Insights View Recording: Implementing Modern CPQ Solutions to Accelerate Sales Processes

View Recording: Implementing Modern CPQ Solutions to Accelerate Sales Processes

In today’s competitive business landscape, streamlining sales processes is essential for success. Join us as we dive into the technical intricacies of developing a modern Configure, Price, Quote (CPQ) solution and how to streamline development, maintenance, and deployments to help the business revolutionize the sales process. 

Our team will cover critical elements of CPQ system implementation, from architecture to deployment strategies while covering hypothetical technical scenarios that may affect the trajectory of the final implementation.  

Key topics to be covered include: 

  • Identifying traditional sales process bottlenecks and importance of efficiency during the quote process. 
  • Architectural deep dive into some of the possible system implementations in Azure. 
  • Technical considerations for ensuring proper core system functionality. 
  • Importance of configuration driven CPQ systems. 
  • How AI can improve CPQ system accuracy and leveraging your system to drive achieving an AI powered CPQ system 
  • Strategic integration approaches for fostering seamless interoperability between CPQ systems and core business applications. 
  • Seamless implementation of CI/CD pipelines in development environments using GitHub Actions. 
  • Using Terraform to ensure a consistent and easy to maintain system infrastructure. 

Whether you’re an associate developer interested in end to end system development or a development manager seeking a path to modernize your company’s technology stack, this webinar aims to draw a path toward an AI capable system focused on solving problems and alleviating headaches – not inducing them. Join us to gain valuable insights, practical tips, and actionable strategies for implementing modern CPQ solutions and transforming your sales operations for success. 

Transcription Collapsed

Good morning, everybody. So welcome to our webinar that’s gonna discuss and in more detail cover implementing modern coding solutions to accelerate sales processes. Uh, we’ll get to intros in a little bit, but before we continue. Let’s talk a little bit. We’re going to be discussing 1st and effectively. This is going to be more technical deep dive into an architecture of a system that or a system that concurrency has implemented in the past. We continue to build upon this systems like this and also. Adapt uh to to different scenarios for for our clients and we’ve done this multiple times over the years and we felt that it was a good idea to start sharing. Uh, some of this this newfound knowledge and and and expose some of these tips for success and get and get going in that area. I’ve been without further ado, we’ll we’ll get to our presenters so we can start with Brian. If you wanna go ahead and introduce yourself. Brian Haydin 1:22 Hey everybody. Brian Hayden I’m a solution architect. They concurrency and I’m really excited to have this conversation with everybody. Mac Krawiec 1:32 Nice. Thank you, Brian. And then there’s myself. Uh, Matt Kravitz. I’m a senior software developer here at concurrency. I’ve taken part in building numerous. Implementations that pertain specifically to this, and I’m excited to just share a little bit more of that. The ends and outs and some of the more technical gotchas that that we’ve run into in the past. Uhm, so let’s talk a little bit about what we would want in the perfect sales scenario, right? Everybody wants to make money and they wanna make money quick and the best way to do it is, hey, a client reaches out to me, asked me, asks for something and an automated process just orders it. And that’s beautiful. But that’s not where we’re at, and in a lot of cases, there’s a lot of steps that have to be taken and make that happen, and even then, not a lot of businesses or business. Ohh concepts apply to that idea and so while this is the North Star, understanding that there is a lot of companies that are maybe not anywhere near there. We’re gonna be talking about a more progressive, incremental step towards this end goal of a very quick sales process and with little overhead. And one of the thing, one of the things that you’re gonna see throughout this webinar are little tidbits that we’ve heard from our clients that make this even more meaningful to us, but also we feel more meaningful to to you and your businesses now. If your sales cycle is longer than some other industries, right? I’ve I from. From what I’ve heard in the past, you know, manufacturing industries have a very quick sales cycle. They turn around in minutes where this kind of a quote applies. If it’s a little bit different than that, maybe it’s a few months this is. Still applicable to a certain degree, in a sense of when you engage with the client 1st and you do it quickly, that resonates with them. So the sale, the company that takes weeks to respond, even though the sales cycle is expected to be six months, if you get there, get back quickly, that makes them feel valued. So these quotes help you know, drive that point a little bit further. Umm, but effectively, let’s talk about our textbook CPQ process, right you we configure or in some way shape or form design or product or service and and make it known that that’s in some