This week, we welcome longtime friend and legal tech veteran Ken Crutchfield, founder of Spring Forward Consulting. Ken brings his extensive experience from major legal information vendors like Thomson Reuters, Bloomberg, and Wolters Kluwer into a timely and candid discussion about the current phase of artificial intelligence in the legal industry. Comparing today’s generative AI surge to the American Industrial Revolution, Ken describes this moment as the “Wild West” era—full of promise, hype, overinvestment, and, critically, few rules.
Drawing historical parallels to railroads, oil barons, and steel magnates, Ken illustrates how unchecked growth and technological innovation can outpace regulation until market forces or policy catch up. He notes the resurgence of large-scale infrastructure investment, now not in steel or steam, but in compute power and data centers. Just as J.P. Morgan helped stabilize chaotic markets in the 19th century, Ken suggests today’s AI frontier needs a similar recalibration, and possibly new rules of engagement.
The conversation shifts toward the practical realities of legal tech adoption. Ken emphasizes that law firms’ expectations of perfection often collide with startups’ resource limitations. Vendors need to rethink how they engage with firms by building credibility, focusing on integration, and delivering actual use-case wins. Firms, in turn, must move beyond the billable hour mindset and consider new metrics like Return on Experience. Adoption is no longer optional, it’s strategic, competitive, and increasingly client-driven.
Ken also unpacks the looming implications of content rights and data ownership in the age of AI. If firms aren’t investing in data hygiene now, they risk being left behind when more sophisticated AI tools demand clean, structured, and secure datasets. AI isn’t just about automating workflows, it’s about being ready to plug into a future where interoperability, metadata, and permissions will dictate who thrives and who gets leapfrogged.
Finally, Ken calls for scenario planning: not just reacting to what OpenAI or Anthropic might do next, but anticipating it. Firms and vendors alike should double down on what works, define success before launching new projects, and invest in meaningful adoption strategies. In a world moving this fast, it’s no longer about who gets there first, it’s about who gets there with a plan.
Listen on mobile platforms: Apple Podcasts | Spotify | YouTube
[Special Thanks to Legal Technology Hub for their sponsoring this episode.]
Blue Sky: @geeklawblog.com @marlgeb
Email: geekinreviewpodcast@gmail.com
Music: Jerry David DeCicca
Transcript
Marlene Gebauer (00:00)
Hi, I’m Marlene Gabauer from the Geek in Review podcast and I have Nikki Shaver here from Legal Technology Hub and she’s going to tell us some exciting news about fund formation, right?
Nikki Shaver (00:10)
That’s right, Marlene.
One of the key problem areas for large firms in particular over the past years has been automation in the private markets or funds practices, especially around fund formation and fund management. Many of the processes in these areas are highly manual and repetitive, such as the negotiation of side letters and the most favored nation election process, which can be real loss leaders for law firms. They’re also risk prone areas since they involve associates
doing things like copying and pasting key information manually. Many solutions have now entered the space in order to automate some of those processes with newer entrants like Smart-esque, formed by a former funds partner, and AI native law firm Covenant addressing areas that were hard to automate pre-generative AI. At LTH, we’ve just released our newest premium category, which is for fund formation.
and that provides firms with an overview of the automation and technology in this space. It also includes a full buyer’s guide. We’ve updated our listings in that area to be complete and informative, and we offer resources to support firms during their tech evaluation process in this area.
So we would encourage you to take a look. You can subscribe to LTH Premium to access all of those resources in one place. And if you’re interested in learning more, write to us at info@legaltechnologyhub.com or visit legaltechnologyhub.com and access the contact form to learn more.
Marlene Gebauer (01:45)
That is fantastic. Thank you so much, Nikki.
Nikki Shaver (01:47)
Thanks, Marlene.
Marlene Gebauer (01:55)
Welcome to The Geek and Review, the podcast focused on innovative and creative ideas in the legal industry. I’m Marlene Gabauer.
Greg Lambert (02:02)
and I’m Greg Lambert.
Marlene Gebauer (02:03)
And this week we’re joined by our longtime friend and colleague, Ken Crutchfield, founder and CEO of Spring Forward Consulting. Of course, we’ve known Ken for his work over the years as VP and of leading legal vendors like Wolters Kluwer, Bloomberg, Thomson Reuters, and others. So Ken, thank you very much for joining us on the Geek in Review today.
Greg Lambert (02:22)
Yeah, good to have you here Ken. You’ve been around the block a bit.
Ken Crutchfield (02:23)
Thank so much for having me.
Marlene Gebauer (02:26)
Hahaha
Ken Crutchfield (02:27)
I have been around the block a bit. Some in legal, some not in legal, but definitely around the block.
Greg Lambert (02:29)
Ha ha ha.
Yeah, so Ken, we’ve been talking with you lately about kind of this, you know, this wild west phase of AI and we thought we’d just bring you in and have a conversation about all of the that we’re going through and how it kind of resembles some of the, you know, social and economic phases that we’ve seen history.
