Gary Palmer-Power: Hello. Hello, everyone, and a very warm welcome wherever in the world you are.
Thank you for joining us today, and we're certainly glad to have you here. My name is Gary Palmer-Power, and I am the Global Head of Account Management here at Acolad Life Sciences, and I'll be the moderator for the next sixty minutes or so.
So here's a situation that I think most of us will recognize. Leadership teams across the world want AI, perhaps everywhere, but that also includes labeling. AI promises faster turnaround times, reduced costs, more languages, and many other benefits. But what's often missing from this request is how and why. But also an interesting question: what and how are others doing it?
As we know, labeling is regulated content. A version error or terminology slip isn't just a typo. It can become an audit problem, a rejected submission, a risk to patient safety. So how do we deliver the benefits that we need to achieve, but do so without taking on the risk? And is AI always the right answer? Today, this is what we're going to be addressing.
Today is not about a vendor pitch, and it's also not generic "AI changes everything" kind of talk. What we're hoping to achieve is a practical conversation about what actually works in regulated labeling and what doesn't. And just as important, how do you know where you stand today and where to start?
To know this, we'll be looking at a couple of things, such as benchmarking: comparing your approach against that of your peers, perhaps even other industries, and seeing what your partners are doing right around the world.
To explore this, I am joined by two panelists who have genuinely tackled this question. Laurie Johnson, who is the Director of Global Product Labeling at Beckman Coulter Diagnostics, faced a major internal problem, and instead of guessing whether she was on the right track, she went out and benchmarked with peers and with vendors. This is one of the spines of today's session.
I'm also joined by Richard Korn, who has spent more than twenty years in medical device labeling, including a stint of seven and a half years leading technical communications at Medtronic. He's seen a lot of transitions up close: what scales, what stalls, and why.
So, Laurie and Richard, thank you very much for being here today.
What we're going to cover first: we'll open the session by starting with a couple of polls so that we can get to see where the audience are at on your journey. Laurie will then move into talking to us about her benchmarking story, the trigger behind it, what she did, and what she learned. And then Laurie and Richard will get into the deep dive a little bit more in terms of what works, what doesn't, and no doubt we'll touch on important things such as version control, terminology, traceability, audit defensibility, and such.
But what we really want you to be able to leave with today are two things. Firstly, a sense of where to start. Do you start with low volume, low risk content, and build up towards high risk, high visibility content once you've earned confidence in the tools and technologies? And secondly, the idea of strategic partners: how can you work with strategic partners to learn what others are doing in the industry, what maybe some of your competitors are doing, and how can you perhaps answer the simple question of am I doing the right thing, or could I be doing it better? What are others doing?
A couple of housekeeping rules before we get started. These sessions are hopefully built around you and what you can take away from it. So we will start with the live polls. Please keep an eye out for when those launch shortly, and vote. There's no right or wrong answers. We just want to see where the audience is at, and that will tailor how this conversation develops.
But secondly, if you have any questions, please feel free to drop them in at any point, and we'll address them either throughout the webinar or towards the end. And yes, we're also recording this session just so that you're aware, so you can also play it back afterwards.
So with that, I'd like to open up the first poll question. Please, Christina. So, where is your organization currently at in terms of labeling today? Are you piloting? Are you in production? Where are you up to? I'll just give you a couple of moments to answer that question.
Okay, so at the moment we have "exploring." Anybody else? Okay, exploring is actually a really important topic that we will be discussing today. It's one of the key things that you could potentially also benefit from benchmarking with. So hold tight, and I'm sure that there will be some interesting information and takeaways to come from Laurie's story and also what Richard has to share.
Should we move on to the second poll? Okay, so the second poll: what is driving the pressure? What is driving the pressure to adopt AI or to change what you're currently doing? Is that centered around cost, headcount, languages, speed, or leadership? I'll just give a couple of minutes to see if we get any more answers for that.
Speed and cost, very good. I think speed and cost will be... okay, we're getting a couple more votes. Cost has just increased. Cost is a driving force that a lot of us are experiencing at the moment, and in regulatory labeling and regulatory affairs in general, it's a very, very important one to consider. Of course, if anything goes wrong, if you're not set up successfully, then that can be very costly, so there's a balance there, which again we will discuss later on in this webinar.
