Episode image for The Role of Technology in Creating Operational and Organisational Efficiencies
LOADING ...
Preview of episode

Want to listen to the full episode and all our other episodes?

Hearsay allows you to fulfill your legal CPD requirements every year.

Our yearly subscription is only $299/year.

With a yearly subscription, you can access all of our episodes AND every episode we release over the next year.

Episode 15 Buy Episode

The Role of Technology in Creating Operational and Organisational Efficiencies

Law as stated: 4 June 2020 What is this? This episode was published and is accurate as at this date.
In this episode, we discuss how technological and organisational innovation open up new opportunities to practice more efficiently, with a deep dive into e-discovery and legal process outsourcing.
Practice Management and Business Skills Practice Management and Business Skills
4 June 2020
Paul Bartholomew and Jon Prideaux
Union Outsourcing
1 hour = 1 CPD point
How does it work?
What area(s) of law does this episode consider?Technology in the legal profession; in particular technology assisted review (TAR) to achieve operational efficiencies, and the use of legal process outsourcing (LPO) to achieve organisational efficiencies.
Why is this topic relevant?Technological advancements have revolutionised the way almost every industry operates; including the legal profession. From software that assists with automation, to advancements in artificial intelligence, to organisational innovations such as legal process outsourcing (LPO), the practise of law is constantly evolving. This episode explores how these advancements are both changing the way legal services are provided, while also creating operational efficiencies that help lower costs and assist in solving the ‘more for less’ challenge faced by so many lawyers.
What legislation is considered in this episode?

 

What cases are considered in this episode?Both guests reference the Royal Commission into Misconduct in the Banking, Superannuation and Financial Services Industry 2017-2019. They each comment on the significant role that eDiscovery and other technology assisted tools had in facilitating efficiency during that Royal Commission.

There are no reported decisions referenced in this episode. Deidentified cases and examples are provided by both guests.

What are the main points?
  • Technology assisted tools can help firms increase efficiency in a variety of areas including discovery, due diligence, contract review and knowledge management.
  • Machine learning can be used to create operational efficiencies for particular types of tasks.  Machine learning is a type of artificial intelligence that uses algorithms to detect patterns in data and apply those patterns to new data to automate particular work flows.
  • Technology Assisted Review (TAR) broadly refers to any technology that makes document review more efficient and reliable. A reviewer uses a base set of documents (referred to as the ‘seed set’) to train the TAR software. Reviewers code the documents as relevant or irrelevant which is then fed into the predictive coding software. This process is ongoing, with the reviewers checking if the software is correctly identifying documents, then inputting more coding where needed. The software analyses the coded documents and creates its own algorithm for reviewing and identifying similar documents.
  • Continuous Active Learning (CAL) is a type of TAR that uses algorithms based on statistical modelling to identify similar documents from a base set often referred to as the seed set.
  • Futurist Richard Susskind (who we hope to one day join us on Hearsay) has written many books where he references the “more for less challenge”, referring to clients being less willing to spend money on legal services but needing more from their legal providers. This is particularly the case for process driven work which resulted in an increase in legal process outsourcing (LPO) from 2014-2016 as firms sought to reduce fixed costs by outsourcing process-based work to offshore jurisdictions at a much lower cost.
What are the practical takeaways?
  • TAR can assist law firms in streamlining tasks, particularly in litigation and also due diligence. This creates operational efficiencies and lowers costs for clients. However, TAR services have limitations, including that TAR cannot assess or consider privilege.
  • Human review is integral to the effective use of TAR.
  • The Supreme and Federal Courts have embraced the use of technology issuing guidance to assist lawyers.
  • Government bodies such as ASIC have issued detailed guidelines on the electronic production of documents.
  • No process or system is perfect, but perfection is not the aim. The goal is to implement processes that produce comparable, better results than your alternative. Human error is still significantly higher than the error percentage of TAR systems.
  • According to Forbes, the disaggregation of legal tasks fuelled by globalisation, technological advances and the GFC resulted in an increase in legal process outsourcing (LPO).
  • LPO was initially considered a solution that enabled law firms to lower their fixed costs, but some clients are now by-passing law firms altogether and going straight to LPO’s for process driven tasks.
Show notes 

ASIC – Document production guidelines issued March 2020

Surden, Harry, ‘Machine Learning and Law’, (2014), University of Colorado Law School

Part 10 of the Central Practice Note (CPN-1)

Part 4 of the Technology and the Court Practice Note (GPN-TECH)

Division 20.2 of the Federal Court Rules 2011 (Cth)

Supreme Court Practice Note No. SC Gen 7

David Turner:

 

 

 

 

 

1:00

Hello and welcome to Hearsay, a podcast about Australian laws and lawyers for the Australian legal profession, my name is David Turner. As always, this podcast is proudly supported by Assured Legal Solutions, a boutique commercial law firm making complex simple.

Just a quick note before we begin, the episode of Hearsay you’re about to listen to was recorded in the midst of the coronavirus crisis and as a result of social distancing measures we had to conduct this interview over remote technology such as Zoom or Google Meet, the audio quality might be a little different than what you were expecting. Still we think it’s pretty good in the circumstances and we hope you enjoy the episode.

Generally one of the most costly phases of any piece of litigation is the discovery stage. In appropriate cases, the use of electronic discovery commonly called eDiscovery, may assist a court and the parties in minimising the cost and burden of the discovery process. Joining me today to talk about the benefits of technology in the legal profession, in particular litigation and the discovery process, is Jonathan Prideaux, partner of the forensics team at KordaMentha. John, thanks so much for joining me on Hearsay.

Jonathan Prideaux:Thank you.
DT:Your background isn’t actually a legal background is it?
JP:

 

2:00

 

 

 

 

 

 

3:00

No that’s right, I neither have a legal nor an IT background in fact. I fell into the discovery space if you like in 1993 where I got a job at Minter Ellison in their document management services group. There wasn’t a lot of electronic or eDiscovery in those days. It was a very manual process. From there I stayed in the industry and worked at Clayton Utz for many years where I was the director and head of the legal technology services group, until recent times when I moved over to join the people at Kordamentha and their forensic technology group.

TIP: We’re about to start speaking about electronic discovery, or eDiscovery. For those of you not familiar with eDiscovery, this is a catch-all term used to describe technology that streamlines the traditional discovery process in litigation – though eDiscovery tools may be just as useful for, and may be deployed just as often for, transactional due diligence work. eDiscovery is usually invoked in contrast to the traditional discovery process which involves an army of paralegals and graduates huddled around a meeting room table working through archived boxes of hard copy documents well into the night. Such fun!