sort of list of data that’s what it is. Brian Haydin 4:22 Thanks. Mac Krawiec 4:26 Then we have a process where we price those items. We have a process where we quoted based on some some inquiry. We draft the proposal, we start negotiating with a, you know, depending on the proposal and then we finalize and we order and each checkpoint in this process that you’re seeing offers in and of itself, the potential to be a bottleneck and the bottleneck most often is is human. Umm. Whether it’s somebody being on PTO or whether it’s somebody being, you know, busy with five other quotes or or six other engagements, the more involvement and the more steps along the way you have, the higher likelihood that there is for a longer drawn out process. And if you think back to that quote a second ago, the first one that to get to respond gets the business, this textbook CPQ process doesn’t really head in the direction that we want to of of. Quick business, turn around. That being said though, keeping in mind the North Star that we just discussed earlier, that’s the North Star for a reason. We wanna get there and and in a lot of ways and a lot of industries, that’s incremental processes that have to take have to occur in order to get there. So one of the things that we’ve implemented for our clients is this streamlined CPQ process, which is effectively taking out and. Leading to a very quick turn around of the quote of the pricing quote system or steps within within the process and effectively going straight from hey, let’s figure out what we’re selling. And then going straight to here is what we can give it. Here’s what we can sell you. How much it costs and what it is exactly, and the beauty of it is that in our in our history with our clients, we’ve literally turned processing the take 1530 minutes in the perfect scenario where there is a sales Rep sitting at a desk looking at their email and waiting to run to take care of a quote. We’ve taken that perfect scenario and turned it into seconds with now no sales Rep involvement. The sales Rep at worst has to hit send and send the quote UM and let’s talk a little bit about before we continue and and get to the more technical fun part. Let’s talk about the importance of of AI in. In this process, in our quoting process, and I only have one word for it, and that that is unequivocal. Uhm, our system, that the system that we’ve built for clients is built around a Python environment that makes use of the use of Azure open AI or, you know, a supplementary LM to accomplish a task. And it’s not always answer open AI, and it’s not always this specific ALM and so on and so forth that really depends on on some of the decisions that we make during the building process or the design process. But effectively what? Well, we’re. We’re we’re we’re at. Is that the the days of rules based engines and you know process maps? If this do that, it was short. It was over shortly after the spotlight of shone onto the artificial intelligence in all business aspects. And that’s what we’ve built around and and we we continue to grow in that aspect and as AI matures and our system matures effectively, the quality of our output does too. And what I what I what I like about this? Visualization is is that it it to me it paints a perfect picture of an AI driven systems capability to provide a quality output. With time and you start small and and and go big and and the speed at which you get there varies depending on the data that you have and how good the data that you have is. We have, you know, clients that have taken have been with us for numerous years that have gotten here. We have clients that have started their journeys recently, but effectively we can take these these years of experience that we have in this area that to allow us to to foster clients to, to getting there in a much better much quicker way. Brian, do you have anything to add any any notable mentions? Brian Haydin 9:31 No, I think you’re doing a really. I mean, I don’t have a ton of ad. You know, I I do see a lot of customers finding a lot of value in that time to quote, you know aspect of it that you talked about and even you know you know the most. The. Yeah, the the 15 minutes to seconds, right? Even if you can get you know that that’s a, that’s a really high volume kind of transaction every you know sort of tail. But you know, these same systems can be used to take something that might take days or hours, which is more common with our customers and reduce that to seconds or minutes. Umm and uh, and the value that automation or the cognitive the Cognizant load that is removed from the salesperson so they can focus on maybe more upselling opportunities? Umm, you know, rather than just the, you know, the transactional quote is also kind of a value add to that ohm and and we’re talking with a lot of customers around that. Mac Krawiec 10:34 Yeah, exactly. So moving on, meaningful data gathering should be at the forefront of any system you implement. That is where it is at the forefront of of this architecture that that we have and the systems that we’ve implemented in this space. But I really think that this is the truth for every single system in today’s world, for no other reason than data strategy should be at the core of a business strategy where we’re at in this today with with data and how it how it operates and the value