Okay, so then thirdly, let's move on to the final survey for the time being, which is how confident are you that you can tell which content is safe to move into an AI workflow and which isn't? So you're confident, you have some idea, or you're completely new to this and have no current ideas? Again, I'll just give a couple of minutes for people to answer that question.
Okay, so some idea. Okay, thank you for sharing. We'll come back to those during the webinar as well. But in the meantime, if I now hand over to you, Laurie, if you could please share your story and the decision to benchmark before investing further. Thank you.
Laurie Johnson: Sure. So my team has been exploring new tooling and adding AI to our content management and translation workflows for about the last two and a half years. We spent a lot of time doing research, investigation, looking around the world, and reaching out to other vendor partners, learning what the landscape looks like and what our tool options were. We had a pretty legacy-heavy environment, and we knew we needed to upgrade, and we wanted to make sure that we were future-proofing ourselves.
And so we'd landed on a new workflow for our team, and we'd gone through our RFI, our RFP. We'd set up an initial schedule for implementation and really driven that to a fairly good point where we had our pilot in line of sight and had an opportunity to discuss expanding that out, and whether what we were looking at might be of use to some other internal customers and business units. But we were really asked whether what we were doing was kind of gold standard, was what the rest of the company might be interested in, and how do we know?
And so, how did we know? How do we talk about our content workflows, our localization workflows, ability to improve? It's the same thing that the audience was talking about. How do we demonstrate that our ability to improve cost control and improve speed was not only good enough, but really best in class?
And so, we thankfully had made some strategic hires. We felt like we had good data points internally. And I'd had conversations with some folks that I knew from other industries, but we felt like we really needed to know, within med device and diagnostics specifically, what our peers might be doing. So this is the point when I really did start having conversations with our vendors, reaching out to see what else was going on in the industry, to the extent that our partners could share what they could about what else was going on.
And I looked through my contacts and who I knew on LinkedIn who might be doing a similar type of work. And I did happen to reach out to Richard, who I had worked with in the past. He and I had worked together at Saint Jude, formerly St. Jude, now Abbott Diagnostics, a handful of years ago, and had a conversation about what his experiences were with localization, what he'd worked with, what his workflows looked like, and where points of comparison were, whether what we were looking at doing was similar or dissimilar, and what I could basically take back to leadership and say, yes, these are the same things that we're talking about doing, or not.
Gary Palmer-Power: This is really interesting. So you had run RFIs, you'd run RFPs, you'd gone to some industry events. I think you mentioned [unclear] just now. And there was still, what I'm hearing, some level of uncertainty as to the right way forward. So were you looking for confirmation of an idea that you had? Was there a particular question that you found you couldn't answer despite having run all of those things that you still didn't feel comfortable answering? Or was there a particular trigger that made you think we need to maybe approach this from a different angle?
Laurie Johnson: Really, the question, the pressure that we were getting that we're trying to answer with the highest level of confidence was, what are our peers in med device and diagnostics doing? So it's not necessarily what is everybody doing, but what are other companies just like us doing? I think a lot of times it's easier to see what tech companies are doing, or what's going on generically, what a localization workflow might look like, or what high-level best practices might look like for content management. But sometimes what a large consumer electronics company might be doing is not going to be appropriate for a company that's operating in a highly regulated environment.
Those workflows are going to be very speed and cost centric. Not that we're not, but they're not going to have as many gates as we do, and the pressures are going to be quite different. So, we were really being asked to benchmark against peers.
Gary Palmer-Power: Understandable. So out of interest, was this also your first time benchmarking? Were you maybe a little bit nervous about going out and asking others what are you doing, maybe what are you doing better than me? Or did you not have any concerns? Because I could imagine that for some people that would be quite a vulnerable position to come from. How did you approach it, and what tips and tricks could you perhaps share with some of the members of the audience?
Laurie Johnson: There's definitely, I think, a level of vulnerability, and wanting to be open to getting feedback. But I think for all of us, there are things that we're probably doing that are better than some of our peers, and there are certainly learning opportunities for all of us. We can all improve on something that we're doing. I think within med device and diagnostics we all operate within the spirit of continuous improvement.
