Latest / Technically Sound Strategy / Episode 37 - AI Readiness for Law Firms with John Johnston

Transcript

Episode 37 - AI Readiness for Law Firms with John Johnston · Technically Sound Strategy

• 0:00
  Welcome to Legal Marketing Radio, a production of Le Fleur Marketing.
• 0:04
  This week, I'm speaking with John Johnston, a co founder at PeopleIT.
• 0:08
  John and I spent some time talking about the ever changing topic of data
• 0:12
  architecture and what law firms should do to prepare their practices for an AI
• 0:17
  enabled future state where artificial intelligence is used increasingly
• 0:21
  across small and large firms alike.
• 0:24
  Born and raised in Northern Ireland, John followed a
• 0:26
  brilliant woman back to the USA.
• 0:28
  Two decades later, he's still with his sweetheart, raising their energetic
• 0:31
  six year old and building people it with his co founder, Charlie Lambers.
• 0:35
  People it is an it consultancy specializing in data intelligence
• 0:40
  and systems integration.
• 0:42
  When not focused on people it, you can find John helping with other startups and
• 0:46
  supporting his local community serving on the boards of Goodwill Industries
• 0:49
  and Holland Educational Foundation.
• 0:53
  Today we've got John Johnston with us with people it, um, had some
• 0:57
  conversations in the past around, especially around AI as AI has taken hold.
• 1:02
  And we have a lot of attorneys that we work with that have dabbled.
• 1:06
  I would say, and John and I got talking about some of the things that his company
• 1:11
  is doing to lead up to implementation of AI across the organization, and it
• 1:17
  got me thinking that this would be a great topic to bring to this podcast.
• 1:21
  John, welcome.
• 1:22
  It's great to have you on here.
• 1:24
  If I got anything wrong in that first lead up, please correct me.
• 1:27
  No, you got it, Chip.
• 1:29
  You nailed it.
• 1:30
  Yeah.
• 1:30
  Excited to be here and have a chat with you.
• 1:33
  Awesome.
• 1:33
  Like I said, we have a lot of attorneys that we work with that have dabbled.
• 1:37
  Of course, we've been pretty active in the AI space.
• 1:39
  We're using the tools that we have available to us.
• 1:42
  Open AI, chat, GPT, Claude All of the different image generation things.
• 1:47
  We've used a bit of everything and we've done a little bit of
• 1:49
  development in this space as well.
• 1:51
  You were a managing partner with longest tail AI, which is a technology
• 1:55
  that allows us to just soup to nuts, build a website, publish it,
• 1:58
  and continue to publish content.
• 2:00
  About 50 to 100 pieces a month.
• 2:03
  And we've trained it on legal content.
• 2:05
  And we've updated it with, bar rules, state by state, things like that.
• 2:08
  So we've played in the space, but we are not a technology company.
• 2:12
  And as I've talked to the lawyers that we work with, one of the things we run
• 2:15
  into, of course, is future planning.
• 2:18
  And so let me start with that as a question.
• 2:21
  What is a good starting point to read the tea leaves?
• 2:24
  What is coming?
• 2:25
  And what should organizations be doing now to prep for not just
• 2:29
  generative AI, but AI as a whole?
• 2:34
  Yeah, for sure.
• 2:35
  1st off, I'll say just so your listeners and your lawyer or
• 2:38
  friends don't get misconception.
• 2:41
  I'm the least technical person on our team.
• 2:43
  So we won't be diving too deep into the actual technicality of things.
• 2:46
  We really come more from the business value side of things.
• 2:49
  So 1 of my roles is okay, can we stay focused on the business value?
• 2:52
  And I think a lot of times, where I feel I add value to the team
• 2:56
  is in keeping our eyes focusing.
• 2:58
  Hey, let's not layer AI tool top of AI tool.
• 3:01
  If we don't have a clear strategy and line of sight to what we're trying to achieve.
• 3:05
  And I think you've probably experienced this as well.
• 3:07
  Chip and maybe some of your listeners where there's a little bit of a.