it provides to a business simply underline the necessity to make sure that the data that we gather is the focal point and very much so in our in our, in our processes and are in approaches to building this systems. That is the central focus. Umm. But with that being said, though, let’s let’s talk a little bit about the architecture and what we can integrate this with. Umm and how it actually really works, but basically I’ve been talking about a system the system effectively what we what we’ve done for our clients is providing a process to ingest requests for quote or or or just inventory checks to effectively turn those and provide and and turn those around in, in seconds. And you might imagine that. You bogus here. You might imagine that a sales representative will receive an email to their inbox and that gets picked up by our system that reads that email and pulls out specific information about the product that they are interested in. It connects with our ERP and then pulls a SKU and price for that and formulates a quote inside of a CRM and sends it over to the client all in seconds, right. A human no longer has to read the email. They no longer have to understand the product. They no longer have to look up the price. They no longer have to create the quote. All of it taking time and all of it being again a potential for a bottleneck. That’ll happen seemingly in seconds, and you can do the same with a phone conversation in today’s world. I mean, everybody’s recording every conversation they have in their business and you can imagine that this very easily extends to your state transcript. And we use artificial intelligence to listen to that and decipher it in the same way that we do with an email. And that also can occur with an external system, right? Any other systems that you know, we start talking about, see there’s potential to to include that. Umm, but let’s focus on one scenario, right the the scenario that we’re gonna focus on is this exchange integration. Let’s let’s integrate this system with our mailboxes and let our salespeople spend the time to Brian’s point, upselling rather than data entry and quoting. So. And we’re going to get back to the the overarching specific points in this conversation. But one of the things that we’ve heard from our client is that it can take a few months to teach sales reps about the products that we sell. And that just underlines the the importance of this system. Once it’s out there, and once it’s trained and AI is performing to do and doing what we needed to do, we effectively have. Made it so a sales representative doesn’t have to understand the INS and outs and every single item on the inventory or if it’s some sort of industry, they don’t have to understand IT industry very intimately to be able to do their job. AI will help and effectively pick up the slack and let those sales representatives be sales representatives rather than specialists in a specific product that they sell. There’s people for that, and sometimes sales representatives are better when they are left to do what they excel at, which is selling. And and that’s why I think that that quote was when I heard it, I remembered it. Right. And there’s a reason I remembered it. And there’s a reason why it’s on the slide. But let’s talk a little bit about the the the as pertains to this architecture diagram. Let’s talk a little bit about the onset of this process. What kicks it off, and you might imagine that in our in our business we have multiple sales representatives and and and those emails that they receive all get funneled to a single shared mailbox and that Shared Mailbox is the focal point and and the and the kickoffs area for our automated process and it allows us to monitor email ingestion. And it’s also easy to to to manage user access to the system. So this shared mailbox is connected directly to the system. It’s the only functionality that it serves, and in exchange admin your your teams can then just at an email to automated forwarding and they now have access to this system and there are a few technical gotchas in this space. We’ve seen issues arise with things like DC where BC is is a blind copy right? So effectively. Umm this approach doesn’t have access to blank copies. That being said, though, there’s other things that you can do to extend it further and make sure that you you you factor that in. And then there’s also email identification. This might not be very clear any right away, but once you start seeing in your processing thousands and thousands of emails, you might find yourself that every 100th every 200th email gets skipped. And that is because exchange offers up conversation IDs and Outlook email IDs. Uh, none of which are necessarily you. They’re gonna be unique between mailboxes as one, but sometimes they’re gonna be unique. Uh. Also, within the context of a conversation in and of itself that poses some issues that we’ve had to parse in the past to make sure that. We are able to identify emails accordingly. Another gotcha here is with the shared mailbox. Sometime the the way that Microsoft clears their cache after removing an automated forwarding process, it might not necessarily remove it for another 24 hours. So there’s a few gotchas in this area, but that is our ingestion process. And some there it doesn’t come without its own share of of concerns. And one of the