In a new episode of SaaS Conversations, we cover technology due diligence. Lauren Kelley (CEO, OPEXEngine by Bain & Company) leads a conversation with Hank Chen (Partner, Bain & Company), who heads Bain & Company’s technology due diligence practice in the Americas. The two discuss how tech due diligence works, AI’s impact on R&D investments, and frameworks for understanding the quality, scalability, and sustainability of an asset that an investor is buying.
Click here to listen to the episode on Spotify.
Technology Due Diligence
“Does the promise of generative AI really follow through? Are there real meaningful revenue upside and cost synergies that are leading to better EBITDA?” - Hank Chen
Host: Welcome to SaaS Conversations, a podcast from OPEXEngine by Bain & Company.
In today’s conversation, we’ll cover technology due diligence. This is an important topic not only for investors, but also for the technology companies that are speaking to investors – it’s critical to understand how investors are going to look at your company and approach diligence before you go and tell your story … or before an investor evaluates weaknesses that you don’t know about.
Hank Chen: “Does the promise of generative AI really follow through? Are there real meaningful revenue upside and cost synergies that are leading to better EBITDA?”
Host: That’s Hank Chen, a partner in Bain & Company’s Boston office and head of their technology due diligence practice in the Americas. In this clip, Hank is speaking about AI’s impact on R&D investments specifically.
In a larger conversation with Lauren Kelley – CEO of OPEXEngine – the two discuss how tech due diligence works, along with frameworks for understanding the quality, scalability, and sustainability of an asset that an investor is buying. They talk through the importance of product differentiation, managing tech debt, presenting a thorough AI strategy to investors, and the key questions that investors have for SaaS companies in 2024.
In this episode, you’ll hear the latest on the tech frameworks that investors are using in 2024 to evaluate SaaS companies.
Let’s listen in.
Lauren Kelley: Welcome, everyone. I am so excited for today's discussion with Hank Chen, a partner at Bain & Company's Boston office. Hank is head of Bain's technology due diligence practice in the Americas, and we're going to be talking today about the value of tech due diligence, how it works, and the importance of a framework for understanding the quality of the asset that an investor – whether that be a private equity or a corporate investor – is buying.
Hank, thanks so much for being here today! You joined Bain a couple of months before my company OPEXEngine was acquired by Bain. Tell me a little bit about your background – how did you end up at Bain?
Hank Chen: Yeah, Lauren, it's an honor and a pleasure to be here. Thank you for inviting me. You're correct. I joined Bain three years ago in July of 2021 to build out this new capability and practice within Bain. My background is 25 years really at the intersection of exactly what we do. Half of my career was in technology, where earlier I was a chip designer in the telecom space.
Then, I spent a number of years in enterprise at IT companies. I was at EMC for a number of years, and then I was at a startup in the hyperconverged infrastructure space called SimpliVity. And the other half of my career has been in strategy consulting, so a lot of what we do really is at the nexus of understanding technology but putting it in a lens that investors can really understand.
Lauren Kelley: That is so cool. I had no idea that you started out as a chip designer. That’s a pretty interesting background – particularly in 2024 given the focus on chip investment right now.
Let's get started. What are the main issues concerning investors today with regards to tech due diligence?
Hank Chen: Yeah, there's a number of perennial topics, Lauren: product differentiation, scalability, tech debt, R&D organization health, cybersecurity. These are topics that we talk about on almost every single asset and these are, I would say, perennial topics.
On top of that, there's another category of issues that are more situational. Often, that is a target that is putting a bet on a new platform – so they're either putting out a new platform or they're transitioning to the cloud and trying to migrate customers from their legacy on-prem application to the cloud. And we're in the process of either assessing their progress against that migration or how well they've done in building out a new cloud solution.
And then thirdly, I would say one of the key new topics everyone is thinking about is generative AI (GenAI) disruption risk – so that's a third area that we spend a lot of time thinking about today.
Lauren Kelley: We see that a lot on our side as well – questions about how GenAI will affect costs and expense structures as well as headcounts, and we're already starting to see that play out In the benchmarks a bit.
We're excited because [at OPEXEngine] right now, we're crunching all the final 2023 numbers from our customers and from private equity firms that contribute their portfolio company data. Definitely in 2023, EBITDA improved across the board for most SaaS companies. I'll be super interested to see when the final numbers come out next week how that has affected headcounts and employee productivity. We hear so much about it on the R&D side, and that's where we really want to dig in and see how much it's affecting cost and organizational structures.