• 3:12
  It's actually so much now, but 18, 24 months ago, it was
• 3:15
  this very big fear factor.
• 3:16
  If I'm not on this wagon, I'm in big trouble and how do I get on
• 3:21
  it and and keep going with it.
• 3:22
  And I think what we found is people have taken a deep breath and said, okay
• 3:25
  my business is not going to crumble overnight here if I'm not adopting AI.
• 3:29
  And what we've seen really for companies and talking about that
• 3:31
  readiness, we really You know, deal in the space of AI readiness.
• 3:34
  We talk a lot about that and we even take a step back before
• 3:39
  even we get into the, tool stack.
• 3:41
  And we actually are seeing more and more.
• 3:42
  1 of the keys to benefiting from AI is really rigorous processing companies.
• 3:48
  So if companies have really rigorous repeatable.
• 3:53
  Accountable processes that their teams are using and that they're
• 3:57
  kept accountable to using, then it's much easier to come and plug AI in.
• 4:01
  If those standards and processes are not in place, then, AI can
• 4:05
  create more chaos than it can value.
• 4:08
  1 of the big things is really looking at how do we make sure
• 4:11
  there's really good processes within a company that we're coming to
• 4:14
  try and do AI implementation with.
• 4:16
  And from there, then we can start to look at, okay.
• 4:19
  What data have you got?
• 4:20
  Where are your data sources?
• 4:22
  Let's start to gather your data, store your data well, architect
• 4:26
  it well, and make it usable.
• 4:28
  But often we're coming back to something as simple as, do you have good process
• 4:32
  standards that are repeatable and accountable within your organization?
• 4:35
  If you've got that, then great, we can start to dig in.
• 4:38
  If you don't have that, we're encouraging you.
• 4:39
  Let's start there.
• 4:41
  That makes a lot of sense, in this space, just to set the stage a little
• 4:44
  bit most legal practices that we're talking in, we work with family law,
• 4:49
  estate planning, but a lot of personal injury, a lot of trial lawyers, right?
• 4:53
  And when we're talking the technology stack that they're using, oftentimes
• 4:57
  case management is at the heart of it.
• 4:59
  Document storage may be a part.
• 5:02
  So they may be on Google or more likely on Microsoft on 365.
• 5:07
  They're using SharePoint more often than not.
• 5:09
  But they're dealing with confidential information within their HIPAA
• 5:13
  protected health care data medical records, things like that.
• 5:16
  So what should they be taking into consideration is there as they're dealing
• 5:21
  with that sort of technology stack,
• 5:23
  right?
• 5:24
  Yeah.
• 5:24
  Yeah.
• 5:25
  And the law environment security is probably, I don't know, at the top of the
• 5:30
  list that they are looking at and dealing with that's kind of their own brand.
• 5:34
  And, their brand is confidentiality for their clients and how they build that.
• 5:38
  And I think the big challenge for companies they come across is
• 5:41
  understanding what data are they storing?
• 5:43
  Yeah.
• 5:43
  And where are they storing it and then how are they going to use that data?
• 5:47
  And if you're a smaller law firm, there's less people.
• 5:49
  There's more accountability.
• 5:51
  You can see that more easily.
• 5:53
  But as you get bigger and add more team members, then how documents are
• 5:56
  managed become a little bit more fluid.
• 5:58
  So again, it comes back to that idea of are there standards and processes that
• 6:01
  are in place that people are adhering to and following and that that is a
• 6:04
  written process that it is audited, that it is seen in terms of the tool stack.
• 6:09
  The thing we come back to is there's 101 tools you can start to use and throw
• 6:13
  on here to actually look at versioning history, look at where documents are
• 6:16
  stored, moved to logged in and out of there's so much there that we can.
• 6:21
  We can do and look at, but again, it comes back to this idea of what are the
• 6:23
  standards and how are you auditing that?
• 6:25
  One of the things we've come across is there's a couple of
• 6:27
  kind of great, law firms out there that are solely specializing
• 6:31
  in the world of cyber security.
• 6:33
  And we've come across a few of them in our travels.