Gary Palmer-Power: No, no. And before we maybe switch over to Richard in a moment, just one final quick question on this from me as well. When you went out there and started to ask those questions and benchmarked, did you meet with any surprising comebacks? Did you learn something that you perhaps weren't expecting, or did it reveal something you weren't expecting?
Laurie Johnson: I think it helped to confirm that we were definitely headed in the right direction. And I think one of the other things it helped to confirm was the importance of making sure that your underlying content is solid before you start looking at ways to improve, add speed, and add AI into your localization workflows. If your content isn't clean, isn't solid, isn't consistent, then there's going to be challenges downstream.
Gary Palmer-Power: Thank you, and we will come back to that point in a moment as well. Richard, if I could bring you into this one. So what do you do when a colleague such as Laurie comes to you and is asking for input and advice?
Richard Korn: It's wonderful to have the collegial environment and to share your notes and also to reconnect with former colleagues and friends from other organizations where you share a common history. I think that's kind of the dynamic within med device, life sciences, med tech as a whole. At the end of the day we're really serving our patients and clinicians, and if we can help one another get to a better place, then that's really a win-win. So to that end, I think that's not unique to this conversation, it's just the way that our industry operates.
I would say, just for a moment tying into what Laurie was alluding to, if I'm putting on a quality hat, I really want to look at content from the perspective of risk. What will lower the risk? And then that kind of helps to guide the conversation. That's looking at other industries, like tech for instance: yeah, there are quality controls and there's GDPR and all these other elements that are part of our industry as well, and they are certainly guardrails, but we have a special responsibility to ensure the safety of the individuals who are using our products.
If I'm looking at content specifically, I would have to look at the maturity of the organization, the labeling organization in particular: how much content is already structured in some format, and maybe in a component content management system. If that's in place already, then I would look at what products are more stable, sustaining products where the content may not shift as much. When you talk about NPI, new products, and maybe content that's not in a centralized database, that's when it becomes more challenging.
I think AI accelerates everything, it automates what we do, but the core tenets of what we were working on, it's not any different really from all those years that many of us have been working on, moving content into a centralized repository, using component content management or whatever. At the end of this whole process, though, the human component is the critical piece. We'll talk more about that, I know, in our discussions, but the human component is key, because as much as you might use AI to automate the generation of content, you need at least one pair of eyes, if not more, looking at what's been generated in context and with the history of how these products interact and maybe other components, safety related for sure, and that pertains to the localization aspects as well.
Gary Palmer-Power: So that actually ties back to what you were saying at the start, that you want to look at content from a risk perspective. So in kind of plain language, when we're looking at labeling, when we're looking at regulatory affairs in general, and if we are considering AI, where'd you begin? What is the safest content, what would you consider to be suitable for AI, versus what is the content that you would recommend people never go near when it comes to AI?
Richard Korn: Well, AI is evolving so quickly, I think it's difficult to say that there's something you would never have AI go near. But I'd go back to being consistent, like Laurie was mentioning: ensuring consistency in your content, that's really what it comes down to. You don't want AI to generate new phrases that are maybe not part of a glossary that you've developed or that you've vetted.
I think it's not unlike if you're migrating content into a CCMS: same tenets, you want your metadata in place, you want to have content mapping, and then you can be more assured that the content that's being generated by the AI, if it's not identical to what one of your writers might have generated, or that a human being may have generated, it's closer, and then you can leverage the benefit of AI, because AI can search in an instant all of the different documents and really put together the pieces. It's really, for me, the missing piece from what we had with component content management. It ties in all of that.
So the short answer is maybe documents that are internal first, and your organization may have a GenAI that's specific to the products that you use, and there may be some ad hoc usage. But also I would look at the integration of AI into the tools that you already use, and there may be a seamless way to integrate AI without intentionally saying "I'm using AI right now." A lot of the technology already has this embedded, but yeah, that's my short answer.
Gary Palmer-Power: Thank you, Richard. Laurie, I have in our notes from our previous discussions the expression "rubbish in, rubbish out," which we've discussed at some length. So I was just wondering, from your perspective and from the exercise that you ran and the benchmarking, when we're talking about AI, how much do you think the problem comes down to AI readiness versus what Richard was just talking about, which is really making sure that you have this content foundation with structured terminology, CCMS? What's your take on that?
