making sure we do it right. But, but I'm, you know, I think the tool stuff is neat, right? Like it's the, it's a proven platform. We have used it for hundreds of projects. And frankly, I'm pretty excited to get that democratized and out in more people's hands. Right. So, so I I'm, I'm optimistic. We'll see that, but we don't want to have to have that work.

Yeah, no, I agree. I think it'd be really interesting to see what, uh, what uptake looks like there.

You know, like, like another thing I think it has been good at over the years and, and obviously like, you know, doing the restructuring is another example of this is, you know, like the mission stays fixed. We want to make biology easier to engineer, right? Like that is, I promise you that that is the goal, you know, that is, that is the mission of my life. Uh, and a bunch of the folks, particularly, uh, early folks and longtime committed folks at Ginkgo, how we go about doing that. You know, it depends what we learn in the market, right? It depends what we learn on the technology side, right? Both ends of that are changing all the time. And so I think we are still excellent. You know, we have an excellent position in the market to pursue that. You know, we have a great cash position. We have a ridiculous technical talent. Over the last couple of years, we rolled up a lot of the core technology in the industry. So I think we have the best technology base. So if you believe that you can fundamentally change how genetic engineering is done, I think Ginkgo is still your bet. And we will remain flexible to where the best way is to do that.

Yes, exactly. I think my next question, which is clearly sort of an emphasis just to add scale, right? And whatever kind of program format that takes. And so, again, if I think back a couple of years ago, obviously a barrier initially was educating customers, both from a technical offering, but also legal structure, potential downstream value perspective, right? So just as you think about going forward, obviously you added programs this quarter with DVS and you're going to continue to, but with an emphasis on scale, how do you kind of balance, you know, which, where you pick your battles to some extent and you're battling your customers, you know, but, but where, where that is really important to you versus, you know, just adding scale and, you know, new customer relationships, you know, that kind of thing.

Yeah. So, so a couple of comments on that. I think for, So I talked about this on the last call, but the two points of friction with customers are our reuse of intellectual property on the foreground IP, like the work we do that they're paying us to do. And us getting a right to reuse some of that. And then the second area of friction can be around sort of royalties and milestones on commercial success of the product. What we found as we've gone back to customers and offered to kind of turn those knobs is that the IP one is a real point of friction. And honestly, when it comes to a solutions deal where we're doing a complex project over a long period of time, putting in a lot of ginkgo scientific effort and taking technical risk, the downstream value share is not as big of an issue, right? Because... They're also concerned about whether the research will work, right? Like biotech is, they call it research for a reason, right? Like, you know, like it sometimes doesn't work. And so when it does work, they're pretty happy to share the upside, right? So I do think that's like the correct perspective. pricing strategy for solutions is to include downstream value share. That's why I did want to point out that, yes, we've signed more deals with it and we'll keep doing that. I think it's great. We report occasionally on the total amount of milestones. We have tons of potential milestone payments in the future. That's all very exciting. It takes a while to get to it in biotech, but that sort of upside potential is really nice to have in the solutions business. We should keep signing it up. The volume problem there isn't really about that. It's just about how much willingness there is to fully outsource a research project. That's something that many companies consider a thing they need to do themselves. And so that's where I do think we have a much better opportunity. We don't necessarily need to re-educate. I think we can focus instead. That's why I mentioned the AI first customer set for LDAS. We can focus on the niche of biotech customers that believe in big data. and just give them every tool they need, right? Because Ginkgo has believed in big data generation for the last 10 years. And so let me tell you, we have built out amazing tools for our scientists, again, at the DNA construct design level, like the ability to build large pieces, to build many pieces, pool testing, you know, sorry, like barcoding and pool analysis, all the automation, flexible automation. Those are all like really great tools. if you believe in generating a huge amount of data to do the work of genetic engineering. And honestly, if you don't believe in that, your bench is sitting right there and the thermocatalog is on the shelf. Like, go ahead and get to work. You're fine. I'm not going to try to fight that right now. I'm going to talk to the people that do believe in it, but are underserved on the tool side in terms of large data generation. And I'm going to just, I'm going to nail that niche and then we're going to grow with it.

Okay, great. Thanks, Jason.

Matt, I was just going to chime in on your points about balance and scale. The other thing I would add is that we do have a more focused on the upfront sort of cash economics on a deal, just relatively speaking. And so that's obviously you can grow here by toggling that. And we did that in the past, but we got misaligned with our cost structure and revenue. And so you will see us, we will be thinking a lot about the nearer term cash kind of service fee economics before entering into a new contract with a customer.

Okay. Understood. Thanks, Mark.

Thanks, Matt. I'm not seeing any other questions in the queue, so I'll hand it over to Jason for some closing thoughts.

Yeah. So again, I'll say again, tough quarter for Ginkgo. I want to say thanks again to the department employees that we had to let go as part of the riff. I think we are doing the right things to kind of create a solid base here at Ginkgo and build on it. And we appreciate all the support of those listening in, in our mission to make biology easier to engineer. So thanks again.