And so ⁓ normally when we have a conversation with the guests, think we have much more of a structure for this. And today we’re gonna have, of break that structure and go much more conversational and see where this goes. So I really wanted to start off with, one of the things we talked about earlier this week when we were prepping for this was that we’re kind of like in early onset, as we were, the Wild Wild West phase.
both in things like regulation, but also in just kind of the similarities in the industrial revolution kind of phases that we’ve gone through in the past. So why we start off there. What are you kind of seeing as we’re getting kind of into, I mean, really the third year of the generative AI phase?
Marlene Gebauer (03:45)
What are the parallels here?
Ken Crutchfield (03:45)
Wow.
Yeah, so I’ve a bit of a history buff and like to dig in and kind of geek out in certain areas. And so I’ve been in the like 1870s for several years looking at and reading about different things. And so the American Industrial Revolution was really there a couple of things that were key to it was one was replaceable parts that allowed operation to be more scale and less artisan than what had been in Britain. So
could replace a part that broke and not have to return the whole machine to be able to get it fixed. So that was a real advancement. And the other thing was advancing at scale. And really that was driven by the railroads or exemplified by the railroads in the biggest way. And so the railroads were the first, if you build it, they will come kind of investment where investors were throwing money to build railroads, literally cities that didn’t exist.
and just believing that there was going to be a return on investment with that. And through that process, it was one of those first things where America and American capitalism was driving things without many domain and other things were being used or maybe even abused. There was also other industries, you know, Rockefeller with oil and kerosene and trying to get kerosene to light lamps and…
know, New York City homes and things like that. And then you had Carnegie with iron ore and, basically creating these massive steel plants and factories that would put rock in one side and on the other side would come out a rail that would actually be used to, you know, by the railroad industry. And these industries competed across each other too. So even though they sounded distinct, Rockefeller ended up building pipelines when he was blocked and they were going to charge too much to ship his kerosene.
And Carnegie created railroads too when he got blocked from being able to deliver his steel to port in Philadelphia. And so there were a lot of different crazy things like that going on. And it was kind of an S curve of when is this going to get done? When is there going to be enough infrastructure there that you get to a bit of a steady state? And J.P. Morgan came in as part of that a little later in the late 1890s and early 1900s.
and help drive some of the rules. basically railroads were charging whoever they wanted to, whatever they wanted to. There was collusion, there was corruption. There were a lot of different things going on there. But my point in this is really that there are a lot of things that play out where we’re in uncharted territory with AI.
but there are some patterns and some things that have happened in the past that if we’re thoughtful about them, it’ll help us inform decisions that we make, how we interpret what’s going on. And I do think that even with the current administration, a lot of the things to deregulate and everything and to re-regulate and to open certain things up and support energy consumption for giant data centers that are gonna push AI is because…
We’re operating at massive scale again, and the United States policies are driven towards making sure that the United States leads in AI.
Greg Lambert (06:47)
Yeah, I was thinking, I saw J.P. Morgan’s name somewhere recently and it was when I was ⁓ actually traveling to Ireland and I went to, in Belfast, I went to the Titanic Museum, which was really interesting because it was from the builders of the ship’s point of view, but the person that financed that was J.P. Morgan.
Ken Crutchfield (07:06)
Wow. Wow.
Greg Lambert (07:06)
So that was,
I was trying to think, where did I just see him? Well, we really do have an administration for better for worse, and maybe it’s good at this point in the phase of saying, we’re not going to really put a heavy hand in regulation, but rather we’re going to push so that, America leads in the AI.
front of it and so that the chips are developed here and sold and I guess that’s where we’re going to regulate is who we can sell those to. But with the whole design that in the international scope of things, America should be the leader and everyone should follow what we’re doing.
do you see this playing out? Do you see it as a nation, know, nation versus nation kind of thing? Or do you see it developing a different way?
Ken Crutchfield (08:02)
Yeah, think one thing that’s an observation is like the CHIPS Act and things like that that we’re looking to strategically move and source more of the computer chip supply to the United States actually happened in the prior administration. So there are some of these things that are going across administrations because I think they’re recognized as being more strategic. The one thing I would say there’s a book by a gentleman named Sean West called Unruly that I’m in the middle of reading right now.
Marlene Gebauer (08:28)
Sean has been on our podcast.
Greg Lambert (08:29)
We had Sean on the podcast a few weeks ago.
Ken Crutchfield (08:30)
Yes, yes,
okay, excellent. Well, my quick interpretation of what he’s saying is the 80s and 90s were kind of a win-win era China was growing their manufacturing, the United States companies were saving money by outsourcing manufacturing and becoming more profitable and being able to sell to a broader middle class around the world.
And with that, laws from the United States and business practices from the United States became a little more pervasive across supply chains and across the world. And now you’re starting to see some, fatigue, if you will, from those actions. And even the things that you see in the United States, ⁓ about nationalism and wanting to do things and nostalgia, I think are happening elsewhere too, is really kind of a bit of the story. So it’s not just the United States that’s running into these things.
And we’re probably in a world national power is being renegotiated for the first time since World War I, to quote something that Sean wrote.
Marlene Gebauer (09:28)
I’m up.