Can you talk more about any early findings you're seeing in terms of GenAI disruption in R&D? Both in terms of new products, and internally how companies are organized?
Hank Chen: Yeah, absolutely. And I fully believe, Lauren, that benchmarking could provide a really interesting avenue at looking at the issue.
I think every private equity investor or corporate investor out there will want to understand whether there's been any real returns given all the talk that's gone on over the last 18 months or so. Does the promise of generative AI really follow through? Are there real, meaningful revenue upside and cost synergies that are leading to better EBITDA?
That’s a huge part of where, I think, OPEXEngine has a great role to play. In the work that we do, I would say there's three categories of questions that investors are asking. First and foremost, investors are very concerned as they think about approaching an exit for a particular portfolio company, or even on the buy side as they're looking at various targets.
And the key question there is: how much GenAI disruption is there for the particular company we're looking at? Will GenAI somehow disrupt and disintermediate the value that the company is bringing to the table? And what does that mean for the value of the company going forward?
That's a critical question. So one type of work we will undertake is performing a GenAI disruption risk assessment. And then along with that, investors are often interested in understanding potential upside in terms of: what GenAI products or GenAI-powered products can I build in my portfolio to further drive growth and my revenue ambitions?
And then on the cost side, certainly thinking about the wealth of different opportunities to leverage GenAI to then reduce headcount costs, whether that's on the R&D side building and using GenAI tools to assist in developing code or other sorts of automation opportunities.
Lauren Kelley: That's super interesting. So really, you look at three things in terms of GenAI:
First, it's whether there's a competitive issue that other companies are going to disrupt a potential asset or acquisition's position in the market because the other companies will be better at using GenAI to be more competitive – that's one thing that you look at and you do a GenAI risk assessment there.
Secondly, you look at what new products and opportunities for market position and/or revenue growth that target might have in terms of using GenAI in their products.
And thirdly, the effect internally on cost structures of R&D organizations.
Hank Chen: That's exactly right, Lauren. On the first one, I would say it's even more than just looking at it from a competitive lens.
It can be as critical as an existential question for the company. Will this company continue to survive in the same way it's survived up until now? Or will GenAI disrupt its business so much that it can no longer generate the same types of value it has and therefore is worth less?
Lauren Kelley: Which is really interesting. If you think about 2021, there was just a massive amount of money invested in the tech industry. It wasn't just SaaS, though it was significant in SaaS. In every product category, so many companies were funded.
And as things tightened up – starting at the very end of 2022 and then continuing throughout 23 – growth rates started to fall, and investors and boards were telling companies: focus on a longer runway with the money that you've got, focus on your cost structures, and improve EBITDA.
So even though in the SaaS sector, the sector overall is expected to continue growing at a 20+% growth rate, not all the companies are going to survive and continue with that growth. And what you're doing is really identifying the ones where there may be existential threats to them because the technology is changing so fast. That's a really interesting point.
Hank Chen: That's absolutely right, and that's on every investor's mind right now.
Lauren Kelley: And do you see a lot of focus on the R&D costs? Have you been able to – and I understand that it's still early – see where companies are utilizing GenAI effectively within the development space to improve their cost structures?
Hank Chen: There is definitely interest there. We are actually in the midst of working with a private equity portfolio company in the insurance space to think about how to deploy GenAI to streamline the overall claims flow.
And this is a product that they sell where when a claim is submitted, often there are photo images that are submitted that substantiate the claim, but there is known to be fraud in the system. There's also known to be time pressure on these claim reps who are manually looking at the claims and have maybe five to seven minutes to adjudicate a claim, which means they don't always get it right.
And what we're trying to do –
Lauren Kelley: As we all know!
Hank Chen: – yes, all of us have experienced some form of that, right? Or maybe there's errors made in our favor where we get a claim paid out and we didn't really deserve that claim and didn't provide enough evidence. But what we're trying to do with GenAI is – if we can get to a much better level of accuracy with GenAI using ChatGPT vision or other tools to look at the pictures that come in and very quickly adjudicate whether a claim is approved, should be denied, or whether there's fraud involved – now, it becomes really interesting. And perhaps with the claim center that the company has, we can start to displace some of the labor there.
Lauren Kelley: Again, that's really interesting. So AI is applied both in terms of how they operate internally, but also to make their product so much better because there'd be less error.
Hank Chen: Very much so. Very much so. We think this has huge potential implications and potential cost savings for the company.
Lauren Kelley: Yes, and in a business like that, the cost of paying out for errors has to be considerable. And if you can reduce that, and make employees more efficient it's a double saving.