DT:

 

Now, a lot of our listeners will be, if not familiar with the process of eDiscovery then at least aware of it, in your various roles dealing with the discovery process what’s the typical experience of eDiscovery and litigation in Australia?
JP:

 

I’m not sure there is a typical experience. There is a standard workflow that eDiscovery professionals follow which is called the EDRM or electronic discovery reference model so we do tend to follow that model. It can be from the very smallest matters to the largest of matters, but the model is still very much applicable so that is first of all we identify potentially relevant information, we then collect what we need to collect, process it in a way that makes it reviewable, then make it available for lawyers to review, and then produce what’s relevant. And then from there on use that relevant material further in the litigation or the investigation process.
DT:

4:00

So it sounds like insofar as there is a typical experience, there’s that first stage review, a second stage review, and then that final stage of production.
JP:That’s exactly right. The idea is that at each stage you’re reducing the amount of information you’re dealing with and increasing the proportion of relevant information so as to get to the heart of the matter by the time you get to the trial, if in fact you get that far.
DT:Now that first stage of search, or that first stage of narrowing down the categories of documents you’re looking at, in years past might have been done by an army of paralegals or junior lawyers, how is that done now?
JP:

 

5:00

 

Well the initial search is done by people like myself who have experience in knowing where to look you know what systems to look at, how to identify relevant systems whether they be electronic or hard copy, or and with the electronic systems whether they’re held in house by the organisation or perhaps externally in the cloud or some other sort of hosted system. When we get to the review stage, yes you’re right it used to be done by armies of paralegals. I was one back in the early 90s that did that sort of work. Today we try and apply more technology to the process if you like to help cull that information down so that by the time it gets to be reviewed by the lawyers, we’ve reduced it as much as possible and you need the least number of lawyers as necessary.
DT:

 

There’s a few ways to access that kind of technology enabled review, some that some of our listeners in large firms would be familiar with the products like Ringtail, but there are also third party providers of eDiscovery processes, can you maybe talk a bit about the difference between using an inhouse piece of software and using a third party provider?
JP:

6:00

Well the large law firms who do a lot of this sort of work do have teams inhouse, that I know because I worked in one of those teams for many years. The sort of what you might call the mid-tier or the boutique law firms may not want to maintain a team of people and all the software and all the infrastructure that goes to, you know, that is required I guess to run these sort of processes, so as a result there’s a large third party industry. What that industry does actually is it kind of creates a level playing field so that the smaller or the boutique law firms with some very specialist and experienced legal skills can actually compete with the bigger firms who have, you know, more resources and more infrastructure to run these sorts of projects.
DT:

7:00

It’s a good way to describe it actually and I think that’s a theme that’s come up a bit when we talk about legal innovation, whether that’s technological innovation or organisational innovation or something else that it’s often about enabling a practice to operate at a scale that was previously reserved for a very large firm. I think what you’ve described, that third party process enables you to scale up or will make the cost of having that scale variable rather than fixed in your workforce, is something that’s really powerful for the mid-tier and smaller boutique firms.
JP:

 

 

 

8:00

 

 

Yes, that’s right. You know there’s some fantastic software out there, it is quite expensive to buy if you’re using it all of the time then it is cost effective, but if you only need it on a sort of case by case basis not an everyday basis, then it’s much more efficient if you like or cost effective to just bring that software in and bring that expertise in as and when you need it.

TIP: I’m now going to ask Jon about the increase in volume of data. According to Forbes magazine, 2.5 quintillion bytes of data – that’s 2.5 billion gigabytes – created each day and 90% of the world’s data ever was created in the past two years. Fun fact – one quintillion bytes is equal to one exabyte.

As the ability to collect and store data continues to increase, so too does our need for technology based tools to assist in the review of that data. At this scale and volume of data, it’s no longer about whether eDiscovery is more efficient than manual discovery – it’s about whether the discovery process is even possible without technological tools.

DT:

 

9:00

On the topic of volume do you think that the availability of technology tools that allow a large amount of documents and data to be scanned and discovered, has increased the volume of material that comes before the court or at the very least the amount of material that is reviewed in the course of a piece of litigation?
JP:

 

Well the technology that allows us to review large volumes of information is really developed to keep pace with the amount of information that modern day corporations are now creating and that’s really the reality of it. With electronic information messaging systems, email systems, other forms and other ways in which companies can create and store large volumes of information, the reality is there’s just more information for us to look at and to be considered for any case. Some of the larger international cases, you know the class actions or the product liability cases, the reality is there are just huge volumes of data to go through and we need the tools that allow us to go through those but even then at the end of the day often what is produced or considered relevant is huge in the millions if not tens of millions of documents in some of the larger cases.
DT:

10:00

John do you think you could tell us about an example of a case you’ve worked on where the availability of those tools was really the thing that made it possible to conduct that piece of litigation in the sense that it wouldn’t have been practical to conduct the discovery process in the absence of those kinds of protocols?
JP:

 

 

 

 

 

11:00

Sure, I think a good example and a recent example is the financial services Royal Commission.

TIP: We’re referring here of course to the Royal Commission into Misconduct in the Banking, Superannuation and Financial Services Industry, known colloquially as the Banking Royal Commission which commenced on 14 December 2017 and concluded on 4 February 2019 which resulted in the production and consideration of thousands of documents.

What the Commission required was the production of quite large volumes of information in very short timeframes. Timeframes that you know we’ve never really seen before in the eDiscovery industry. Projects that would have taken weeks or months were required to be done in in days literally, and you could not have done that without those technology tools, the tools to extract the data, transfer it to the law firm, or the provider, process review and then produce it you know that had to be done using the technology tools that were available and the associated processes.

DT:

 

 

12:00

And I think that’s a great example because of the timeframes involved. I think the traditional model of using an army of manual reviewers, is capable of reviewing a virtually limitless amount of material given enough time, but, and this was well publicised, the Royal Commission was operating at a pace that was difficult for many firms to to keep pace with and it sounds like it really wouldn’t have been possible in the absence of those technology abled solutions.
JP:

 

 

 

 

 

 

13:00

No, that’s right and in that case it was really a factor of time when it comes to volume some of the larger, again the international class actions, the product liability class actions, involve huge numbers of documents and again from the volume point of view you would not be able to get through though that sort of volume of documents, you know, millions or tens of millions of documents were it not for the technology and the techniques that you can apply to help reduce those volumes and speed up the actual review.

TIP: I’m about to ask Jon about Machine learning. “Machine learning” refers to computer algorithms that are designed to detect patterns in data and then apply these patterns going forward to new data in order to automate particular tasks.

Machine learning has been used outside of the legal space to automate tasks that were once thought could only be done by a human, including language translation, fraud-detection and data-mining.

If you have an interest in machine hearing and the law, there is an interesting article by Harry Surden from the University of Colorado Law School – we will provide a link to this article in the show notes.

DT:I want to ask you now about the use of machine learning as a replacement for manual review, do you have a kind of industry term that’s usually used for that?
JP:

 

 

14:00

 

 

 

 

 

 

15:00

 

 

 

 

 

 

 

 

16:00

 

 

 

 

 

 

17:00

There are a number of terms and technology assisted review or TAR is certainly one. The latest iteration of TAR is what’s called continuous active learning or CAL for short, those tools are very useful.