• 6:35
  Maybe we can post.
• 6:36
  Those along with your podcast summer, people can access those, but we find
• 6:39
  those firms to be really good and look into specifics of different
• 6:41
  law firms, what industries they're in and what they're doing in terms
• 6:44
  of actually guidance practices.
• 6:46
  And it's the idea of doctor heal thyself.
• 6:48
  Sometimes what we do for others is not as good doing for ourselves.
• 6:50
  And we need somebody else to come in and help with that.
• 6:52
  But again, the biggest challenges we see, and breaches that we see
• 6:59
  for folks with data, they don't come always through the technology.
• 7:03
  They come through human error, right?
• 7:04
  They come through a human mistake.
• 7:06
  And it is about really that idea of how are you training your team?
• 7:09
  How are you helping your team stay accountable to your
• 7:12
  expectations and standards?
• 7:14
  Yeah, that makes a lot of sense.
• 7:15
  I was having a conversation with a lawyer just this week, and we were
• 7:19
  talking about just generative AI and they're using that to they're leveraging
• 7:23
  that to help them ideate and then come up with article outlines, right?
• 7:28
  And so oftentimes our team is creating the content or they're taking
• 7:31
  those outlines or those drafts and we're optimizing them for search
• 7:35
  and doing some other things there.
• 7:36
  But in our conversation, what we found out was that they were
• 7:40
  using a pro version of ChatGPT.
• 7:42
  And this was actually a surprise to me.
• 7:45
  I thought I was, I thought I was pretty literate on this, but I
• 7:48
  didn't realize that even Pro ChatGPT is using your feedback and your
• 7:52
  data to train their model, right?
• 7:55
  I had been operating under the assumption, not even the assumption, I do read
• 7:59
  the terms on most platforms, but.
• 8:01
  I've been operating on the assumption that if you're paying for the GPT, if
• 8:06
  you're paying for the AI, they probably don't use your, data to train the model.
• 8:09
  But with Chad GPT, they do unless you specifically go in and change that.
• 8:14
  Now, that hasn't been an issue for us because we use the team's version.
• 8:18
  And then I'm pushing the team here to start using open AI from within Azure.
• 8:22
  Because I think that's, where there's guarantees for Microsoft
• 8:25
  that data will not be used to train.
• 8:27
  And I trust that a little bit more.
• 8:28
  But I guess outline a couple of ways that a firm could potentially
• 8:32
  get in trouble, using some of the tools that are out there.
• 8:35
  mentioned human error.
• 8:37
  What are some of the ways that can happen?
• 8:39
  And then what are some of the ways that a firm can protect against it?
• 8:42
  Yeah, I'm going to be like a broken record here in terms of
• 8:44
  talking about, your people, right?
• 8:46
  And in terms of how you train your people and how they're using the tools.
• 8:50
  I think to your point, it is the generative AI.
• 8:52
  It's all of the copyright space, right?
• 8:53
  Which again, a lot of your lawyers are going to understand and know.
• 8:56
  And there's still so many debates over that.
• 8:58
  How are people's data protected as these language models just build
• 9:02
  and build and all they want to do is they just need more food right there.
• 9:06
  actively looking for food.
• 9:07
  And that's why, things like even ChachiPT Pro and, the next levels,
• 9:11
  I think it's you're hoping that they're keeping you confidential.
• 9:14
  And if you're paying for a service, well, then the data is yours.
• 9:16
  It's not been exposed, but they are, they're certainly always trying to find
• 9:19
  ways that's not the truth and the case.
• 9:21
  I think, one of the things I found in this reading of my brother is
• 9:24
  Attorney over in Europe and Europe are a little bit ahead of the game on this
• 9:29
  and maybe ahead of the game wouldn't be even language people would use.
• 9:32
  There may be a little bit more stringent or cut a little bit more rigor around
• 9:38
  how they're managing it, but they're putting a lot of confines on how folks
• 9:41
  are using data, how they're sharing data, how they're asking for permissions.
• 9:45
  And it's certainly I think one of the challenges that's slowing down.