Greg Lambert (09:28)
think last
week we saw a Japan first party make headways in Japan for the first time. yeah, it seems to have caught on. ⁓ We’ll see.
Ken Crutchfield (09:37)
Yes. ⁓
Marlene Gebauer (09:38)
But what are the parallels here? mean, you had, when you were talking railroads where everybody was trying to be the best and the most powerful and look where that kind of ended up. so how does that bode situation now
In this case, sort of nations are sort of racing towards that.
Ken Crutchfield (10:04)
I think where I’m looking at is trying to assess, you know, what does the new normal look like? When do we get to a place where scale and rules need to be settled down? And where are you going to see the, competition? I think one other dimension of competition that isn’t really talked about a lot is, the Magnificent 7 have worth and value that is equivalent to the domestic production of like a
almost like a G30 type country. So there’s significant that have that rival what governments have. So I think that’s another part of this that plays through. And I’ll go back the Civil War with Vanderbilt, who was really central to the shipping and then shipping. And he had a tremendous amount of influence as the
richest person in the world at his time, where he was basically transporting and shipping people through a lane that he had figured out how to go through, cutting across Nicaragua to get people to California. And more people actually, rode ships out to California than came across in covered wagons, as we always heard in, American school systems. And on the way back, Vanderbilt brought gold, and that gold helped continue to keep
the Union solvent during the Civil War. So there are parallels there even, I think, to some of the things that are going on in the cross collaboration and competition between different, seemingly different industries. So I think that’s a real key part to it. What I’m looking for and what I’m hoping to do in the next month or two is begin to put together a of a hypothetical roadmap for what OpenAI might do and what some of these big
organizations that are working at scale might do. I remember Gary Marcus, who’s one of the AI experts, and he’s a little bit sour on AI right now, but an April Fool’s joke, I think a year, year and a half where he was talking about just wait authors’ compensation and rights to authors.
and know, links and sources. And there are links and sources right now that are part of, you know, ChatGPT that didn’t exist six months ago. So I’m just thinking through and trying to understand what are those different things that could play out that would inform, law firms that are trying to figure out what to do and what’s a good priority and legal tech startups that are trying to invest to build out their next functionality. So many companies, built fine tuning of
of models on GPT-35 and ended up having 4.0 leapfrog that. And therefore, it was just a little bit of a waste of time. And if that would have been known, people might not have done it. So if we can understand some of the things that might play through, I think that’s going to help. One example that I’m thinking through are the different roles that can play. So what if you had a one to many and many to one with ChatGPT?
or Perplexity or something where you have multiple people talking to and asking questions on the same thread like Slack. was something that was going on in a LinkedIn post recently. And then the flip side, you know, with what Sam Altman’s talked about with we don’t have privilege and other things, which there’s a little backstory to that. what if you had different personas for your ChatGPT If you’re asking medical questions that are personal, maybe.
HIPAA rules might apply and there might be ⁓ an MD style interface where if you’re talking to a lawyer, it’s different, or if you’re talking as a therapist or just asking general questions, you’ve got different roles and personas all in your same ChatGPT. So those are things like that that I think are healthy to be thinking through and saying those are plausible. You know, what if those happen? Would it make sense for me as a technologist to try to build that?
and recognize that wake up one day and it’s all changed because it’s embedded in the scalable systems.
Marlene Gebauer (14:01)
Ken, you raise a good point in terms of the what ifs and how things are happening quickly? I think again to your example where eventually shipping was regulated, railroads were regulated, they were slow going. The regulatory bodies could keep up with that. And here we’re in a situation like how do we figure out how to
when we’re not even sure what’s on the horizon and then the horizon sort of shows up in the next two or three months.
Ken Crutchfield (14:32)
Yeah, think that there were a lot of things that were completely unregulated. know, there was no, you know, if you wanted to blow up a mountain to, you know, to try to put rails down, you just did it. And there wasn’t really much concern as to who owned the land. Was it owned by the government? Was it owned by somebody else? If there was nobody there, it just happened. And I think that’s kind of illustrative of, some of the things. Uber didn’t get permission and go say, hey, before we,
start doing Uber this great idea, let’s go get permission from the New York medallion, know, taxi medallion and taxi association, whatever it’s called, before we did it. They just did it and then they dealt with it later when they had a little bit more power and they had the backing of consumers behind them. And I think that’s a lot of what’s going on on now. But there’s enough big players involved and there’s enough
visibility that I think you’re going to see more lawsuits and things that will probably get settled out of court around copyright and theft of data or fair use or whatever it’s going to be interpreted as.
Greg Lambert (15:32)
Yeah, decision a few weeks ago with Anthropic, where a judge basically said it is okay that you basically stole all these copyrighted materials because the output that you’re doing is significantly different than the material itself. So let me pull back onto your experience of working with
with legal information vendors for so many years. The amount of content that is currently held and just tightly possessed by essentially a handful of big legal vendors is really
the source material that enables them to produce great products and to be such big powerful players within the industry. If for some reason the AI systems were to just gobble all that up and be able to be your own Wolters Kluwer, or TR or Lexis
Bloomberg we apply the same standards to that If you were back leading one of these companies what would be your biggest concern as far IP?