concerns that when I was talking through this with Brian was, you know, we funnel into this logic app, right, that we’re going to talk about a little bit and. There the concern that we’ve heard in the past was, hey, you know, you’re listening to all of our emails. This process is going through and and and potentially sending every single email in our organization to an artificial intelligence platform. And it doesn’t quite work that way. Uh, the the the design is such that there is a logic application with some entry level rules based filtering to ensure that, hey, what we’re reading or what we’re what we’re scanning is effectively relevant to our topic. So you know it, it may scan, it may look for specific words like quote or some specific terms in the subject or you know it can’t be from this email address and that is a way for us to one solve some privacy concerns, right? There was. We’ve had concerns with HIPAA and so on and so forth that we that we’ve had to take care of, but two also offer a low cost filter for the remainder of the system. Umm. It allows us to to free up some precious resources downstream, umm, and then moving on from the logic application. We have our Azure function and that is the central piece of of this entire conversation. It is what integrates with every other downstream process. It is what it integrates with the logic application. Umm, because of this, we’re able to effectively do this. It it the way that on a more technical note it’s using a more recent days, we’ve upgraded to .net 8 making full use of of you know dependency injection. I I like personally to build with the service repository build pattern. I personally think it’s. It’s a nice clean way to separate your areas of concern. Umm. And this Azure function as I mentioned, because of its centerpiece in integration, you know, makes use of graph API to further integrate with everything that happened before you know with our shared mailbox and and so on and so forth. But then downstream right where if, for example, we’re integrating with Dynamics 365, you know we’re making use of the CRM SDK to make sure that we can, you know, creatively create an opportunity. And and things of that nature. And then moving on to. So one of the things as a quick aside, I I when I was thinking through this, I was like you know what is the focal point here? What’s the most important thing? Is it Azure functions? Is it the database? Is it the akas? There is no easy way for me to say, but everything is important. But there are degrees to which there are there are more important and less important, and I think one of them is. Our our database data is at the center of this application and as I said before, your business strategy is your data strategy. That’s something I’ve heard from from very reputable people and the this architecture and this this the way that the system is designed is meant to facilitate the collection of meaningful data. What does that mean? If you have good data at store at hand, you’re going to improve your artificial intelligence output quicker. If you are collecting the right data, you may turn around and iteration of an improved LM from, you know, potentially waiting months to get data to weeks. Because of the the sheer fact that, hey, you know, we have this at our fingertips and there’s there’s engagements that I’ve been on where we’ve had so much data that we had, we ideated and said, hey, let’s create a model to do X and you know our teams went went back and and and implemented the system, taught, taught in LLM. And did it, uh, because they had data and then we had engagements where, you know, we started and found ourselves, you know, we can do this, but we lack the data to to, to, to be able to, to fully implement it and effectively what that put us at is is a position where we needed to keep that in mind, put a pin in it. We’ll come back to it once we launch and gather data and then we can start to iterate. So data is. A very important topic in in the system and then. There is the money maker. I did mention that AI is unequal, unequivocally important here and. We this is a webinar in and of itself, but effectively we’re where we’re at with this is we run a containerized Python environment that receives things like the the transcript of the conversation, the the body of the email, the subject of the email to effectively you decipher it and attribute specific rules and translate it into something that we can use to work with. Right, so a huge BLOB of text email that is requesting us to, you know, give me the price of a cup of coffee. You know, a large cup of coffee, obviously I’m. I’m minimizing the the the the problem here, but we’ll get an email requesting that this system, by virtue of of of being trained to read emails like that, will take that and and and transfer and translate it into a tabular format that we can then use to query against other database and push it downstream and integrate it with other systems therein. And then once we’ve went through the flow that that we’ve roughly discussed, right. So keeping in mind that we’re we’re at the end of our journey of of this of this process, but you might imagine that at a very high level a new email comes in, it goes through the logic app. We confirm that it is that it is my apologies that it is a indeed a quote and we do that here. We send it through our