Laurie Johnson: Well, I think, again, as Richard was saying, there's a lot of similarity to being ready to go into content management in general, and being ready to leverage AI effectively. That's making sure that you have clean content and it's well governed. And so in order to be able to leverage AI in an intentional way, you do really want to know the state of your content and make sure that it's prepared for whatever your intent is with AI, whether that's leveraging AI for a productivity purpose or using it in a translation workflow in some manner.
So, as you said, rubbish in, rubbish out, you do need to make sure that it's in a known good state ideally before it becomes part of a workflow that's leveraging AI. Having good terminology, having structure, as Richard said, having metadata defined and in place, and governance, so that you know what you're doing on a go-forward basis. We've had a lot of discussions within my team about types of documents and content, and the level of risk and sensitivity for those documents, and whether it's appropriate to go through an AI-leveraging workflow or not.
Gary Palmer-Power: Thank you. So it's a really good point, and maybe a bit of a provocative question: in the polls, a number of people said that cost is obviously a driving force of introducing AI, but if we're talking about really investing in that foundation, that structure, whether that be CCMS, language assets, so your translation memories and glossaries, they can be very costly things to invest in. So I'm wondering, if anyone on the call identifies a weak foundation that could therefore amplify errors that AI could make, is there maybe a real case for some people to repurpose the budget for AI into cleaning up that terminology, putting in place the solid foundations as an initial step, and then maybe looking at AI in the future? And if so, how would you respond to pressure from a leadership team to get AI now?
Richard Korn: I don't know, Laurie, do you want to give that one a go? I think, yeah, not unlike many other projects that involve legacy content in particular, or content that's being taken from one format to another, say you're in InDesign or FrameMaker and moving it into an XML-based platform, yes, there is usually a project at the beginning.
I think one of the key areas that we need to consider is what products will still be in place. First of all, is it worth touching that content? If it's stable and the products are actually going to be phased out, then you can cut out a bunch of content. But it's not as easy as that, and maybe that's more of a CCMS approach, because sometimes that content may be relevant and the AI search engine, or what's integrated into your tools, may need to make use of it. But that's one consideration.
I would say repositioning the team, and at the same time looking at your content and having a big project to assess how consistent the content is: sure, that's an investment, and I think that's a valuable investment, but at the same time the team needs to be retrained, placed into new roles that are focused on AI and the quality aspects, because that's what makes it unique, I think, to our industry.
There is reticence, this sort of reluctance to adopt new technology until it's been out there for a while. This is the departure, I think, from that model. A lot of med tech companies are embracing AI in a big way, they're seeing the value of that, and the value of repositioning those team members who were formerly on maybe operations or labeling tech comm teams, localization teams, looking at them as becoming subject matter experts in AI.
And the other thing I would say is if your costs are going up initially, okay, there's an investment that needs to be made. But over time there should be a model that ensures you're leveraging the goodness of the AI, and that you don't have, say, a full team of writers or localization professionals in the same way, that they're facilitating the work that can be automated and doing the work only human beings can do. And that's really the big challenge. That's where I think most organizations are going to face a big challenge.
Gary Palmer-Power: Thank you, Rich. I think that brings us nicely onto — we have one more poll, which I think has already been opened, but just to draw the audience's attention to it. One final question: when it comes to adopting new technologies or AI changing workflow processes, whatever that may be, which of these worry you the most? Is it version control? Is it terminology drift, which Richard and Laurie were just talking about? Is it traceability, or maybe the ability to defend during audits?
Okay, terminology drift has just popped up, as well as audit defense. We also, just to go back whilst we're waiting for maybe some more answers to come in, I just wanted to take it back to what you were talking about then, Richard, in terms of repurposing resources and still keeping that human in the loop being a non-negotiable, particularly with higher risk content. So what does meaningful human control actually look like in this process? Where would it take place? Would you still keep it in, say, pre-submission? Is it during audits? Is it throughout the whole process? What is your experience?
Richard Korn: Well, sure, quality is useful at all stages in the life cycle of a piece of content. But upstream, if you can address those issues upstream, eliminating inconsistencies in content, you also eliminate the inconsistencies in the translations that would follow. That's where I would say you can gain the most value. And also, because AI is learning, if you allow AI to learn from clean content, it's not as muddled, it gives you more precise responses. The AI can learn as well, in the same way that human beings can. I mean, not at that level, but it's got the capacity to weed out some of the rubbish or the garbage, but why not help it along?