Ken Crutchfield (16:47)
One thing I would do is, you know, before ChatGPT 3.5 turbo came on and everybody is aware, there were a lot of organizations that would put information, allow Google to index it so that it was discoverable and findable. I think now after this, you’re going to see contract law play out a lot more and licensing terms and conditions for access to information.
and you might see things no longer be published to be indexed or discoverable and things like that. So I think that’s one place where protecting content is going to be important. The second thing is, I think Thomson Reuters had said a few years ago, right before all this happened, they had 250 editors writing and helping improve and find parallel citations and things like that to improve Keysight. And that’s…
a fair amount of work. Now, maybe that could be done in an automated way without that much more efficiently, but it’s still, if it was 90%, you know, if the work goes away, that’s still 25 people working for several years full-time, you know, trying to do that work. So I think there are places like that where we have to recognize the potential scale of what may exist and where there’s proprietary information. But I also think you’ve got to look at it and say,
What’s the future state? What’s the eventuality? And how long is it going to take you to get there? Is it a couple of years? Is something like MidPage.ai for example, going to be good enough? Is the case law that is on LII or from the Harvard Project or other things, does that give you good enough results? Or do you really need to have the combination of that case law with the structure and the citations and the page numbers and
Access to code and regs and the specific acts and, you know, understand the context of when is the issue. Is the issue in 2012? Okay, what did the law look like in 2012? Those are all things that lawyers need to know. And I think there’s going to be varying degrees of scale of what systems are going to be able to do. And.
if I were CEO of one of those companies, I want to look at that really soberly and say, is there a good enough solution that’s gonna be out there in two years? Or do I have enough of a mode that’s gonna keep me solid and protected for a decade or longer?
Marlene Gebauer (19:04)
Ken, given your experience on the vendor side of the house and sort of what you’re seeing now, I’m really curious.
Greg Lambert (19:04)
and
Marlene Gebauer (19:12)
to get your insights on how vendors are navigating the challenge of fitting into highly customized firm ecosystems. I mean, there’s a number of hurdles. There’s a desire for interoperability. There’s the, is it going to support the billable hour or is it gonna hurt the billable hour and how does the firm feel about that?
what are you seeing or what have you seen? that, happening on the vendor side.
Ken Crutchfield (19:38)
Right. think big firms want it the way they want it. So that’s, I think, one of the challenges. they often, you know, their standard is perfection, even if good enough could be good enough. So I think that’s one balance. And I think a lot of, especially startups, underestimate that, underestimate what it takes to get access to these firms. So part of my advice would be, you know,
consider mid-market firms if you’re looking at law firms or is there a corporate legal angle that you can go after because just the standards that are being required for, you know, SOC 2 audits and conforming with policies of corporate clients are pretty significant. The other thing that I think is that there is that whole infrastructure of what a firm has in terms of its back office. And if you want a 360 view of clients,
And a lot of clients want that. And I think a lot of firms do if they are effective at cross-selling or want to cross-sell, you really need to have a little bit more homogeneity in how things work. So that demands more integration. But point solutions and best to breed is a way that you can, you probably really have to do it within legal because there’s so much fragmentation.
It’s possible that a practice area might be able to adopt something within the practice area that’s relatively siloed. And that might be just fine, except you lose that 360 view the client.
Marlene Gebauer (21:03)
I’m curious, like, is it even possible for something to go to market, you know, perfect? you know, as, as you said, I mean, again, like some of these vendors are not, big operators and they’re coming up with sort of small solutions and, really looking to partners to kind of help them flesh it out because they don’t have the capital or, know, they, they don’t have the, you know, the expertise.
Greg Lambert (21:08)
You
Marlene Gebauer (21:29)
necessarily to, to put something together is perfect. And I mean, let’s face it, you know, every firm you go to is going to be slightly different. like anytime, anytime there’s something out of the box, you know, there always needs to be some tweaking. So, are these expectations even reasonable? And. if not, what do you know, what do firms need to do about that?
Ken Crutchfield (21:37)
Right.
Yeah.
So one of my observations is I’m just trying to understand how many things are truly game changing. I think there’s a lot of incremental solutions out there. There are things that can be automated. Getting rid of the 20 to 30 % of non-billable time that can be squeezed out without affecting the billable hour, that’s great. Being able to automate part of a process, maybe in an SEC filing or other things, awesome.
but I’m really not even hearing, even like with Harvey and others, what are the specific things that are repeatable that everybody’s saying, I gotta have that? And my frame of reference for this is, know, dating myself, I started in 1982 with Lexis. And at that time, Lexis salespeople could still knock on a law firm’s door and never heard of Lexis. And it was back.
when you could literally walk out with a signed contract because it was so game changing what they were doing. What do mean? don’t have, I can search this, show me this, know, Boolean search. Wow, if I would have known this, I would have won that case. Those sorts of things are what I think, people have kind of hyped about what AI can do, but we’re not quite, I’m not seeing a lot of those that are, you know, that killer app yet. I think there’s a lot of incrementalism and a lot of things that show promise long-term to do that.
it may be a little bit larger investment and a longer runway. I think to the point that Greg and I have been talking about about crystal ball questions from a few years ago, is there going to be some winners and losers?