Hank Chen: Exactly right.
Lauren Kelley: When you think about giving advice to a company that is starting a process and opening themselves up to investor scrutiny or to a corporate acquirer, how important is it for them – in their presentation – to present an AI strategy specifically on their tech side?
Hank Chen: Yeah, that's a great question. I think today in 2024, it's critical. If you do not have your story straight on GenAI, that's going to present challenges for buyers looking at you. In general, I think there's maybe three pieces of advice I would give to anybody preparing for a process.
I think first and foremost, I would say it's really critical to have a very compelling vision or strategy that you tell others of what you've been trying to do with the company, how you've achieved the growth that you have, and your ambition for how to take that forward for whoever is going to be the owner for the next leg of the journey. Along with that, having that GenAI story and understanding what the potential risks are, what you've done to mitigate those risks, and what you've done to explore GenAI as an opportunity for your company I think is going to be a critical part of that story.
I think secondly, it's important to be honest and to not try to minimize, or hide the areas of the company that aren't working so well. I think as a diligence professional, we are in these conversations every day, and it's not very difficult to pick up where the management team is trying to deemphasize an area that is really critical to the investment thesis, but hasn't been well thought of or well-characterized. I think it's important to just be honest and just call out the risks. I think everyone will be on the same page on that. And it's better to be clear there and move forward rather than minimize or ignore the potential problem.
And the last thing I would say is to get your ducks in a row. I think it's really important to have your documentation in place and to have completed all of the necessary compliance code scans – all of the types of work that one will need to do to get their house in order prior to putting the company on the market. It's much like selling a house. You have to make sure to clean up all that clutter and stage the home appropriately. We're really talking about the same thing here.
Lauren Kelley: The parallels between tech and general business due diligence are striking. I was speaking last year with the head of America's Growth Capital, a big tech investment bank based here in Boston. He said that in 2021, the typical diligence and negotiation to acquire a software asset was roughly – from start to finish – two to three months. That could be an exaggeration, but it was definitely fast-paced.
He said starting at the end of ‘22 going into 2023, diligence was taking between six and nine months. Part of that was the depth that investors were going to in terms of looking at companies and that there was a new criteria that had come along – if a company couldn’t produce the information quickly and participate in the diligence process, that became a metric to evaluate a company by. It would be a mark against the company if they couldn’t do it quickly. So for example, in SaaS, if a company can't produce its SaaS KPIs or quickly analyze the unit economics of their customer model, it looks bad. The company is just not seen as being well-run, and it's interesting to hear that's kind of the same on the tech side.
Hank Chen: Very much. We are part and parcel of that process and participate in that process. Often, I think maybe one of the unique hallmarks of tech diligence is we actually spend a fair amount of time with the management team talking to them about their business and really going through some of the documentation they provided.
So we gain a pretty good sense up-close of whether the company is well-prepared or the management team really hasn't thought about some of these questions, which sometimes can be the case.
Lauren Kelley: What I heard from you as your advice to companies that are entering into a process – they need to be able to articulate a compelling vision for growth both on the product as well as on the business side.
They should have their GenAI story down – both the risks to them and the opportunities to them.
And then you also recommended being honest; don't fake it or minimize problems and have your ducks in a row.
We say the same thing on the business side – it's better to be honest and make sure that the numbers add up even if they're not quite as good as the way you want to spin them.
If I could play devil's advocate, have you seen that work well for companies? Where they've been honest and they got through the diligence process and it was actually a credit to them that they were open about it?
Hank Chen: Definitely, Lauren. I've been in hundreds of debriefs after the management meeting with the private equity investor, and part of what we're always sussing out as a consolidated team is: how did management really present themselves? Were they credible?
And there's always some sort of temperature-taking on that front. Usually, we're either in the room nodding our heads together saying they were well-prepared, they were honest, they answered the questions, they didn't minimize, and they were very straightforward.
Or, everyone's looking at each other and, not saying that anyone is lying, but gosh… there were some really stretched truths there that we called into doubt. And that will always just end up in more questions back to the company itself, right? So it's better to be upfront. It's always the easier way to go, I think.
Lauren Kelley: You don't want to be the company where in the debrief afterwards, everyone's looking around saying that somehow the story doesn't quite hold up.
Hank Chen: Oh, yeah. We deal with all sorts of different tactics for obfuscation, if you will. Whether that's bankers trying to manage our times to be a very short conversation, which then just requires follow up conversations or that's poor documentation where you have one-line answers that don't really answer the question or you get information that's redacted. There's lots of different tactics people try to employ, but at the end of the day, really what we're trying to do is have a good conversation and understand your business as an extension of the private equity investing team.