And if you've got a team that's highly familiar with your products and the content, it may not just be within the traditional labeling organizations, it could be groups that do your in-country review or other teams that could add some value to your larger data-vetting project. I think that would be my take: the upstream work is where you're going to gain the most value in the long run.
Gary Palmer-Power: Thank you. And maybe just for awareness, we probably have about ten minutes left or so before we'll move over to questions and answers. So I'd just like to bring the conversation back to the benchmarking piece that we were discussing before. Laurie, from the benchmarking piece, we mentioned the ability to go and speak to your colleagues and peers, such as Richard. There's also then the possibility to speak to your vendors, as you mentioned earlier. I'm just interested to see: did you get complementary information? Did you get conflicting information? How did the benchmark provide that level of clarity in terms of next steps for you?
Laurie Johnson: So in general, we got fairly complementary information. We were looking for kind of the environment, the workflow that we're standing up, is generally kind of gold standard, what the industry is using. That's fairly commonly what we found. We found a little bit of a different tool set, but the tools themselves were pretty much what we were using. So the specifics varied somewhat, which is what you would expect, whether it's which vendor, which TMS you're using, that type of thing, whether something is internally hosted versus externally supported by a vendor, that type of detail. But in general, the workflow that we were looking at was fairly common.
Okay, and that's largely what we were looking to validate, and we did largely validate that across our industry. That was a good finding. And it was also really good to start having more of those conversations, and to start making more of those connections, or in some cases, like with Richard, reconnecting. So again, it's something that I think, for myself, I could be doing more of, having more of those points of connection, more of those conversations.
Gary Palmer-Power: No, brilliant. And so it's good that it was largely a validation or a confirmation of what you were doing, but just interested to hear, was there anything that came out of it that you've implemented that perhaps you wouldn't have done without benchmarking, or has it helped you somehow internally to align different teams in a way that wasn't the case prior to benchmarking?
Laurie Johnson: Trying to think... I think, again, by and large, we got a lot of validation. I think we ended up with a much larger dataset than we got through our RFI and RFP process, which was quite interesting, because we did quite extensive research beforehand, but the industry and the playing field has changed quite rapidly. So, again, it was interesting research. We found out a lot about additional tool sets and additional vendors we didn't have exposure to before.
Gary Palmer-Power: Yeah, good additional research for sure. I just maybe want to mention one of the elephants in the room, perhaps: of course, I come from the vendor side of things. We work with also maybe some of your competitors, and we can see maybe what they're doing versus what you're doing. So how did you find approaching the vendors? Was it a good vantage point for you? Were they also open to talking about what you're doing versus the vendors? Were they as open and forthcoming as maybe Richard was as a peer? Did you see any advantage of consulting them?
Laurie Johnson: I think by and large, our vendors that we've worked with have been as open as they're able to be, given the NDAs and agreements that they have with various levels of clients and partners. So I know some of our vendors are able to disclose more than others. They've shared what they can, and I feel like that's been very helpful. I think by and large we have pretty positive relationships with all of our vendors. We've had good discussions about where we sit with regard to our tool sets, our productivity, and where we can make improvements. And I think they operate fairly transparently, so yeah, all positives.
Gary Palmer-Power: Great. And I would encourage, for folks who have not, and I know I've had some conversations with some of my peers who haven't made that next-level connection with their vendor partners, to really do that.
Laurie Johnson: It's been truly helpful in some of our transformation and helping us refine our processes internally too.
Gary Palmer-Power: Great, thank you. And maybe one last question to Richard before we look at what the audience is asking. So from our notes, I also have the comment that companies can often diverge internally, business unit to business unit, in terms of their approach to AI readiness, what they're willing to do and not to do, perhaps also related to risk appetite. From your experience, have you ever used benchmarking to expose those differences, or have you ever used benchmarking yourself in the same way that Laurie has?