Azure function which. Amongst data transformations and and storing it in databases and logging events, the the main focus of what it does is it also sends that data to AI. AI is able to decipher that email brands lated into something that is pertinent to us, to our product in spits it back to our Azure function, and then what we need to do something with it. It needs to provide value to our business and we do that by. Integrating with some other external system that that is in use today, which brings me to our integration piece. The way that we’ve Mark is added this system. We allow for the integration of uh of different products at the I’m set during the ingestion. What also, during the downstream processes? What does that mean? Uh, if, for example, all of your products back going back to the CPQ concept, if all of your products are stored in SAP, we’re able to integrate our system with that and feed our product data from from there and use that to drive good data being sent into back into your CRM and effectively uh, the possibilities are endless in terms of integration here. We’ve done this in multiple in multiple ways with multiple vendors with. Multiple platforms. Uh, sometimes custom creams as well, depending on on on the engagement. But effectively, if if there was not an integration piece and we couldn’t take a quote and get it into some shape or form and push it out to a CRM, then this would be moved and and and the ability to do that it it is is is you know imperative umm and that is the architecture really is at a very high level. Of of how it works and what it does, and the pieces that come in come together. Uh, well, let’s talk a little bit about how we actually deploy this and manage this maintain this. We’ll talk more about. We’ll talk a little bit about Terraform for those who haven’t heard of Terraform in the past Will will give it a little one on one on one session, but then we’re also going to to discuss some strategies for that for deploying your infrastructure and really how, how, how all the intricacies of that. So without further further ado, let’s talk Terraform one is Terraform. Terraform is an infrastructure’s code software tool that can easily be used to deploy applications to multiple cloud providers. Azure being the one that we we specialize in. What are the benefits of Terraform? There is version control integration. You can deploy an entire application landing zone in seconds. In fact, you can deploy an entire enterprise landing zone in in seconds as well. You see standardize deployment workflow and honestly for the folks on this call, we could have a webinar just for Terraform as well and and a few hours of that. But it is it is imperative in the system as effectively the Holy Grail of in the in the Terraform ecosystem is is its state as it applies to an application, Terraform State is is effectively a way to determine if changes have been made. It’s it’s used for change management. It keeps it’s a log of hey, you know, the last time I executed these scripts, this is what I deployed. And when we run and when we go ahead and deploy a second time and checks against those, the Terraform State to know what changes to make? Uhm, and it is you, and we’ll get to that a little bit later on. But effectively it is storing a lot of those keys and data. That’s that. The deploys in a very raw format. It doesn’t obfuscate any keys or passwords, and when I say it’s the Holy Grail and I mean it because in some way shape or form. If for whatever reason it it found itself to be. You know, hacked then you know that could pose some issues and that is why we separate that areas of concern and in our architecture often find that in a storage account of its own lock down it’s various uh firewall policies and access control. We’ll talk a little bit more about that in a second and that here on the screen I have an example of a of a Terraform script used to deploy a Azure SQL Server and Azure SQL database and. At a glance, right even to those untraining Terraform, I think it’s rather easy to determine, you know, kind of see what’s going on here. First of all, that we’re gonna deploy a SQL Server. We’re gonna point it at some sort of resource group name and the and then we’re gonna deploy a SQL database and it’s going to, umm, be tied to this SQL Server AD and and and the point I’m making here is Terraform for its rather simplistic way of of reading. It serves to improve our business or our our deployment process is significantly in that we can write a Terraform script to deploy an entire application landing zone once and then never have to worry about it or at least minimize the overhead with maintaining it. And let’s say for one reason or another, we start with a dev environment and a production environment and we find ourselves to need a third environment for staging or QA. You can deploy 1/3 environment in seconds versus having to go out and and ask your IT department to create a new environment and then they have to, you know point and click and create all these resources on their own. So these scripts are are or Terraform as a whole or a hugely helpful to to streamlining development processes. And let’s talk a little bit about GitHub actions and how they tie in with Terraform. Umm, one thing I was mentioning before is that this concept of Terraform State and how it’s the Holy Grail and