Greg Lambert (23:21)
Yeah, since Ken brought that up, before you jumped on Marlene, we were reviewing his crystal ball question ⁓ from a legal tech from 2022, so pre-Chat GPT. And Ken was very on point with saying that there’s some AI technology that’s coming around that’s really going to help. I think at that time it was like,
un-silo a lot of information that is going to be able to let us kind of get the free flow of information. Ken, what stock do I need to pick for next year?
Ken Crutchfield (23:53)
I don’t know.
I think my answer was sufficiently vague and I said AI so it sounds like I was smart.
Greg Lambert (24:00)
Yeah, yeah, yeah.
I was talking with somebody actually earlier today and we were talking about just the amount of AI products that people are knocking on law firm doors today trying to sell. And I think when you said these are not huge game changers, these are kind of point solutions that fix.
incremental things. And it’s interesting because the person had framed it as this was that I’ve got these vendors that are selling me these software packages that are actually just one little slice of a software package. And I need that whole package, but I can’t buy like 23 point solutions in order to get most of that package. how do you see things changing or do you
Is it through consolidation or do you see there’s going to be a new Google type entity that comes in and figures it out for law firms?
Ken Crutchfield (24:57)
Right. So an observation since, you know, we talked to AALL, Greg was I looked at all of the different, you know, booths there and Lexis had AI fairly prominently highlighted and one or two others had AI mentioned on their banners. But if you went through and looked by and large, AI was not the headline. Thomson Reuters had co-counsel.
Wolters Kluwer had vital law and their vital law AI, but it’s kind of fading into being like an expected component of what’s there. But I didn’t really see anybody jumping out and saying, here’s something dramatically new. So I do think that there’s going to have to be a little bit more stitching together, integrating, pulling together things to do more complete workflows. And some of the really large firms that are doing things that are custom with Harvey, that’s my outside in perception.
that’s not repeatable yet. So I think that’s really one of the things that I’m looking to see is how far do you have to go with that. And there are some really good products out there if you can get adoption, but you know, adoption isn’t always easy.
Greg Lambert (26:02)
What do you think about a company like Microsoft coming in and buying a Thomson Reuters?
Ken Crutchfield (26:11)
Yeah, so I think Microsoft’s market cap is 50 times what Thomson’s is. So more or less the last time I checked. So I think there’s interesting things like that when I talk about the railroad analogy and doing things at scale that if data really is the new oil to quote Damien Riehl or others that have used that as an analogy, it’s something that is going to be either
accessible and discoverable and found or they’re going to acquire it. They’re not going to let somebody get in the way, if you will, of the railroad coming down and getting from point A to point B. So I think that things like that are the sort of things that I would encourage people to not get overwhelmed with, but, you know, have on their scenario map as to what ifs, you know, because I don’t think that’s too, too unrealistic. I have no knowledge, but it just seems like intuitively that
there would be somebody thinking and asking those questions of are there data sources in particular industries or areas that have, pervasive, important, deep vertical like medical or other things, you know, and does it make sense to go after that data?
Greg Lambert (27:20)
me follow up on that with something else that I heard recently, and that was when it comes to the information, and the person I was listening to was talking about the press, about news, and that they foresee that we may end up with this almost Medici type of environment where we have just a few families.
that are running the whole of the industries. Now, families could be nation states. it could be, and I think the way that he had phrased it was, there would be a conservative view of it, a liberal view, an American view, a European view, a Chinese view. we would have this kind of information that…
Marlene Gebauer (27:46)
It’s the family, Greg. It’s the family.
Ken Crutchfield (27:49)
Yeah.
Greg Lambert (28:05)
is being
set up and presented in a certain way based on the we want. So do you see potentially something like this happen? mean, kind of, we always talk about the big two in legal.
But do you see that even kind of getting even, maybe legal is not even the main focus anymore, that legal is just a piece of these pies and that the information starts getting categorized into certain groups.
Ken Crutchfield (28:33)
Yeah, I can say like my wife and I have entirely different Facebook feeds, you know, it’s like we’re on different planets. They’re so different. And so I think at the big social mass market, you may end up having some of those tendencies that just kind of play, especially as they’re learning and having a feedback loop. But in the legal space, I feel like there’s so much investment, so much fragmentation, so many different solutions and different things to attack that I kind of feel like
That’s a long ways away. know, going back again to another kind of industrial revolution area, when Rockefeller had 90 % of the oil market locked up, there was still 10 % that was doing very unique and different things like creating machine oil to keep machinery working and things like that. And he did not have a play in Texas like Texaco or Chevron out in California.
an argument to say that the point where his company was broken up and his monopoly was broken up was right at the time that it was going to be broken up naturally by competitive forces. So I kind of bring that back and say, think, the Thomsons, the Bloombergs, the Lexis’ the Wolters Kluwers the Literas the iManages of the world, I think they’ve got really strong positions, and I think they’re going to continue to have strong positions.
but there’s so much more that can be automated, changed, and tools can be applied that I think you’re gonna see a lot more companies before you see a lot less.