• 9:48
  Adoption and it's slowing down what people would say is creativity
• 9:51
  and that's one of the challenges I think we have with some of the tools
• 9:54
  and I suppose law firms as well.
• 9:56
  It's how much do I lock this down?
• 9:58
  How much do I allow creativity and opportunity to emerge as opposed
• 10:02
  to really be very legalistic.
• 10:03
  And that's a conversation we often have in our company as well.
• 10:06
  Are you asking for forgiveness?
• 10:08
  You know, a little bit this idea I'm trying to, take some liberties with
• 10:11
  things or are you really trying to understand all of the aspects of it first?
• 10:14
  And obviously we're not, dealing in the legal space quite as much.
• 10:17
  So we probably are a little bit more on that freewheeling side of it.
• 10:20
  But I do think there is, it's almost a full time job to try and work out
• 10:25
  what, what your capabilities are or what your approach can be and what's
• 10:29
  going to keep you within the bounds.
• 10:30
  But I would also say folks trying to catch up with this and yeah what's the
• 10:35
  word I'm looking for in that they're trying to litigate against some of
• 10:38
  this, even with the status we've got, it's so overwhelming at the moment
• 10:41
  that there's, it's hard to litigate against in any in any generalized way.
• 10:46
  Yeah, no, that makes a lot of sense.
• 10:48
  I think a lot of our clients, a lot of people in this space are
• 10:51
  trying to push the envelope, right?
• 10:53
  Looking ahead to see, well, there's there is a competitive
• 10:56
  advantage to rolling that in.
• 10:58
  I want to talk a little bit about a specific use case.
• 11:01
  Like I mentioned, they're using case management software, right?
• 11:04
  Case management software basically is a CRM.
• 11:07
  But then it's a combination CRM project management, so you're
• 11:11
  moving a case through the system.
• 11:13
  It's all built on top of a database, right?
• 11:17
  And so can you give me some ideas?
• 11:19
  This is a very open ended question, but can you give me some ideas of how?
• 11:23
  A database like that could be leveraged and, what sort of outputs could a law
• 11:29
  firm hope to use with something like that?
• 11:33
  So you're saying the case management side of it, it's acting like a CRM though.
• 11:36
  It's holding all of the client's information and documentation.
• 11:39
  Is that how I'm understanding it?
• 11:41
  Yeah.
• 11:41
  So when a new potential client comes in, they come in through intake.
• 11:44
  The intake person is going to enter them into their case management system.
• 11:48
  And if it becomes a case if it becomes a client, they sign an
• 11:51
  agreement with the firm and get hired.
• 11:53
  And then that usually sets a trigger that initiates the
• 11:57
  next stages of a case, right?
• 11:59
  Medical records retrieval police record, interviews, insurance coverage, all
• 12:04
  that gets entered in there and then they use that to Formulated demand letter
• 12:09
  the demand letter goes to the insurance company that says hey, you should settle
• 12:12
  this case for this and here's why and then it follows that case all the way
• 12:16
  through pre litigation through litigation if it goes to it to trial, et cetera.
• 12:20
  All that data is contained in there.
• 12:22
  It does include sensitive medical records potentially.
• 12:25
  So just using your imagination.
• 12:27
  I know it's a out of the blue question, but what are some of the things that,
• 12:30
  that a firm could look at using something like that to actually do using AI
• 12:36
  ? Right, so yeah, I'll deal in the realm of the possible and not
• 12:40
  necessarily what's allowed for that.
• 12:42
  But yeah I think we love those kind of data sets because I think what
• 12:45
  we start to do is we start to say, okay, we want to find similarities in
• 12:49
  those cases that are moving through.
• 12:51
  So it might be things like, hey,
• 12:53
  can we actually have an aggregate of the specific insurance companies?
• 12:56
  And is there a volume of certain insurance companies that help
• 12:59
  our cases move through quicker?
• 13:01
  Or they pay out faster or whatever that metric might be.
• 13:04
  And now we know, okay that's a great company for us to work with because
• 13:08
  it's a more efficient process.