architecture. It sits elsewhere, completely segmented on its own and it’s own storage account for no other reason than nobody really bought our pipelines should be accessing this. Only our pipelines should be going in and saying, hey, do I have I updated? Have I changed something? If I haven’t cool. If I have then I know that I need to overwrite this. What effectively in our system somewhere we have a Terraform State file and there is multiple methodologies to deploy infrastructure. I’m using terraform. Some teams do it using uh, approaching it from a completely separate repository, completely separate, uh, area that the the the deploy their code from that they’re the they deployed their infrastructure and their application phone. Some of them do it alongside so effectively as part of the your exact same code repository and the exact same deployment pipelines, you’re always redeploying your infrastructure. Uh. And there’s pros and cons to both, depending on the size of your implementation, you might find yourself. It might be that a a large enough project will not. It will be better off to deploy it separately for no other reason than just visibility and making sure that we have a good handle of what’s going in and what’s what’s changing. If it’s a smaller project, then oftentimes you’ll see it deploy alongside where and there is just no need to segment it to that degree. And then there is the the next level of of. Of Terraform segmentation and something that I’ve seen is deploying infrastructure in pieces. So you might imagine that in line with these with these highlights that you know the logic app is deployed separately because in this system it’s it serves as a as its own area of concern. There is the uh application at large the.net API. The database that’s deployed in together and then you deploy your AKS along with your Python environment separately as well. And there is no. It’s kind of like react development where where the best practices are fluid. The same applies with Terraform. Are these standards are sometimes more fluid depending on on on on your use case. But with that said, let’s talk a little bit about getting my actions. Getting up actions, offers and this this getting offers a native support for Terraform. So what you’re seeing here is a is a snippet of code that can run my terraform scripts and deploy my infrastructure in a total of three steps, four steps if you count the check out of the checking out of a branch, which I wouldn’t, but effectively the reason why this is so powerful is because. I’ve had the pleasure of doing this in 80 and and Azure DevOps and I’ve spent years working in Azure DevOps and I and I love it to death, but setting up Terraform is a little bit more of a of a of a an involved process. So much so that it’s something we consider into the overhead of setting up the project, and depending on the size of the project, you just might not do it. But with GitHub actions you can do it all in this amount of effort of work and and and that really opens it up to the viability of using Terraform with GitHub actions and underlines. Just how, uh, how helpful it is, it’s going to be going forward and effectively. It completely integrates with our Terraform Scripts as it is, we can set the workspace that we’re in from within these arguments in our GitHub actions and effectively use this to very quickly deploy entire infrastructures while setting up a brand new project. Brian Haydin 36:25 Where you going? On what about our customers and DevOps? Like, how would they solve that same problem? Instead of GitHub actions. Mac Krawiec 36:37 How do you mean? Brian Haydin 36:39 Like uh, for the the deployment automation, I mean, can we still do that in eight in Azure DevOps? Mac Krawiec 36:45 Yeah, absolutely. So I’ve done that multiple times. Uh, it it’s 100% doable once you do it once. It’s not. Not at all difficult. That being said, it it does takes it. It takes a little bit more of an effort to implement that in ADO versus GitHub actions, but with Azure DevOps you could solve this problem by executing some PowerShell scripts from within your pipeline agent. That effectively set up your Terraform container and and set your workspace and then run your plan and then and ultimately apply your your Terraform plan. Those are all things that you have to specify and write on your own as a matter of fact, depending on your on your pipeline agent, you might even have to install Terraform from the beginning on your agent, so it’s doable. It’s been done for years now. We’re used to it, but I’ve had the pleasure now of also coming to the new age, which is this and and and now I can effectively do it in three steps. Does that answer your question? Brian Haydin 38:03 Oh yeah, yeah. I was just planting a question to to get the conversation going. Mac Krawiec 38:09 Yes, Sir. Brian Haydin 38:09 Yeah. And there’s other ways to do it too. You know, you and I were talking yesterday about being able to do some of these deployments natively and Terraform Cloud or other ways as well. So great. Mac Krawiec 38:21 Yeah. And and and like I was saying, Terraform is a webinar in and of itself. Brian Haydin 38:26 Yeah. Mac Krawiec 38:26 Umm, one, you know I’ve I’ve. I’ve used it multiple times. I’m adept at it, but to you know, there there’s a lot of other facets to Terraform that that need to be uncovered. Umm, let’s