Richard Korn: I've used benchmarking many times in different scenarios. I think there's quite a bit you can learn from your peers, and especially in this industry, it's more reassurance that what you're doing is on the same path, and it tends to work out that way, because with the state of technology, a lot of groups are looking at the same types of improvements and enhancements within the same time frame. It's kind of uncanny how that works. I would say with suppliers, for sure, for localization, for content management, some quality-related vendors as well. It gives some insight very quickly. So as you mentioned with the work that you do, you're going to see the landscape of all these different labeling, tech comm teams, and that can provide some insight as well without doing a ton of extra benchmarking.
But I did want to delve a little bit more into the piece about divergence within an organization. Some companies are not very large, so this wouldn't apply to them, but the ones that are larger, or have had recent acquisitions, it's important to keep on track with the larger initiatives that you've undertaken. If it's AI driven and you're trying to incorporate that into your labeling, that doesn't mean that those newly acquired organizations would necessarily use that right away, but there is value in having those larger projects that could then maybe apply to new organizations at a later date. So I would encourage that.
The other thing is that wherever you are in your AI journey, and it applies to all facets of labeling, but certainly for AI, I would jump in and experiment, and it's very easy to do. You could take low-risk documents and work with them, maybe non-regulated files or internal documents, documents that you just use for your team, and use those as case studies: what worked, what didn't work. Encourage your team to experiment. Go into Gemini and Claude and ChatGPT and all these different AI options that your organization may allow you to look at, because I think that will spark ideas that are more relevant to your specific scenario and situation.
The other element of this is, if you don't have infrastructure, say maybe you didn't work on a CCMS in a meaningful way, or maybe not at all, maybe your content is kind of in disparate locations and not centralized, this is an opportunity to create some consistency and leapfrog ahead, and not go through all the pain that a lot of other organizations have gone through. At the same time, if you've developed a robust CCMS, yes, you can leverage that, you can leverage what you've done with content mapping and all the different facets of CCMS to guide you on the AI journey.
But on the flip side, it's okay, I think sometimes people are a little bit... I don't know if afraid is the right word, but it's more about, do I want to invest my time in this technology right now because I've got all these deadlines and all this work to do? So I would say, at least part of your team, and the teams that are going to be repurposed and repositioned and retrained, that's an opportunity for them to give some input and say, okay, yeah, I think this would really work well for us. And certainly working with suppliers and vendors that you're already working with, they can help you as well with what they know from their journeys.
Gary Palmer-Power: That's great, thank you. So for the last five minutes or so, we'll answer some questions from the audience. I can see a couple have come through before we wrap up. So the first question, and I think this touches back on what you were mentioning a little bit earlier, Richard: when an auditor asks how AI has touched a piece of labeling, what do you actually need to be able to show them? Have either of you gone through that process at the moment, and feel comfortable answering that?
Richard Korn: Well, I have not gone through that specific question with an auditor, but one thing to keep in mind is, at least the way things are set up today, the content is released through a PLM, and then that would be the source of truth for your audit. So maybe it's a PDF file, it might be within your PLM or on a website, but I would say that if you are trying to prove that the quality is in place, that you are using the same quality steps, that should be incorporated into your quality system before you start releasing AI content. So you'd have a new paradigm to address AI-generated content. That's my take. Maybe Laurie has some other insights here.
Laurie Johnson: Yeah, I was going to say pretty much the same thing: you have a mechanism to ensure quality output, so you would have a way to demonstrate that it's been appropriately reviewed, approved, and released, and it meets your quality standards. And that's what I would show.
Gary Palmer-Power: Great, thank you both. That actually leads us quite nicely onto maybe the next question, which is: how do I make the case to the leadership team for a crawl, walk, run approach, when they're expecting maybe big savings this quarter? Laurie, maybe should we start with you? I don't know whether that comes back to your story and your experience.
Laurie Johnson: So I would want to know first what type of savings you're looking for this quarter. You'd want to define what savings you're trying to deliver, so that you know what goal you're trying to meet. Are you looking for cost savings or time savings? And then what can you deliver in-quarter? So, if it's cost savings, then what are some AI-based tools or processes, productivity enhancements, that you can use AI for that could help get you closer to that goal, that can help move the needle? It doesn't have to be huge necessarily, or a transformation. You could set that crawl goal, and then the walk for the next quarter or the next year, and then run is further down the road.