• 13:09
  We actually, be able to monetize that better in the back end and
• 13:11
  our client gets a better output.
• 13:13
  There's also other things to do in terms of.
• 13:16
  Internal looking at, hey, what was our billing for these light cases which
• 13:20
  of our team members we're actually managing them and where they see, hey,
• 13:24
  cases of a similar nature took longer for certain attorneys in the firm, it
• 13:28
  can be saying, okay, we need to look at a better training and development
• 13:31
  process, or we need to look at how the process was actually being followed for
• 13:34
  that person to gain efficiency on that.
• 13:36
  But again the data that you're talking about is.
• 13:38
  Thanks.
• 13:39
  Okay.
• 13:39
  It gets exciting to us when we start to internally start to cross reference that,
• 13:42
  and you look at the demographics, you look at where people are coming from, where
• 13:46
  the cases are coming from, what insurance companies they're using, what members
• 13:49
  of staff are managing those projects based on skill set and experience,
• 13:53
  and then start to look at that.
• 13:54
  Law firm or any firm, we're still trying to, make a profit.
• 13:57
  We're still trying to grow our company.
• 13:59
  So how do those things tie back to actually what other, what insurance
• 14:02
  companies do you want to work with more?
• 14:04
  What kind of cases do you want to take more , because
• 14:06
  we can very specific nuances.
• 14:08
  And I think that's what large data sets allow us to do instead of
• 14:11
  having to be able to see things like, oh, I can see a 90 degree,
• 14:15
  change there that makes a difference.
• 14:17
  You're able to see 1 and 2 degree changes because you analyze the data and say, oh,
• 14:21
  now we see there's a real idiosyncrasy that comes into play there with these
• 14:24
  cases when they enter the door and the further back upstream we can go, then
• 14:28
  the more efficient that is for a firm.
• 14:29
  And the big thing that we've noticed with companies we work with is when they can
• 14:34
  identify sooner, a case that isn't a fit for them and they can let go of that.
• 14:40
  It really helps.
• 14:40
  And often that case is already taken a halfway through
• 14:43
  their system before they're.
• 14:44
  Oh, my gosh, is it going to be a nightmare?
• 14:46
  How do we go back on that?
• 14:48
  1 of the interesting kind of stories that we tell is that
• 14:50
  there's a client we've worked with.
• 14:52
  Over time, and they have, I think, 60 locations across the country,
• 14:55
  and they're dealing with the public in their real estate environment.
• 14:59
  But 1 of the things they find is when they started to overlay data about
• 15:02
  the performance of their different offices amongst those 60 offices
• 15:06
  they were trying to find, hey, which offices are performing better.
• 15:09
  And, in what geography and based on what, but they started to overlay a
• 15:13
  bunch of things like weather patterns in terms of these locations, the age of the
• 15:17
  employee base, in terms of how many there were new hires and young hires versus
• 15:21
  experienced hires, those that had gone through a technology training program.
• 15:25
  There's not so there was so many data points they put in, but after all of that,
• 15:28
  they started to realize there's a. Bunch of things they can now start to make an
• 15:31
  impact or make more predictive model off.
• 15:34
  And once they had a predictive model, it really started to help with their hiring
• 15:37
  plan, with their sales and marketing plan.
• 15:39
  But it takes a lot of playing with that.
• 15:41
  And I think that's the thing with AI and with data, you can't just
• 15:45
  dip your toe in it's really a business transformation journey.
• 15:48
  It's not an AI journey, and you've got to say, we're going to really set
• 15:52
  our business up to really use data in a very intelligent way, but we know
• 15:56
  it's going to be a two, three year journey before we even start to see
• 15:58
  some of the big impacts of that and they're able to understand the data.
• 16:03
  That makes sense to me.
• 16:04
  That's why I want to have this conversation with you now.
• 16:06
  A couple of years ago, we had put together a model, and we used Azure and we were
• 16:11
  given basically, about 40, 000 records from an injury from down in Louisiana.
• 16:15
  We stripped out all of the personally identifying information.