Marlene Gebauer (30:00)
expansion than contraction.
I wanted to talk a little bit about data hygiene and how that impacts AI implementations in legal. I think that some of the tools we have have fantastic capabilities, but it’s oftentimes these products don’t have access to
⁓ the data that they need because, either it’s unstructured or because they are not allowed access. what, anything, are you aware of that, that firms are doing to, clean up their data to make it accessible and actually working.
with their clients and their security teams and their risk departments to actually allow it to be accessible.
Ken Crutchfield (30:49)
all right. let’s helicopter way down from the big picture that we’ve been talking about. And I think one thing that any organization can use as a direct piece of advice is focus on your data. If there’s one thing that’s going to be work that’s valuable that will survive a change in language models, Google leapfrogging open AI or AWS coming in and Amazon coming in with something,
this disruptive, it’s going to be working on your data. Get your data right. Get the metadata straight. Figure out your permissions and who can have access and get access to that. That’s something that I’m hearing not just in the legal space, but more broadly. have ⁓ one of my contacts that I reach out to periodically, ran the Google’s self-driving car AI initiative for a few years. And that’s been his advice for the last few years is get your data right.
Make sure that you’ve got the metadata, that you’re thinking about hygiene. If your knowledge management systems are fragmented because you still have some rogue attorneys that like to keep things on a flash drive or coddled away somewhere, those things have got to break down. Those have got to break down in order for the firm to be able to do things and to be able to make the right decisions and advise clients properly.
Greg Lambert (32:05)
Well, and actually, one of the things that I think that we’re facing is ⁓ things like ⁓ with Microsoft, you can put things into their shared drive, right? And actually, you need to put it into the shared drive in order to use Copilot.
Ken Crutchfield (32:22)
Right.
Greg Lambert (32:23)
So I think there’s still some processes that we need to work out on where the data is in order to take advantage of the tools that we’re bringing in. ⁓
Ken Crutchfield (32:34)
Right.
I think that’s a great one too, like with copilot, what I’ve heard from some firms is that, asking copilot to look through email, well, one of the challenges is that there’s a lot of speculation in email, not all email is official documents. So it’s a great example of, okay, what is the data and how do you categorize data so that if you’re asking a question that requires a substantive answer that’s got to be correct, that it doesn’t get, polluted with…
passing information or jokes or hypotheticals aren’t true. I think those are perfect examples.
Greg Lambert (33:10)
I’d hate to see what it does with our Thursday happy hour emails that go out to everybody. ⁓ well, let me ask you about the vendors that most law firms deal with I think we spent the last year and a half, everyone added, AI to whatever the end of their product name was. And that was great. Everyone, I think it was a good experimental phase. And I think a lot of,
Ken Crutchfield (33:14)
Ha ha.
Marlene Gebauer (33:15)
You
Greg Lambert (33:34)
of us were ready to experiment and having fun doing that. But at some point, we’ve actually got to show some type of return on investment. I don’t know of any firm that could honestly come out and say the amount of time, energy, and people and money that they’ve put into ⁓
products has given them a huge return on investment. And I think we’re still okay with that for now, but that’s not going to go on forever. So if you’re an AI vendor dealing with this industry now, how do you build credibility with this industry?
Ken Crutchfield (34:00)
Right.
You know,
I’ll answer it more from the firm side first. So, like, one of the things I see is, the billable hours is certainly always going to be a bit of a challenge for those firms that have that mindset. Their infrastructure is structured to perpetuate that. You know, they don’t have metrics on improving margin and fixed fee engagements and things like that typically. So, there’s that challenge. And then,
You know, one thing that really stuck with me was I had a firm tell me that they wanted a deal because of machine learning. Their client said, I want you to do this engagement using machine learning. And so when the other vendors or when the other firms were asked, they did not have a machine learning capability. So guess who got the business? And the point was is that I find that ROI is always going to be a bit of a challenge.
with the billable hour in firms. So I view it, more as being competitive. You’ve got to be within that window of being close enough. Some firms are going to be technology leading and want to be that way. And they’re going to be out in front and do some of these things because that’s part of their reputation. And then there’s going to be the other ones that are going to have to be fast followers because when clients shift from saying, don’t use AI or don’t use machine learning or don’t do this to do this,
I expect you to do this. now it’s professional misconduct ⁓ or it’s outside of our engagement policy. If you don’t use AI or if you don’t do these certain things, I think that switch flips like that. And I think that’s going to be something that’s going to be very important for firms to gauge on. On the vendor side, I do think it is going to be a kind of a, you know, let’s.