TIP: Ok we’ve got a lot of acronyms here, so let’s start with TAR, or technology assisted review. This refers to any review system where it’s the computer software itself that is classifying or coding documents to assist with the lawyer’s review. TAR software is based on machine learning. At first, the lawyer teaches the software what to look for in a large set of documents, and then the software goes off and classifies the documents according to what it has been taught.

Then we have CAL, meaning continuous active learning, which is a type of TAR. Now to understand continuous active learning, we have to understand what came before it, simple active learning or SAL. In simple active learning, a human lawyer would train a machine till the machine’s coding algorithm became stable, usually by showing the machine an initial sample of documents and then checking its work in the early stages. But SAL, simple active learning, machine learning algorithms, stops getting better once they get to that stable point; they don’t keep learning after that. The continuous active learning software keeps learning throughout the entire discovery process. It’s constantly getting feedback from lawyers and updating its finds and classified documents, getting better all the time. That’s CAL.

Our discussion with Jon will focus on TAR, or technology assisted review which will be explained in more detail over the next few minutes.

They’re proven to be useful by the statistical output, you know, we do a lot of tests to test the accuracy of those projects where we do use TAR or CAL and that kind of justifies the use if you like. There’s lots of different ways in which you can apply TAR though and also TAR is not always the right solution on every matter. It’s not a silver bullet if you like. There are certain matters where the types of information or the categories of information sort lend themselves to using TAR and then there are others where they don’t. For example, TAR’s not very good at picking privileged documents because TAR looks at content and not purpose, and privilege tends to be more about the purpose rather than the content in many cases. So we have used TAR or technology assisted review and it is a fantastic tool if used properly and in the right circumstances.

TIP: Let’s talk a bit more about the concept of predictive coding. Predictive coding uses an algorithm to recall documents that are conceptually similar to a sample set, referred to as a ‘seed set’. For example, in the sample set there might be a plane ticket from Sydney to Los Angeles, and documents such as a travel itinerary, passport, boarding pass and hotel voucher might all be returned as conceptually similar by reason of their connection to the SYD-LA journey. This is distinct from an algorithm that identifies documents which are formally, if not conceptually, similar. For example, like an algorithm that can examine a sample set of leases and then find more leases.

Then there we have the concept of clustering or mining, which focuses on the organisation and display of document sets or search results in an interactive, graphical format, where documents that are similar (that is, conceptually similar) around each other. That can also include software that assists with email threading, where common emails are indented and linked together.

DT:

 

Let’s talk for a minute about how TAR works mechanically and I think part of this really came out in your example about privilege, TAR works using natural language processing doesn’t it?
JP:That is correct.
DT:

18:00

It effectively reads the document the same way a human being would to try and identify content, but as you say identifying content and the meaning behind it can be very different things. One criticism I’ve heard of technology assisted review, and I don’t think it’s a fair one for a reason I’m hoping you’ll be able to tell me, is this idea that it has a margin of error because it can understand content but not meaning. Can you tell me a bit about the margin of error? I know that you collect a lot of statistical data as you say about the accuracy of the process. How accurate is it?
JP:

19:00

 

 

 

 

 

20:00

 

 

 

 

 

21:00

 

 

It’s much more accurate than a traditional human review, let’s put it that way.

TIP: The imperfection of human review is best exemplified by a 1985 study completed by Blair and Maron, researchers from the universities of Michigan and California, respectively. In that study, a team of lawyers and paralegals searched for documents using search terms and iteratively searching, that is, reviewing discovered documents and then using new search terms based on what they found. The lawyers and paralegals estimated that using their iterative search process they had found at least 75% of the relevant documents. How many did they actually find? Just 20%.

So the discovery process has never been a perfect science, it’s always been about doing a reasonable job and a job that is as accurate as possible. At the end of the day, it’s a subjective process so there perhaps is no right or wrong answer, but what you’re looking to do is collect and produce as much relevant information as is possible in a reasonable time frame for a reasonable amount of money. So there’s a term in the industry which is proportionality and you always want to take a proportional approach to these exercises. So the good thing about TAR and active learning is it’s very consistent. You know human beings are different, if you have 10 people reviewing documents one of them might be having a bad day or they might not be well, they might have other things on their mind and they’re therefore very inconsistent and they’re generally inconsistent in the way in which they interpret a document or some categories of discovery. The good thing about the technologies, it’s very consistent in that respect so it’s a defensible process. It’s also not used in isolation. You still need the human reviewers to actually teach the TAR model what is and isn’t relevant. So it’s not about one or the other, the best result is normally where you get some subject matter expertise and then apply the technology as well to help with the review.

DT:

 

 

 

Can you tell me a bit more about the process of teaching the TAR algorithm to find what you’re looking for? I’ve always found this idea of teaching a natural language processing engine really fascinating, it seems like such an interesting interaction between man and machine to kind of teach it over the course of weeks, I’ve heard of using the due diligence context for example that you teach a natural language processing engine this is what a change of control clause looks like go out and find change of control clauses.
JP:

22:00

Sure. So it’s based around sort of conceptual analysis and those algorithms that we talked about and the algorithms identify similarities between documents at a conceptual level so they don’t rely on the existence of specific keywords. They understand the similarities and they learn from looking at the content of large volumes of documents, they learn what are conceptually similar documents and similar concepts within different documents even though the language might be a little different.
DT:What is an example of a concept?
JP:

 

 

23:00

So the concept might be for example an automobile. So in one document for example it might talk about a car or automobile in another document it might talk about Bronco or Mustang or something like that, and over time the algorithm will understand that a Bronco or a Mustang is in fact the same thing as a car or an automobile even though the language is quite different.
DT:And so the engine looks for concepts rather than specific phrases, it’s more than just a search engine?
JP:

 

That’s right, and where a reviewer identifies a relevant document, the TAR system will go away and find documents that are conceptually similar and then assume. So it’ll go away and find documents that are conceptually similar and then present those the reviewers suggest that are also relevant because of the conceptual similarity. And it continues to learn. It’s what’s known as a binary process so yes the document is relevant or no it’s not. On that basis it will find other similar documents that are both relevant and not relevant.
DT:

 

24:00

Thanks for explaining that, I just always find it so interesting this idea of teaching the machine to do the work and developing into a more accurate reviewer at a time. Just before we leave technology assistant review, you raised a great point about the accuracy of the review process not being perfect, but certainly no process is ever perfect and certainly the manual review process is significantly less perfect, and of course the federal court has given it’s imprimatur to technology system review in a number of decisions. Speaking of the federal court, because I want to talk now a bit about how the courts facilitate the use of eDiscovery through the protocols that they enacted, the federal court is really a leader in this area in Australia isn’t it? Can you tell me a bit about the federal courts eDiscovery protocol?
JP:

 

25:00

 

 

 

 

 

 

 

 

 

26:00

 

 

Sure, the protocol is referred to as GPN dash tech. It’s existed in many forms for quite some time even in the other court jurisdictions there tends to be more compliance with the federal court protocol. It is probably the gold standard if you like.