talk a little bit about the maturity curve of AI in our products and applications and some of the things that I I recently just I’ve come to a realization is there is a there is a significant amount of companies that are still down here. There’s a lot of companies that still operate out of Excel, manage their business and excel some big large, you know, some smaller than others, but they do and. We’ve taken companies just like that. Uh to about this area of the maturity curve, just by virtue of implementing this system that we’ve been discussing for the last 30 minutes. Uh. Effectively it there is little little effort overall to get to the machine learning aspect of hey, we are now able to predict that. Mac wants to buy this, and it’s gonna cost that much money. And let’s quote him that all in seconds the the overhead to jump from here to here is relatively low and the amount of effort that we need to put to put to make that happen is relatively low. And that just just one thing I wanted to highlight because you might find yourself to be that that, that client, that company right where where you are out of operating out of Excel and and using AI seems like something so abstract and I’m telling you it’s not, it’s something that can be done in a relatively short term and you can see the fruits of of of of that work turn around pretty quickly. Brian Haydin 40:22 You bring up a really good point. I mean you you’ve been working on some of these CPQ solutions for years now and I remember, you know some of our earlier projects, you know taking years to get to a point that you cannot get to in months, you know or even weeks, right, you know the innovation, the newer innovations in the Microsoft Stack absolutely enable this to to go so much quicker. Mac Krawiec 40:39 Yes. Brian Haydin 40:53 Uh, and we haven’t even scratched the surface. I mean, you know, semantic kernel’s coming out. That’s gonna help with a lot of the agent and orchestration work. You know, there’s there’s so much, you know, coming out in the pipeline and with Microsoft build coming up next in the next couple of weeks, I was probably gonna be a lot more announcements as well. Do you have any thoughts about that Mac in terms of how much quicker this might actually this might actually be? Mac Krawiec 41:20 Yeah. And so to your earlier point that you know, we’ve seen our clients, some, some take years and and now we’re at a point where it’s months, there is no reason that this can’t be reduced to weeks now too, to your point, there’s things coming down the pipeline and and there’s a huge focus in the Microsoft space to Accelerate this even further and bring our everyday companies to this to this stage. Uh and and my thoughts are, you know, let it happen effectively and and we’re here to ride the wave. Uh, and Brian, if you wanna talk a little bit about some of the next steps that we can offer to make this a reality for for anybody on the on the call. Brian Haydin 42:08 Yeah, absolutely. So we do a lot of these uh workshop sessions with with our customers directly. So if you want us to come and speak with your organization, we are more than happy to do that. One of the first steps that we would do is getting started with a group envisioning session or an executive envisioning session and where this could be anywhere from a couple hours to half day, sort of a workshop environment where we introduce you to some of the capabilities of AI. What? What art of the possible things can you know? Get the creative juices you know flowing within the organization and then maybe do some breakout sessions. We can do this virtually. We can do this on site and and an output of this that we deliver for our for our customers is a value analysis where it’s kind of a breakdown and we help you stack rank some of the ideas that come out of this and work on next steps. Here’s an example of a an agenda for like you know, that sort of half day, you know, few hours, you know, type of workshop. Again, we’ll do an art of the possible sort of expose you to some of the things that we’re doing and other customers are doing in this space. And then you know, get out into the the breakout session. You make it really fun here in this agenda. We’ve got, like, prizes and winners and stuff like that, but we’ll work with you to figure out what makes sense for the organization. After we’ve gone through the envisioning session, typically what our customers want to do is break out, you know, one or two items that uh and the value now ISIS might be, you know, low to medium to implement, you know, kind of get your feet wet and we’ll help evaluate the scenarios in look at how you might want to pursue that. So there is a survey included at the in the link here in the chat and if you wanna work with us on one of these envisioning sessions and maybe explore what a pilot would look like, fill out the survey and indicate that you’d like to reach out to you. Ohm again, you know our program starts with usually with the executive envisioning session. Uh, and uh, we would like to get the leadership of the organization before we do a full group full group and we can also as a Microsoft partner in the AI space, help uh, look at possibilities from Microsoft to fund some of these events. And with that, thank you very much and look forward to hearing from you. Mac Krawiec 44:49 Thanks everybody.