That's really what I would think about doing, and it's sort of what my team has been working towards, where we have really huge aspirational goals, but we have smaller incremental goals. We talked about AI-based workflows, or where we can put AI into tooling, but you can also think about how AI can help with your productivity, like could we have some writers on my team experimenting with some tools as a peer review partner, so it's your first-pass peer review, and could that help with your productivity in generating a draft? So, what are some creative, out-of-the-box ways that it could help with achieving your delivery metrics or something like that.
Gary Palmer-Power: Brilliant, thank you. We have one last question before we can move on to closing, and I think this maybe touches back on what you were mentioning before, Richard: we're a small team and leadership wants to see AI this year, if I can only fix one thing first, the content foundation or the tooling, what would actually move the needle? What would your recommendation be?
Richard Korn: I would look at the content first, and the tooling is something that could come later. That's my take. And Laurie, do you feel the same way?
Laurie Johnson: I do. I mean, your tooling is probably, especially if you're looking at ground-up tooling, that's RFI, RFP, it's going to be a longer term engagement, and potentially a lot of resource allocation as well, time, money, people. Your content you can address more directly.
Richard Korn: Yeah, and there are groups also that could help you get a jumpstart with that, that do this type of thing, especially in the content management world. But I also think it's good to take an assessment of your data, how much content do you have, and then you can start figuring out what's realistic to do. Maybe it's a subset of that, which applies to the previous question as well. You can make a case for something smaller, maybe a pilot, or setting aside maybe one team member to focus on this if you have a really small team, but just looking at what's out there, that can guide you, inform you quite a bit. A lot of the time it becomes this amorphous thing if you don't know what you're working with, and once you do, it becomes more about breaking it down, and you can use methodologies that have been proven within our industry as well, like a Six Sigma Lean approach or something like that.
Gary Palmer-Power: No, brilliant, thank you. So we have just a couple of minutes left. Maybe one closing statement from each of you: if a member of the audience were to reach out to you tomorrow to ask for some advice or some benchmarking, what is, maybe in one or two sentences, the thing that you would recommend everyone here should go and check or do, starting maybe tomorrow, in their organization? Let's start with Richard.
Richard Korn: Okay, sure. I would say strategy. Look at your strategy, treat your organization, your labeling organization, as it is a part of the critical path, and think about what is your strategy for content, what is your strategy for your team dynamic, your org, and what's your vendor strategy. If you lay out those three, you'll have a path that you can follow, and it might be a multi-year path, but that's also a path that you can start selling to leadership. Although, in my experience, although labeling is critical, it's regulated, it's part of the critical path, so you can't sell product without it, there's still a lot of selling that needs to take place internally, that internal selling of the services. So that's my recommendation.
Gary Palmer-Power: Thank you, Richard. And Laurie?
Laurie Johnson: You touched at the very end on something that we've been working on very diligently within my group recently, which is thinking about product labeling as not only critical for our registrations and for getting us into market, and required for all of our products, but it is also sales enablement. Market expansion and sales enablement is a core function of our product labeling, so it really is part of what we do. And I think that's part of our job and part of our strategy, as we build out all of this: our content strategy, our AI strategy too, is to think about how we enable product sales and market expansion as well.
Gary Palmer-Power: Brilliant, thank you very much, Laurie. So, to the audience, hopefully you're leaving today with some clear direction on potential next steps to take, rather than a specific product or a specific solution. But to go back to what Laurie and Richard were both saying, maybe more strategy. And when it comes to AI, when it comes to content, as we've said throughout today's webinar, start where the risk is low, build some proof, build some confidence, and then climb towards the higher risk content, rather than judging our position in a vacuum. And then, of course, benchmark. Ask the people you see across the organizations, across the industry, peers, and also don't forget about your strategic partners as well. From our point of view, we work exclusively with life science teams, we run exactly this kind of benchmarking conversation, so feel free to reach out to myself as well if you're interested in seeing what we could do from the vendor perspective.
But finally, Laurie and Richard, thank you very much for participating today, sharing your knowledge and your experience. It has been invaluable to the audience. And to the audience, thank you again for spending the last hour with us. I hope you found it useful. Thank you, everybody, and enjoy the rest of your day.
Laurie Johnson: Thank you.
Richard Korn: Thanks very much.