TIP: The Federal Court of Australia has issued guidance on the use of electronic discovery which can be found in:

  • Part 10 of the Central Practice Note (CPN-1)
  • Part 4 of the Technology and the Court Practice Note (GPN-TECH)
  • Division 20.2 of the Federal Court Rules 2011 (Cth)

Similarly, the Supreme Court of NSW has issued Practice Note SC Gen 7 which broadly covers guidance around the use of technology in courtrooms and jury deliberation rooms, as well as in civil litigation and discovery more generally.

We will put a link to these references in the show notes.

What that protocol does is set out some guidelines for the way in which party should exchange documents between them. It suggests that they consult early to make sure that they’re both, that all parties are using a similar type of process and that they agree upfront on the way in which documents should be exchanged so that’s the data that’s exchanged with them, the format of the documents themselves to be exchanged, whether they are exchanged for example in native format or converted to a PDF, and the numbering system that is used.

DT:Do you have any tips for practitioners who are using eDiscovery tools, particularly technology assisted review, in narrowing the category of documents that the tool will produce? Do you have any tips for using the software in a way that’s going to produce truly relevant material?
JP:

 

 

 

27:00

Yes, there are ways in which you can do that, if you’re the recipient of a large volume of information because you’re the plaintiff in a class action, you can use the tools to find key documents if you like what used to be called the proverbial smoking guns. If you have a document that is particularly relevant you could use that as a seed document and ask the software to go away and find anything which is similar. So that can be a useful technique. So again the same sorts of review techniques to prioritise or find documents which are most likely to be relevant can be used to go through a large discovered set, or if you’re the producing party to identify documents that might cause you a problem or might be key to either your defence or may be problematic for your defence, to find those documents early on in the case, again you can use that search capability whether that’s conceptual search or keyword search, you know the more traditional types of searches.
DT:We’ve talked a little while about the receptiveness of courts to eDiscovery tools and more recent tools like technology assisted review, but we also talked earlier about the Royal Commission and how important it was to the efficient running of the Royal Commission that these technology assisted tools were used by respondents. What are other regulators doing in terms of using technology to assist with production notices for example ASIC or the ACCC?
JP:

28:00

 

 

 

 

 

29:00

The regulators, ASIC and the ACCC, are actually very progressive in this space. They use a lot of technology themselves to analyse and review documents that are produced to them, but they are also very much aware that people responding to them will likely use these sorts of technology tools, and by doing so they are likely to give a better result and a more timely one. ASIC recently issued a revised set of guidelines for the production of documents.

TIP: In March 2020, ASIC issued document production guidelines that provide detailed guidance how to respond to a notice to produce from ASIC. The majority of these guidelines focus on the electronic production of documents, including the treatment of metadata as well as acceptable file formats for both documents, emails, text messages and voice recordings. We will provide a link to ASIC’s guidelines in the show notes.

So normally with ASIC, it’s a section 30 or 33 notice that’s issued to a party to produce documents, the ASIC guidelines prescribe the sorts of technologies that should be used and the protocols for producing those documents. So really it’s a very progressive and very useful, and a really good document I think if you look up those guidelines. The ACCC similarly has protocols that it likes responding parties, so typically a section 155 notice, to produce documents and again it’s a very similar format.

DT:

 

 

30:00

 

Let’s talk about the trial now and the use of an electronic court book, again the federal court is really where the gold standard for this seems to be at least that’s what I’ve heard. By using consistent referencing you’re able to consistently refer to documents from the discovery bundle through lay affidavits and expert evidence, then through to submissions and the electronic court book, do you have any insights on how much time and money that saves by having that level of consistency throughout the process and having a technology enabled hearing?
JP:

 

 

 

 

31:00

 

 

 

 

 

 

32:00

 

It is hard to put a finger on it of time and cost, but we do know anecdotally that using those sorts of techniques that you just mentioned as well as the trial systems does save a lot of time and therefore cost. I once worked on a federal court matter that had six parties to it. The parties agreed on a 60,000 page court bundle. They actually exchanged all the documents using the proper processes using the federal court practise note and following the guidelines all the evidence had been prepared with common referencing and then when it came to the trial for some reason, the parties decided not to use an e-book, instead they reordered the 60,000 pages into chronological order. Renumbered all of the documents from one through to 60,000 and then printed them out. There were six parties, each party printed 4 copies and then there was a copy for the judge. So if you do the maths, that was 1.5 million printed pages before day one of the trial. Then what happened is from day one they started to insert new documents, supplementary documents, into the bundle. Now because the documents were already chronologically sorted and sequentially numbered, they had to be slotted in at the right point in the chronology. The mechanics of doing that costs lots of money because there were 25 separate copies that had to be updated rather than a single central electronic source that could then be communicated to all parties or replicated to all parties if you like. Practically it didn’t work either because the folders were full and you couldn’t actually fit anymore paper into the physical folders, so they then started creating supplementary bundles, and that then had the effect of obviously meaning that the entire set was no longer chronological. So it wasn’t a great idea, it’s one of those examples unfortunately of where the technology wasn’t used and that trial I think went from 10 weeks to around about 14 weeks. So those extra 4 weeks would have cost a lot more money than would have been expended had a new trial provider been engaged.
DT:It is a really tangible example of how much time and money using the technology can save. Can you tell me a bit about what they could have done by using the e-court system?
JP:

 

33:00

 

 

 

 

Yes, sure, that 60,000 document bundle could have been uploaded to any court system. The numbers from the discovery could have been retained. All of the references about where those documents had been used in evidence would have been retained. The meta from the discovery process also could have been reused because the e-trial environments are essentially very similar to the review platform environments; they look pretty much the same. There would have been no need to renumber and then actually change all of those references in all the copies of the evidence as well and of course there would have been no need to print 1.5 million plus pages. The critical thing about the e-trial environment is because it’s centralised and run by an independent provider is that all the parties actually then get access to exactly the same thing. I could imagine that those 25 copies of the court bundle by the end of the trial were certainly not identical anymore because of the manual handling and the error in inserting new documents throughout the course of the trial.
DT:

 

34:00

 

 

I think it’s always a source of anxiety for trial counsel when you’re handing up a court book or maybe 10 or 20 volumes of a court book and you’re asked, and the copy is obviously identical to all of the other copies, it’s supposed to be of course, but you can really only say that on faith when it’s 8 copies of a 60,000 page court book because there’s no way to know with any certainty. Whereas as you say, using technology you can be absolutely certain that everyone is working from the same set of documents and you can also ensure in the room that everyone’s actually looking at the same document because the technology in use in the physical courtroom enables parties to be referred to those electronic document IDs.
JP:

 

 

That’s right, and it’s a great time saver because as counsel reads out a document ID the court operator types it in and by the time they finish reading out that ID the document has appeared on everybody’s screen instantaneously. That doesn’t happen in a paper environment as you know each of the counsel or their juniors goes to the relevant volume, turns the pages to find the document and everybody waits until everybody has the document in front of them.
DT:

 