Marlene Gebauer (35:33)
you
Ken Crutchfield (35:54)
Let’s kind of pause and catch up here a little bit because there is so much going on with, ChatGPT agents and other things. You’ve had a lot of people hyping agents, but you know, now ChatGPT can log into your Thomson account or your Lexis account and do certain things on your behalf. So I think there’s, there’s a little bit of, step back and retrospect and also just focus. Don’t, don’t try to cover the waterfront and do everything. Focus and
solve a specific problem, build out from that, build out again, build out again. And if you have to go deeper and then deeper and deeper, that was one thing I learned when I was at Bloomberg. Some of the people in New York in the financial side were like, if you go after something, go after it. And if it doesn’t work, go deeper. And that was one of the things I really liked about Bloomberg is that they were willing to, not give up. would they would play the long game. They wouldn’t just stop after one release like a lot of publicly traded companies will.
they’d go deeper and they’d dig into a second, a third, and a fourth release.
Marlene Gebauer (36:54)
Ken, I read something the other day and instead of ROI, was ROE. So it was return on experience. And I’m wondering if firms or organizations are actually sort of measuring or really care about this. But like you’re talking about, the workflow is being more seamless and possibly this is non-billable work that sort of it’s just making stuff easier. Maybe you’d administrative stuff making it easier.
looking at this and making determinations sort of based on those type of metrics.
Ken Crutchfield (37:22)
It’s a good question. I kind of rephrase it and say it’s the new cost of doing business.
trying to, my mind’s going to automotive analogies in the early days of automobiles and automobiles took a lot of, you wanted to drive a car, okay, cool, it’s really beneficial, but you better be strong enough to do the self-start crank. I know my great-grandfather broke his arm trying to start a 1916 Mitchell, which was a GM-acquired company that doesn’t exist anymore. So there are different things that you.
have to put up with to learn if you know that this is the promise of the future and the future state. Now we don’t think about that. You click your remote and your car starts outside and it’s warm by the time you get into it. And it’s got all these other backup cameras and other things going on that make a car so much more automated than what it was. I think going back to the Wild West in the early days, we’re still very much in those early days. So I think it’s a recognition that if you know,
If you want to live in the suburbs, you’re going to have to have a car and you’re going to have to learn and invest to do that. if you want to be, practicing law in the 21st century as the 21st century progresses, you’re going to have to be proficient in these tools, put up with some things that are, know, not pleasant, but know that every day that goes by, every year that goes by, things are going to get better and it’s going to get better fast.
Greg Lambert (38:42)
Ken before we get to our since you did such a great job as the crystal ball question the when we had you on at legal week we get to that those there any other topic that we Failed to hit on that you think we might be missing
Ken Crutchfield (38:47)
you
Yeah, I think one thing that, you know, both for legal tech startups and also for law firms in particular is adoption, not underestimating that, you know, for a legal tech provider, the sale isn’t over until you have users actually using the product. Otherwise, you have a contract that’s likely to get canceled. So if you want to be in a position where you’re going to be successful and have a truly successful sale, you have to help the firm.
get through the process of changing their process, ⁓ understanding what the skill set changes are in the training and make sure that they’re getting the outcomes that are needed. And I think that’s something that firms need to recognize too, because in a world where, anything that’s not billable is, you know, a drain on partner profits, there needs to be a little bit of acceptance of that in the planning upfront to be able to do that. The other part is just making sure that, especially with the gauntlet of
IT and committees and other things that need to be done to evaluate a product and make a decision, that that, you you really try to find a sponsor for that work and recognize that sponsors within a firm or even a corporation, are dealing with things that are both strategic, kind of tactical, like I work through these certain things to be able to get the product out there. expending political capital and putting their own personal
you know, reputation on the line. And there’s a personal aspect of these projects. Is it rewarding? Is it going to get me the next job that I want? Do I get to see my kids earlier if we get this implemented and go to a couple of baseball games, little league games? Those are the sorts of things that I think, are important to be aware of and to think of both from the firm side and then also from the vendor side.
Marlene Gebauer (40:42)
I was going to follow up with that before the crystal ball. What sort of skill sets do you think are required on both sides of that equation?
Ken Crutchfield (40:51)
That’s a great question.
Marlene Gebauer (40:53)
And if anybody knows it, like they, like they have bottled the secret sauce.
Ken Crutchfield (40:57)
Yeah.
I think, you know, even with a out-of-the-box SaaS solution that’s supposed to be low implementation, I think there needs to be strong onboarding customer success sorts of approaches. I’ll go back to my Lexis days again, because it’s a, I’ve said this one other time.
I kind of deconstructed what Lexis had done back in the early, early days. And they manufactured terminals, you know, before there were computers. And they ordered the phone lines for the firm in the firm’s office so that those terminals could access the private data network that Lexis had created. And then that accessed, the world that was the mainframe computers and the actual service.
Well, eventually personal computers became a thing. IT departments and firms became a thing. You didn’t need the phone lines anymore. The internet replaced the data network. There were a number of things that did play through, but that was a full service, complete thing to be able to adopt. And you even had Lexis reps back in the day that had an office in the firm that sat there eight hours a day and were on the beck and call to be able to help. And so…
those are the sorts of things that I think if we’re really looking at how do we get somewhere, we should be thinking about models, you know, maybe not that extreme, but what we need to be thinking about on both sides, what does it take to adopt? And there’s probably a cost and a charge for some of that work to truly get to adoption with some of these solutions.