And if we can't do our jobs, then a private equity investor can't ascertain value.
Lauren Kelley: Yeah, you have these wild stories of WeWork or some others where people raise huge amounts of money, they do amazing transactions, and there's not as much there and everyone thinks that they can do that also.
And what people don't realize is: that's one in a million, and that there are a million other transactions going on all the time that don't operate quite that way.
Hank Chen: Yeah, definitely.
Lauren Kelley: Let me follow up on the tech debt question – that's something that we've been tracking as a benchmark at OPEXEngine for almost 10 years.
CFOs find it useful as an indicator of how their R&D investment is being spent. Usually we see a range of 20-25% percent on tech debt is a good benchmark. If there's less being spent than that, then there might be too much building up unless you’re an early stage company. And that's not good and it will cause problems at a later date. And if there's more than that on a regular basis, then it could indicate a problem with the overall team and structure.
How do you view what's good in terms of a company's tech debt efforts?
Hank Chen: Yeah, I think that sounds right, Lauren. The rough benchmark I have in my mind is 25% to maybe a quarter or a third of the R&D team's time and resources should be devoted to tech debt.
And I agree that it's very much a balance. Every single company has tech debt of some sort. It is impossible to have 0 tech debt. Therefore, it may be tough from a financial vantage point to look at that and say gosh, I'm spending so much money on R&D. Why am I not getting as much product out the door as I would like?
And certainly often that's a question that we get and we will spend time with a portfolio company to dig into that. But some of the R&D spend must go towards tech debt. Otherwise, exactly what you described will happen. It will accumulate over time and it will start to really overwhelm the company.
And you may not initially see it, but then eventually what you find during the diligence process is you have a collection of many different platforms that haven't been consolidated – some of which need to be sunset but you can't get the customer off because they haven't thought through migrations – and you're just left with a really sprawling tech estate that's costing you millions of dollars. That happens far more often than I would imagine folks would like to see.
Lauren Kelley: If you don't deal with tech debt then at some point you may end up in a roadblock where you just can't get out of it because too much has built up.
Obviously that's not where you want to be, which leads into my next question as we start moving towards wrapping up. If we had this conversation two years ago, we probably wouldn't have been talking so much about Gen AI, so 2 years from now, who knows what the next technology challenge or opportunities will be.
I’m assuming part of what you look at is your best estimate of how well this tech company is structured for scaling and sustaining their technology and adapting to change, given that investment horizons are typically a bit longer than 2 years.
You started this conversation by saying that at a very high level, tech due diligence gives the acquirer a framework for understanding the quality of the asset they're buying. And part of that has to be an analysis of the scalability and sustainability of the technology – how do you do that in light of fast-changing technologies?
Hank Chen: Yeah, absolutely. That is a very common question that we receive from investors. Often, as you can imagine, they're trying to acquire a platform so that they can scale and grow that platform. Whether that's through M&A and a buy-and-build type of scenario, or whether that's through organic growth, you can be sure they're wanting to scale that platform.
We look at it a couple different ways. We start with the lens of: what are the foundations? How has the house been built? What sort of technology and underlying architectural underpinnings are there? What sort of frameworks or code and languages are being used?
And then from there, we'll take a look together with the management team at the architectural diagrams and how they think about the overall design principles of the application. We'll sit down with them to have that longer conversation.
We talked a little bit earlier about tech debt – that's another lens we'll take a look at. In terms of sustainability and scalability, often what you'll find is there may be various components of what's been built that are outdated, no longer in support, or have been on older generations of languages that are very difficult to continue to maintain.
And then we also talk quite a bit about cloud, and if they have products on-prem versus cloud products, how they think about the two different types of deployment models. Have they made that transition to the cloud, and how are they thinking about scalability?
So those are some of the different avenues we would go down to try to assess scalability and sustainability.
Lauren Kelley: Awesome. I should check with our tech team and do a quick review of some of these topics!
I want to thank you so much, Hank. This has been a terrific discussion.
I'm Lauren Kelley, CEO and founder of OPEXEngine, the leading SaaS benchmarking platform. I am so pleased to have had this SaaS conversation with Hank Chen, partner at Bain & Company, on tech due diligence. Thanks so much.
Hank Chen: Thank you, Lauren. It's been a pleasure being here. We see huge value in the benchmarks that OPEXEngine provides and the value that you bring to our investors, so thank you so much for the work you do every day.