35:00

We spoke a bit earlier and again talking about the federal court because the federal courts process does seem to be gold standard in Australia, that the federal court’s protocol is technology agnostic that its desire to facilitate using whatever platform or software the parties choose to use. There are a number of providers that are pretty well established in Australia but in your experience how malleable to new products and new technology processes are the courts’ processes? Is the court receptive to parties using new technology?
JP:

 

 

36:00

I’d say there’s a mix. The federal court can be very progressive in some cases as in some of the Supreme Court jurisdictions as well, there are initiatives such as e-filing that have been around for a while that make filing legal documents with the court a lot more efficient, if you like so those are fantastic initiatives. In terms of adopting technology within the court rooms, that’s perhaps been a little slower, you know courts are very traditional environments if you like. So you know that technology has taken a while I think to be widely adopted. It’s often on a case by case basis. Each judge will have their preference or their opinion on how a trial should be run. Some are very open to using technology, some are agnostic they’ll let the parties decide and some have an opinion that they don’t want to use it.
DT:

 

 

37:00

I think you’ve raised a really good point there about some complaints that lawyers, and also judges, have about trolley load litigation; a glut of information put before the court. As you say, corporations produce more data today than they ever have before and that creates an initial sample size that’s of unprecedented size but I think the challenge for us as practitioners is really to use the tools that are available to us to find what is absolutely relevant and germane to the real issues in dispute from that large sample. And I think there’s nothing inconsistent with the just quick and cheap resolution of the real issues in dispute starting with that live sample so long as there’s not then 50,000 documents that are just for background. I think there’s a temptation sometimes to say well we use this tool to review 10,000,000 documents, it found 100,000 that are relevant, we might as well put them somewhere in the bundle even if we never refer to them again.
JP:

 

 

38:00

 

That’s right, and there is a tendency when it comes to trial to, if you like, for the parties to cover all bases. They’re not sure which documents they necessarily want to refer to even though there is that core set that’s been referred to in the evidence preparation process so there those documents are obviously key, but there may be others. As well, given if you have an electronic court environment there’s no limit to the number of documents you can include in a trial model because there’s no physical space requirements in a traditional or a hard copy courtroom you know cupboards for, book shelves full of folders of hard copy documents may be a little more prohibitive. In an electronic environment you don’t see them, so it does actually allow a large volume of information to be considered or at least sit there in the background as you say during the trial and then being referred to if necessary.
DT:

 

 

39:00

 

 

 

It sounds like technology could be an enabler in terms of avoiding trolley load litigation in the sense that it could be used cleverly to identify documents that really are relevant rather than identifying a class of documents that are tangentially relevant but ultimately not referred to a hearing.

TIP: Now privilege is a contentious area when it comes to eDiscovery. Some have concerns that the technology is not sophisticated enough to identify documents containing privileged material and worry that production of such documents might inadvertently waive privilege, while those on the other end of the spectrum might believe that that the risk of error, or making a mistake and perhaps inadvertently waiving privilege, is higher actually when humans are involved.

DT:Can you tell us a bit about an example of a situation where discovery or disclosure was inadequate?
JP:

 

 

 

40:00

Certainly. I mean without going into details, I have been involved in matters on both sides where the discovery has ultimately turned out to be inadequate. And that might be because the right data wasn’t collected in the first place, obviously if you don’t collect the data then you can’t possibly produce it, or because the way in which the searches were applied or the review was done meant that relevant documentation was excluded. That can also because problems that means parties maybe haven’t discharged their discovery obligations and that can bring an adverse inference, or it might mean that all of the parties don’t have the relevant information before them to further their case and ultimately end up with a just and fair resolution. So it certainly is problematic. It is important that the process is done well because if it’s not done well and the case continues, a lot of work might have to be done again, and that causes a delay and also causes extra cost.
DT:

 

 

 

Just from a practical perspective, how do you know if the process has been done well or not? I imagine with such a large volume of material the process initially undertaken might still produce a large amount of material from that initial sample that appears relevant to the proceedings and it might be difficult to identify what has or hasn’t been missed. Do you have any advice or experience in terms of the QA process? How do you ensure that your eDiscovery has been adequate?
JP:

41:00

 

 

 

 

 

 

Yes, I mean QA is critical throughout the whole process and there are techniques you can apply to make sure that you have produced the right information. So for example you could look at all your documents in a timeline and you might see that at a certain point in time weeks or months that there are less documents than you might expect to see compared to the rest of the timeline. And that might be that you know you haven’t collected those documents in that time period, perhaps they were archived or put into a different system or something like that, so you can do those sort of statistical QC checks. Often what happens is parties to a dispute, they both have the same documents on both sides because they’ve been emailing each other and if one party produces the document and the other doesn’t then that raises questions again about the quality of the process when a document hasn’t been produced. So that often highlights gaps if you like or inadequacies in the discovery process.
DT:

42:00

Before we finish up, some of our listeners might be very familiar with the discovery process, some might be embarking on the discovery process for the first time, and some might be faced with a matter or discovery process where they’re contemplating for the first time using a tool like one of the ones we’ve discussed today. What tips do you have for our listeners who might be dealing with eDiscovery for the first time or if not the first time who are relatively inexperienced with eDiscovery to get the most out of the process?
JP:

 

 

43:00

 

 

 

 

 

 

44:00

 

There’s a number of tips I would say. Firstly, make sure you’re aware of any relevant guidelines or protocols if it’s the federal court, or the Supreme Court or a regulator or a commission of inquiry, they all publish relevant guidelines and usually there is a template protocol that you can use as a starting point, if you like if you don’t have one yourself. Critically talk to an experienced eDiscovery professional. It’s important not to rely on IT professionals, IT professionals have a role but eDiscovery is quite a niche specialist industry if you like and it does really help if you have the right expertise in the right advice. I would say you know also make sure you’re aware of the available technology in leverage that technology as we said you can leverage technology or the smaller the boutique firms can leverage the technology to help level the playing field with the bigger firms and to also help comply with the requirements of the regulators or commissions of inquiry as well which can be can be quite onerous. I would say that adopting some of the more simple basic technologies can bring or give the best return on investment, if you like. So consistent referencing once you’ve done your discovery consistently referencing those documents throughout the litigation life cycle will bring a huge benefit in a number of different ways and that that shouldn’t be underestimated, but also be aware of the more advanced technologies such as technology assisted review and apply those in the appropriate circumstances and again that’s not all circumstances that lend themselves to you using those sorts of tools but that some certainly they do.
DT:Those are all great tips, and Jon thanks so much for joining us today on Hearsay.
JP:Thank you very much, thank you for your time. You’re welcome.

Part 2: Paul Bartholomew

DT:

 

 

 

 

 

45:00

 

 

Those were great tips from Jon that demonstrate how technology can be used to improve legal processes, in particular the discovery process. But technological innovation is not the only way law firms can innovate. Talking with us today about innovation changing practice of law, I’ll now be speaking with Paul Bartholomew, the COO at Unison Outsourcing to talk about organisational innovation in the form of legal process outsourcing.