Marlene Gebauer (42:23)
that’s a great answer because you know I think that’s sort of one of the things that that I think people just sort of assumed for a long time that okay we got the contract and you know and it’s in and now people are just going to use it and we have found that that that is not always the case despite everybody’s best efforts and testing and whatnot so very good.
I will ask you the crystal ball question. And since we’ve been saying this stuff is happening so fast, certainly don’t go out the next few years. Maybe we should just do the next few
Greg Lambert (42:53)
happened on my end.
Marlene Gebauer (42:53)
So, you know, what
should firms, vendors, the industry in general, what should we be thinking about?
Ken Crutchfield (43:02)
I think we should be thinking about focusing a little bit more on fewer tasks, fewer pilots, and trying to get a few more of them, really kind of progress them further. I think also just stepping back and looking at things like work on your data, things that will be universally beneficial to whatever solutions you select and whatever solutions went out in the marketplace. And just be eyes wide open about
the potential for the next thing that OpenAI does or Perplexity or Google Gemini or Anthropic Claude or any of these does because it might leapfrog and fundamentally change what you’ve done. So just think through those things. I’m actually gonna try to write an article in the next month or so highlighting a few things that I think might be realistic in the next 18 months to two years.
for these large language model companies to actually do and deliver on that might help people with a more concrete guide and checklist of things to consider. Yeah, that makes sense. All right, well, what does that mean to the projects that we have today? What would that mean to our investment and things that we would consider? That’s really where I would go.
Marlene Gebauer (44:11)
we got to do some scenario spinning on a regular basis.
Ken Crutchfield (44:13)
Yeah, a little bit of scenario planning. Keep it simple. Yeah.
Marlene Gebauer (44:18)
Ken, you’re talking about adoption and, and I think part of that question is like, what’s really important and you know, in your experience, like, how do you determine what’s important? Like, what do you focus on in terms of, a solution and, for who and doing what, because, know, there’s,
Either nobody’s interested or maybe a million people are interested. you know, as a professional in this area, how do you make that determination and move with it?
Ken Crutchfield (44:46)
I think this is an art that’s so straightforward but hard at the same time is is think about those things before you decide to pursue a project or to buy what would success look like? In a firm make sure you’ve got some leadership skin in the game on it. I think that’s that’s really important. Otherwise, it’s like pushing a rope You know, you’re pushing and trying to sell something instead of having
the organization want to buy it and pull it in. And I think that’s really an important part, as you look at what are those metrics that really matter. And if those metrics don’t matter, then maybe it’s not time to do the work or to do the project aside from that argument of, well, what happens when all of our competitors have this and our clients start asking for it, then what? And if you can kind of reframe things in that.
in that regard and what would be the areas and situations and listening signals we want to know to be able to be ready for that, get the buy-in around that to say, okay, we’re starting to hear, we had our first client actually ask about this or, these two benchmark firms that we like to compare ourselves have this capability, we know that competitively. Now it’s time that we have to have that in our portfolio and there’s no question about it, the metric is competitive parity.
That’s the way I would look at it.
Marlene Gebauer (46:05)
think that’s a really good point, competitive parity, because so often, even at top levels, what is important can change. ⁓ And so I think having that competitive intelligence being able to raise that in terms of getting the buy-in is critical. Thank you.
Ken Crutchfield (46:13)
Yes.
Absolutely.
Marlene Gebauer (46:23)
Well, Ken Crutchfield, thank you so much for talking with us here on The Geek in Review. And of course, thanks to all of you, our listeners, for taking the time to listen to The Geek in Review podcast. If you enjoy the show, please share it with colleague. We’d love to hear from you, so reach out to us on LinkedIn and Blue Sky.
Greg Lambert (46:41)
And Ken, for listeners who want to learn more about Spring Forward Consulting, or connect with you, them to go?
Ken Crutchfield (46:42)
Awesome.
Awesome. Well, first off, thank you Marlene. Thank you, Greg, for having me. This has been a great conversation. Loved it. Springforwardconsulting.com is my website. You can look me up on LinkedIn, connect with me there. I write articles for Above the Law and then also for Law Next and Bob Ambrogi. So those are other places that you can find my writing and some of my thought leadership. And happy to talk with people. So reach out, send me an email, and I’d love to talk. Thank you.
Greg Lambert (47:18)
Did you see Marlene that we dropped to Bob’s third favorite podcast? Because they’re now counting his Friday gathering as a podcast, so we’ve dropped to number three. We’re still up there. We’re still up there.
Marlene Gebauer (47:22)
No, I’m very sad.
Ken Crutchfield (47:25)
You
Marlene Gebauer (47:28)
We need to have a talk. We need to have a talk.
Ken Crutchfield (47:32)
Nice.
Marlene Gebauer (47:33)
And as always, the music you hear is from Jerry David DeCicca Thank you so much, Jerry.
Greg Lambert (47:38)
Thanks, Jerry. All right, bye everyone.
Marlene Gebauer (47:40)
Bye.