• 16:19
  But then we were able to take that model and the identifier was just
• 16:22
  a case number, or a client number.
• 16:25
  And then we were able to take that model and train it based on which cases
• 16:30
  were paying cases and which were not.
• 16:32
  And even at that point, probably about five or six years ago, the
• 16:35
  model was about 96 percent accurate as to whether it was going to be a
• 16:39
  paying case or not, the prediction.
• 16:41
  So we were able to take that model, enter new information
• 16:44
  using about 16 data points.
• 16:46
  And using those data points, we could say this is a pain case or
• 16:49
  not a pain case to your point.
• 16:51
  Like these models can actually start to inform they can inform the intake team
• 16:56
  and, intake teams require some training.
• 16:59
  They need to get spun up.
• 17:00
  And in any questionable case, it goes to an attorney to figure
• 17:03
  out is this a case or not.
• 17:05
  So even something like that.
• 17:07
  So, Just to kind of wrap up, I guess one more question is, what
• 17:11
  should these firms be doing now?
• 17:14
  In light of the fact that it's a two or three year process, what practical
• 17:18
  things can a firm do right now?
• 17:20
  Better position themselves for that period of time?
• 17:23
  So that they can.
• 17:25
  Really make use of the tools that are available now and what will
• 17:29
  be available in the coming years.
• 17:31
  Yeah, great question.
• 17:33
  It goes back to where we started the conversation, make sure there's
• 17:36
  just really good process within your company and make sure that
• 17:38
  everybody's accountable to it.
• 17:39
  It's repeatable.
• 17:40
  And it's followed by everybody in the team.
• 17:42
  And if that process there, then start to like, okay, where's the
• 17:45
  data going from that process at step?
• 17:47
  Let's just say it's a 10 step process.
• 17:49
  Data is transferring to somewhere along that process.
• 17:52
  Where is that data going?
• 17:53
  Where is this housed and stored?
• 17:55
  And, can you look at security of that?
• 17:56
  But once you know where that is, you can start like, okay, how do
• 17:59
  we consolidate those data sources?
• 18:00
  How do we build a single source of truth?
• 18:02
  And then once we're there, we can start like, okay, the world's our oyster.
• 18:06
  We have reliable, consistent data.
• 18:08
  Now let's.
• 18:09
  Get chip in the room and say, okay, chip, we want your and your
• 18:12
  team to come in and tell us how we think differently about this.
• 18:15
  How do we really drive a point of differentiation around this?
• 18:18
  And I think it really starts there.
• 18:20
  And I think then look at the data architecture.
• 18:22
  The biggest thing that we see companies struggle with is if you
• 18:24
  have not architected your data right in those early stages, it's
• 18:27
  very hard to come back and do it.
• 18:28
  You can certainly come back and do it.
• 18:29
  Everything's repairable, but it becomes quite painful.
• 18:32
  I think just taking the time to engage with a firm or specialist to say,
• 18:36
  hey, how do we really architect our data here where it's being stored, how
• 18:39
  we're storing it and what we're doing.
• 18:40
  And so I think that's where it is.
• 18:41
  But if they have that process and their data is being stored in an
• 18:45
  efficient, effective way, they can bring a company like yours in and be
• 18:48
  ready to go much more efficiently and what they're doing and Get the inside.
• 18:52
  So I think that's the big thing I would encourage, but human behavior
• 18:55
  is the hardest thing to change.
• 18:56
  But yet it's one of the most important things to AI readiness
• 18:58
  and benefiting from AI.
• 18:59
  So, yeah.
• 19:01
  That makes sense.
• 19:02
  One of the ways that we've approached at least a portion of that, when we
• 19:05
  talk about data architecture, things like that, I think that those things
• 19:07
  can sound more intimidating than they actually are, for my company.
• 19:12
  One of the things that we've done is we revisited just what
• 19:15
  our file storage structure is.
• 19:18
  And so when you think of that, we're on SharePoint.
• 19:20
  We use Office 365.