Technological change constantly changes the way just about every industry operates, and the legal industry is no different. Automation is changing the way we do discovery, contract review and knowledge management. Meanwhile the theoretical smart contract is fast becoming a practical reality, but these advancements in technology have also increased the attractiveness, and correspondingly the growth, of a kind of organisational innovation in the form of legal process outsourcing or LPO. LPO providers use a flexible, and often virtual, low-cost workforce in different locations around the world to perform tasks that traditionally were performed in-house or at a law firm. This includes tasks such as research, contract review and due diligence. Now while LPO can reduce costs, some have been reluctant to jump on board because of concerns about confidentiality, quality and unreliability. Joining me today to talk about legal process outsourcing is Paul Bartholomew, COO at Unison Outsourcing. Paul, thanks so much for joining us on Hearsay.

Paul Bartholomew:Thanks for having me David, pleasure to be here.
DT:Now let’s start with the basics: what is LPO?
PB:

 

46:00

 

 

 

In its simplest form, LPO is legal support services provided to law firms or in-house teams by an external organisation, usually not a law firm. The concept really is the disaggregation of legal services or legal functions into smaller parts. Those parts are most likely to benefit from being performed at scale. Firms come to us really because they want process efficiency, economy of scale, and just general flexibility in how they deliver those services. I think most people though when they think of outsourcing they imagine a large factory like office in some emerging markets with endless cubicles and you know a workforce of drones running around piling over documents, but slowly though I think people’s perception of what LPO is, is changing because that’s just not the case anymore.
DT:

 

I’m glad you mentioned that traditional conception of LPO because I think a lot of people think of LPO, or just outsourcing generally, as something that’s driven by labour costs in an emerging market that ‘oh well this is made possible because the minimum hourly wage in another country is lower’ and therefore that that’s how it’s achieved. But it is actually as you say about scale, isn’t it?
PB:

47:00

 

 

 

 

48:00

Yeah absolutely. I think if we would be honest in our own industry that it was a labour arbitrage play initially but technology has just decimated.

TIP: Paul has just referred to ‘labour arbitrage’. What is that? Arbitrage refers to taking advantage of a price difference between two or more markets. Paul is referring to the saving on labour costs by using cheaper labour, usually offshore. For example, if a large law firm is hiring 100 paralegals to help review documents in a royal commission, it would be cheaper and result in labour arbitrage to engage a LPO in the Philippines to complete the same task because labour costs are lower in the Philippines than in Australia.

I mean if you talk to someone in a discovery or due diligence team you know that when a project used to take 20 people a month to complete, it’s now done with two people in a couple of days, the software does so much of it. So LPO over the years has adapted to that as well and it started to offer different services and offer its own technology along the way.

DT:

 

It’s interesting to see them as a complementary service rather than as a competing one. I think sometimes people think in terms of dealing with a large discovery task being they’ve got eDiscovery on the one hand and that’s an exclusive option to using LPO as an alternative. But it’s interesting as you say that one has influenced the other.
PB:Yeah look, I think the market force has really just driven where firms have gone and you know if you don’t innovate and adapt, you die. You know LPO got its start by that labour arbitrage but it’s quickly sort of pivoted I think.
DT:

49:00

Now LPOs have been around globally for some time, but when did it really become an option available to firms in Australia?
PB:

 

 

 

 

50:00

I’d say circa sort of 2005/2006 we started to see it in Australia. Was adopted earlier on by some of the global law firms that came to Australia, obviously they’d already been doing it, but I don’t think it really got any traction here till the GFC, post GFC. I think firms then became concerned, needed to look for ways to reduce their cost, so that they embraced it more that way but then also too I think of that time everybody was under the same financial pressure to client with demanding more for less and they still are 10 years later.

TIP: ‘More for less’ is a phrase we hear often as lawyers, referring to clients being less willing to spend money on legal services but needing more from their legal providers.

This is particularly the case for process driven work which resulted in an increase in legal process outsourcing (LPO) from 2014-2016 as firms sought to reduce fixed costs by outsourcing process-based work to offshore jurisdictions at a much lower cost.

DT:

 

The more for less challenge, those words we talk about constantly in legal innovation, and I think you’re right in those years immediately after the GFC there were those knock on effects where clients were experiencing financial pressure and you do need to deliver more for less when the legal budget is under greater scrutiny.
PB:

 

51:00

I think what will be interesting too is to see whether or not, you know the pandemic that we’re in the middle of here with Covid and people go into work from home arrangements, whether that forced flexibility that’s been pushed on all of us to adopt, whether that will now change people’s opinion to having remote legal teams. I just think in general they’re going to be more comfortable with it, so I think if anything it’ll really speed up the adoption.

TIP: Historically the take up of technological innovations is influenced by a number of factors including the availability of pre-existing alternatives. The COVID-19 pandemic has created a burning platform where many industries were forced to embrace technology, such as Zoom, DocuSign and other cloud-based platforms, in order to continue business. Similarly, state and federal governments passed temporary legislation to enable the electronic signing and attesting of documents, but that might be there to stay. For example under the Electronic Transactions Amendment (COVID-19 Witnessing of Documents) Regulation 2020 (NSW) some deeds can be signed electronically. COVID-19 has in many ways forced the advancement of technology upon the legal system, regardless of one’s preference for more traditional methods.

DT:

52:00

It’s an interesting effect that the coronavirus crisis has had, some of our listeners will notice if they’ve listened to our episode with Nicola Martin the employment practice group leader at McCabe Curwood, that many employers who would never have considered having a large scale remote workforce or a large scale experiment of remote working have really had no choice but to trial it and might be surprised, pleasantly surprised, at the results.
PB:Anecdotally that’s certainly what I’m hearing from our clients as well, and it’s helping us do our job. I think the adoption of technology, even just the simplest of technologies like people going into the cloud and people starting to use products like Zoom and things like that, it’s just getting everybody more comfortable with that remote workforce.
DT:

 

Now we talked a bit about LPO providers being based in offshore locations, particularly South Africa, India, the Philippines, Ireland is one, how much outsourced work stays in Australia? And perhaps you can talk about Unison’s offering in particular there.
PB:

53:00

 

 

 

 

The LPO market is still growing at a, I think quite extraordinary pace from what I saw recently in 2019 it grew another 30%. The latest set of figures I saw are around $6.5 billion globally so it’s quite large. Realistically though, for options that are based here in Australia there’s not a lot. On the whole I would estimate we would be lucky to retain half a percent of what’s outsourced here in Australia. Which is a real shame because I think everyone’s sort of acknowledged that the people most impacted by outsourcing has been junior lawyers, grads, paralegals, the big firms used to employ them in their thousands and now it’s in the hundreds and it’s just because that need is not there anymore. Great, talented and educated people are coming through the ranks and they’re just struggling to get their foot in the door and get that start.
DT:

54:00

 

Absolutely. That onshore/offshore dichotomy brings to the fore a lot of the traditional concerns about LPO, I suppose which I mentioned earlier in the episode those three being quality, confidentiality and data security. A lot of those concerns are driven by an offshore provision of services because there may be different data security regimes applying in that jurisdiction in which the services being offered and there may be a perception that legal training is of a different quality in that jurisdiction as well, but you avoid a lot of that with onshore LPO, don’t you?
PB:

 

55:00

It’s interesting you bring up, that’s certainly part of our pitch in Unison is that you get an Australian based team that for the vast majority have all been educated in Australian universities. You know we’re here in Wollongong, so particularly for Sydney clients you know if you want to come and see where we do our work well it’s an hour’s drive and you can work with the team closely, you can meet all the lawyers involved, the project managers involved, and that sort of option is just not there for an offshore.
DT:

 

 

 

 

56:00

I like the way you describe the difference in the work in terms of work that requires a particular sensitivity to the unique features of an Australian jurisdiction, compared to something that might be common to common law jurisdictions all around the world. You know If you look at a process like discovery, I imagine that finding documents in particular categories that much is the same in every common law jurisdiction, but there might be some tasks where you do need a unique understanding of the way Australian courts operate or the way the Australian authorities have been applied and it’s interesting to have that dichotomy. That really brings me onto my next question about the kind of tasks that are done through LPO, tell me a bit about what you can and can’t do through the legal process outsourcing regime.
PB:

 

 

 

Certainly, the work lends itself to anything that’s process driven, anything that traditionally is high volume, so yes discovery and due diligence has at least traditionally been the lion’s share of the work. I think moving forward though, particularly as in-house clients are starting to embrace outsourcing more, we’re starting to see legal functions being outsourced that require a level of expertise. So we have some clients who we manage contract negotiations from start to finish within a defined parameter, so we authorise to negotiate to set amount, set terms, it’s all set down in the playbook, nothing revolutionary about it.
DT:

57:00

It’s interesting you mention the use of LPO by in-house teams because I think it’s probably a service that originally only law firms and perhaps quite large ones would have availed themselves of, but it is now available as a viable option to in-house teams, isn’t it?
PB:

 

 

 

 

 

 

58:00

 

 

Absolutely. I think their needs are very different. So I can’t speak for all outsources, but in Unison we’ve probably got a 50/50 mix of clients. Our law firm clients come to us for all the traditional things, because essentially a law firm client wants the ability to have flexible resourcing, so when the client comes to us interestingly enough, cost is not usually the first thing. It’s a ‘I’ve got massive projects but they come in sporadic bursts, can you help me with 30 bodies on the ground if we’re doing hard copy, can you provide me with you know 20 people here doing eDiscovery.’

TIP: This is an interesting observation here by Paul. Law firms have high fixed costs because labour is expensive in the legal services market. We’re paid pretty well us lawyers. These costs are fixed because they can’t be scaled up or down, at least in the short-term. Now often when law firms look to cut costs, they’ll consider redundancies to reduce their labour costs, giving redundancy packages to those who leave. But when they need that labour capacity back again, then they have to pay matching costs to find new staff, such as recruiting commission, advertising and the cost of training new people. This is a delicate balancing exercise. Law firms need a large enough workforce of permanent staff to attract and build client relationships, without having such a high fixed cost that it can’t withstand turning down revenue. It’s a very common challenge for all professional service firms.

DT:

 

I’ve never thought about it that way, but that just solves such an economic problem in professional services, which is particularly for lawyers, that so many of our costs are fixed but our revenue is so spiky.
PB:

59:00

 

Absolutely, and the headaches that partners and HR managers and legal project managers face when a large project hits and there’s scrambling all across the firm to try and find a paralegal to join his team to do work, what we give them is the flexibility to say we don’t need to carry a headcount of 100 for the 30% of the time that we need them, we can carry a headcount of 50 but then we can scale up to 150 if the projects demand that. Certainly resourcing the last few years for the Royal Commissions has been particularly difficult for a lot of the firms.
DT:

 

 

1:00:00

In terms of converting fixed salary costs to variable costs that can scale as work increases or decreases, it’s quite similar to some of the contract platforms for contracting lawyers, gig economy sort of platforms if you like, for lawyers that we’ve seen offered and perhaps by some of the large firms but also coming out of the UK such as Lexoo that offer that flexibility directly to in house teams.
PB:

 

 

 

 

 

1:01:00

 

Absolutely. So the vast majority of our workforce would be paralegals, and then the vast majority of the paralegals we employ actually are law students out of one of the local universities predominantly but then also some from Sydney. The beauty of the model is that while our law firm client experiences spikes in demands, we get to sort of flatten that out by having scale across various law firms.

TIP: This is a great example of diversification – Unison as an onshore LPO generates work from two key groups, by selling to law firms and selling to businesses. By targeting these two different groups, Unison reduces its risk because the sudden drop in demand from one segment is likely to be ameliorated by stable profits from the other segment.

So if one law firm’s busy they might take up more of our resources but next week they’re quiet, but then there’s another one that picks up, so we get the opportunity to level it out across multiple firms.

DT:We spoke a bit about some of the tasks that you can perform through legal process outsourcing and the two that often come up, and you mentioned before discovery and due diligence, that touches on both litigious work and transactional work, but what are some of the tasks that you see coming through Unison that many lawyers wouldn’t expect to be done by a legal process outsourcing provider?
PB:

 

1:02:00

Sure, so Unison is again you know a bit of a niche in the market, but an interesting development we’re starting to see is an increase in the small firms that are coming to us. So we see a market of you know that sort of 2-4 partner suburban or regional based firm coming to us saying ‘hey, you know, what can you do for us?’ For example with conveyancing transactions in residential property we now have a service where we handle basically everything from exchange to settlement and with the advent of PEXA, it’s just making that even easier across all the states. We have services where just a certain function of their team we perform as well. So firms might have a very large estate and estate planning inheritance team and they need to produce lots of wills and powers of attorneys and all those sorts of things and probate applications, all of that is actually quite standardised. I don’t think quite frankly that most LPOs want to play in that space. I just think it’s just too hard, I think there’s not enough volume.
DT:

1:03:00

 

 

 

There’s a lot of different tasks that a paralegal workforce at scale can perform and one of the concerns I suppose that we were talking about earlier was about quality and sometimes that’s driven by a perception right or not, that the quality of legal education might be different in an offshore jurisdiction where that service is being provided, but I suppose a different aspect of the quality concern is that as you said many paralegals are law students and may not have a lot of experiencing in legal practise. How do you manage the quality of output as an LPO provider and what can users of LPO services do to manage and check quality on their own end?
PB:

 

 

1:04:00

 

 

Sure, whenever I hear this question I always laugh over thinking of that saying ‘you don’t know what you don’t know’ which is very often true I think with junior lawyers and paralegals once they start to get a bit of confidence in what they do. Having said that though, we’re a big believer in the ability of our Australian law schools to produce high quality students, excellent attention to detail, excellent education. And then we just design processes with clients that have checks and balances along the way. So not every service we provide for clients requires the same level of oversight and that’s, we give that decision to clients. So for big projects for example we’ll do spot testing so we’ll be ‘OK, we’ve reviewed 100 contracts let’s pull out 10, let’s see of those 10 where’s the quality levels at?’ And then if we start to identify problems we look for patterns and we need to do retraining, we do re-training, if we need to re assign people to different functions we do that as well. We try to manage the whole process in a best practise method which we adopt.
DT:

 

There’s sort of a quality control on the individual project, but there’s also a quality control in terms of making sure the service on subsequent projects is optimised in terms of the people that you’ve got working on it.
PB:

1:05:00

 

Yeah, absolutely, and we actually involve the client particularly early on in an engagement in that process too. So we say to them up front, you know, we will go and build this process for you, we’ll run the pilot, here’s the work and now give us your feedback. Because I think LPO had a problem initially where people think they could just sort of throw it over the fence and expect to come back perfectly.
DT:

 

Now some of our listeners, because our listeners are lawyers, we’re always thinking the worst. When that quality control process doesn’t go right and there’s a professional indemnity liability issue there, how does liability work with legal process outsourcing?
PB:

1:06:00

So regardless of whatever the contract will say, it always falls on the lawyer. If you’re advising a client on whatever the matter is you will be in the eyes of the law you will be liable to the client. However, if you’ve engaged an LPO to assist on that, the liabilities are going to vary depending on who the LPO is, what the contract says. So 90-95% of LPOs are not law firms and they will tell you upfront they don’t provide legal advice and they usually in their contracts will seek to limit as much liability as possible.
DT:

 

Returning to the benefits of an LPO, it’s estimated that utilisation of LPO particularly in the litigation context, can save costs on particular tasks by up to 70%, I think the task that statistic is talking about in particular is the discovery process. Why do you think LPO isn’t used more frequently in litigation?
PB:

1:07:00

 

Actually I think it’s used quite extensively but predominantly by the US in the UK firms. I think many of the global firms that are operating in Australia have now gone down the path of setting up their own local centres abroad as well. So you mentioned earlier many firms will have centres in Manila, or Belfast, or Glasgow, or Newcastle in the UK, and they do rely on those centres to help them with the discovery. I just think in Australia maybe the pressure to go down that path isn’t there. There’s still a feeling of trepidation amongst partners of these firms that, you know, how can I trust an offshore provider? It’s not ingrained in now our profession here yet that it’s perfectly fine to outsource things.
DT:

 

1:08:00

That’s true there’s no impetus, financial impetus, to change. It’s funny that you mention that it’s not really yet a part of our culture. I remember being at a conference a few years ago and it was about automation in the law, but I think the comment applies equally to legal process outsourcing, they were talking about an automated discovery tool that had a 90% accuracy rate in terms of picking up documents that fell within the categories described. Someone at the conference said well 90% sounds good but that’s a 10% margin of error and 10% could be a lot, that’s 1 in 10 documents that it misses. And the person who was talking about the product was saying well that’s true but your own graduates have a higher margin of error than 10% so it’s not about whether the process produces perfect results, no process produces perfect results, it’s whether it produces comparable, better results than the alternative.
PB:

1:09:00

That’s right, I think in that the discovery space, I think the single biggest thing has been the advent of the technology, the review tools like RingTail and Relativity have almost reduced the amount of work that’s involved by 90% and then so all of a sudden instead of reviewing 800,000 documents you’re only going to review 80,000, and then if your second pass review is your graduates, the two combinations together got to be a far superior outcome for the client.
DT:We’ve talked a bit today about the way legal process outsourcing has changed over the past ten years or so since it first became popular in the wake of the GFC, and how technology has refined the product that legal process outsourcing offers.  Where do you think LPO is going to be in 5 or even 10 years time as it continues to grow both in popularity and in sophistication?
PB:

1:10:00

 

 

 

 

 

1:11:00

 

 

 

If only I had a crystal ball! I think it’s a really interesting question, I like to ponder from time to time. I’m a big believer in that the LPO will transition to having best in breed proprietary technology and I think it takes money to produce that. So I would expect that a lot of the large global players will start to merge and we’ll see a consolidation in the market. I think it doesn’t matter what market you look at the moment whether it’s social media or industrials where it just emerges and then consolidation to having one or two really big players just seems to be the way that it goes. And I don’t think the LPO market will be any different. I could see that in the LPO space, and probably the legal managed services space in particular, I think the accounting firms will start to gobble a little bit of that up too which will be interesting to see. A little bit off topic I guess but I think you know all of it though I think the trojan horses are the information companies. They are collecting data on a scale I don’t think we can comprehend. Not to mention all the legislations obviously documented, all the case law and the outcomes of the cases is documented, but then they run our practise management systems for the most part and document management systems all the data that would be throwing off into work flows and how to run matters. One day when the AI is ready and they need someone to feed the data to it, that’ll be the source that they tap. And so I could see you know the Thomson Reuters or the Lexus Nexus of the world coming back into LPO, partnering with a technology partner perhaps, and that’s why I think there’ll be some sort of a merger there.
DT:

1:12:00

 

 

 

 

 

 

1:13:00

 

 

 

 

 

 

 

1:14:00

Yeah I think that’s definitely a prediction I’ve heard before, that in so many aspects of what we do, the legal services market is moving away from bespoke work billed by the hour every single time, to converting our knowledge into a product that can be sold and I think if we don’t do it, then as you say its software companies that will come and eat our lunch and do it for us. But I think when we talk about this kind of future prediction it’s easy to get a bit gloomy and pessimistic about it. I think the mindset that I try to have at least is that these are all opportunities, even for smaller firms, to use tools like legal process outsourcing to operate at a scale that they couldn’t before and compete and win work that otherwise wouldn’t have been available to them. I think if you view those kinds of technological changes and organisational innovation changes like LPO as opportunities, then there’s a lot more work out there for you.

Paul, thanks so much for joining us on Hearsay.

You’ve been listening to Hearsay The Legal Podcast. I’d like to thank our guests Jonathan Prideaux from Kordamentha and Paul Bartholomew from Unison Outsourcing for joining us on the show. If you liked this episode about innovation and process, why not listen to my interview with Talitha Fishburn about digital advocacy. Or, for something different, try my interview with Leona Bennett and Shelby Timmons about another innovation in legal practice: collaborative work. If you’re an Australian legal practitioner, you can claim 1 continuing professional development point for listening to this episode. Whether or not something entitles you to claim a CPD unit is self-assessed, but we suggest that this episode constitutes an activity in the practice, management and business skills field. If you’ve claimed 5 CPD points from our audio content already this CPD year, you may need to access our multimedia content to claim further points. Visit htlp.com.au for more information on claiming and tracking your points on our platform. The Hearsay team is Tim Edmeades, our audio producer, Kirti Kumar who does all our episode summaries and quizzes, Araceli Robledo who’s responsible for all things marketing, and me, your host, David Turner. The Hearsay producer is Nicola Cosgrove. Hearsay is a project supported by Assured Legal Solutions, making complex simple. You can find all of our episode summaries, transcripts, quizzes and more at htlp.com.au. That’s HTLP for Hearsay The Legal Podcast.com.au. Thanks for listening and we’ll see you again next time.