• 19:22
  We've moved our files around so that we have folders that only
• 19:26
  have active materials, right?
• 19:29
  We'll have a client folder within that folder.
• 19:30
  We'll have branding materials, headshots, content, things like that.
• 19:34
  But we made sure that we separate that from archived,
• 19:38
  materials that are outdated.
• 19:39
  We're not using anymore.
• 19:40
  Whether it's, base files interviews, things like that or final products.
• 19:44
  So we take the things that are no longer in use and we move them into an
• 19:47
  archive folder and then archive folder.
• 19:50
  We can do what we want to.
• 19:51
  So we can actually archive it, through office 365 products or elsewhere.
• 19:56
  But what that allows us to do is then use copilots that are built into
• 20:01
  SharePoint so that we could train a copilot on just the live data.
• 20:05
  I don't want the folks listening to this thinking, Oh, data architecture.
• 20:09
  We need to bring in your data scientists to move things around.
• 20:13
  We use these terms, I think, because we're in there all the time.
• 20:15
  But I think, for a lot of people, data architecture sounds much more
• 20:19
  intimidating than it actually is.
• 20:21
  A hundred percent I think one of the things you're bringing to my mind
• 20:24
  as we talk about that is just even thinking about naming conventions
• 20:28
  . Another good way for folks to start is bring your interns in this
• 20:31
  summer and have them look at naming conventions within your file structure.
• 20:34
  And what does that mean was like, hey, some again, maybe a silly example,
• 20:37
  but you've got folders that contain pictures, but some of those folders
• 20:42
  and files have the name pick on them.
• 20:44
  Some have pictures, some have photos, some have PHO like there's different
• 20:47
  conventions are named by and some of them that have a code going after them.
• 20:50
  That doesn't sync up that inefficiency makes accessing your data and
• 20:54
  processing a little bit more difficult.
• 20:56
  And I think if you just think of the consistent naming conventions of files
• 20:59
  that we know, it's called a case file.
• 21:01
  And if it's related to a certain category, it's got.
• 21:03
  2 initials related to that everybody uses the initials and a number and there's
• 21:07
  a validation process for that simple step is it's data architecture and it
• 21:13
  can be massive and its impact in terms of, how efficient you can use your data.
• 21:17
  I'm going forward.
• 21:18
  So you're right.
• 21:18
  It's certainly not rocket science.
• 21:20
  And the big thing, here's the thing I say, and it seems, redundant sometimes,
• 21:24
  but anybody who's listening to something on this podcast and there's
• 21:27
  terms we use, or, they want to dig deeper, literally just ask GPT a quick
• 21:32
  question , like, just jump in and say, Hey, what is data architecture?
• 21:35
  And as a law firm, how would I even start?
• 21:37
  , it'll give you, I don't know what answers it'll give you, but it'll
• 21:39
  certainly give you four bullet points to start to think about and tease
• 21:42
  out and talk to people about it.
• 21:43
  So, yeah.
• 21:45
  Awesome.
• 21:45
  Well, John, I appreciate it.
• 21:46
  I appreciate your time and your insight.
• 21:48
  Even if you're not the most technical resource within your company I love the
• 21:52
  way that you're able to kind of translate, and I think a lot of the work that we
• 21:55
  do is a translation service, right?
• 21:57
  To say, okay, here's what the technical folks are doing and here's
• 22:00
  how we're going to execute on it.
• 22:01
  So I appreciate your time.
• 22:02
  I appreciate your insight too.
• 22:03
  And I just want to say thanks for being on with me.
• 22:06
  Yeah, thanks.
• 22:07
  Appreciate the partnership.
• 22:08
  Appreciate the friendship.
• 22:09
  And yeah, this is fun.
• 22:13
  A big thank you to John for joining us today.
• 22:15
  You could find John through his website, peopleit.
• 22:17
  com or on Facebook and LinkedIn.
• 22:21
  Legal Marketing Radio is a production of Le Fleur Marketing.
• 22:23
  Find us at lefleur.
• 22:25
  marketing and on LinkedIn, Facebook, and Instagram.