Hi everyone and welcome to today's Product Office Hour for FirstSpirit. I'm Timo, I'm your host for today. And these are the items on our agenda. We will start by explaining what the session is all about, then talk about the new community, and I will give you an update on how the migration from webforms to form-centric is doing. In terms of releases, we will have a look at the past nine months and go through a few items there. After that, we will get our hands dirty and look what the newest advances in AI and FirstSpirit are all about. And then tie things back to GEO, Generative Engine Optimization, and MCP, Model Context Protocol. If those terms don't mean much to you yet, no worries. There's going to be educational material on those two topics and we'll discuss why they are relevant today and definitely trending topics. After that, the roadmap is going to be what we focus on. And then let's talk a bit about upcoming events, for instance, the DMX go in Cologne or the DEF CON in Dortmund, Germany. Today, we also welcome DXM customers. Thank you for joining. In addition to the FirstSpirit customers. So what is the FirstSpirit Product Office Hour all about? It is about discussing new and upcoming features. So we have demos. There's going to be an outlook at what happens next. And at times, we invite experts. Not today, but if you look at the recordings, because this session is recorded, you will see that we have plenty of partners, third-party vendors join, also experts from within the engineering teams of GroundPeak, for instance. And we have a Q&A section. And we have a Q&A section. And a chat in Zoom. So if you click on that, type in your question if you have any questions or remarks. And on the right-hand side, for the ones who are watching the recording, you will find QR codes. Those will lead you to registration pages. We have the English session today, German session tomorrow. And every three months, we have a Product Office Hour. On the Community. So that is what was previously known as community.groundpeak.com. It has moved. The move is complete to Zendesk Knowledge and Zendesk Community. Different platform, different product. But the key message is that all those posts that we deemed as relevant, because there's there are good insights, there's good knowledge in those. There are good insights, there's good knowledge in those. Those question and answer sections. Selectively for all of the products, and in its entirety for First Spirit, we're moved to the new platform. Why do we do it? To consolidate and to focus. We have a big portfolio, and we wanted to have all of those products shown in our new Zendesk and then as community. Also, if you're looking for release notes, you'll find those here as well. As you see here, once you go to those pages, you have access to the content without logging in. So you don't need to create an account. But if you want to follow an item, so you have a whole folder of topics, you can follow that. Or you have a single conversation you can follow that as well. And if you want to follow that as well. And if you want to leave comments on a specific topic, you need to log in to do so. One last remark, the redirects are also here. So let's imagine you still have that old URL or a community.groundpeak.com post. That should still, as you click on that, lead you to that very post. Now within the new portal. On another note, for our First Spirit customers, we are also starting a quarterly newsletter. You see the German version here on the right-hand side with enough interest to reach out. We are happy to provide that one in English as well. Informing you to help us. Informing you to help us. Informing you about recently released features, upcoming webinars, and on-site events, for instance, the Customer Connect Day. If you haven't received that newsletter yet, or if you're interested in getting the next one, do reach out, please. You'll find my email address at the very end of the session. On the topic of migrating from Web Forms to Form-centric, an update. Web Forms, of course, being your form solution in First Spirit. This one getting a big update with Form-centric. And here we have a timeline for the three major pieces. Form-centric without customization, so custom solutions in your project that you might have running in Web Forms at this point. If you don't have those, this is your path. The newest version of Web Forms is being your form-centric. The newest version of Web Forms is being rolled out with this month's release. And this one is to allow to update to Form-centric. And for the first half of next year, the idea is to have that migration piece fully covered. The key topic to tackle here, of course, is that we want to have all of the data points migrated, but also during migration, we want to make sure that data is still reliably connected. Basically, this applies to the with customization path to, except for the fact that this then happens after we've covered the migration without that customization piece. And then last but not least, we want to cover the headless part as well, which is not directly tied to the without customization and with customization. It's a separate web app. And here, next year is what we are aiming for. Let's have a look at the releases, the recent ones. So within the first nine months, quite a few things happen on the first period side. The code system was released. That is, of course, the newest, the next addition to the AI suite that helps you identify issues in your code, in your project, and it will help you jump directly to those to then fix them from within the next one. And fix them from within the project itself. Prompt management, that's the big one in terms of compliance and scalability. You want to make sure that the teams that are working with the AI suite have the right prompts, AI actions available in those projects of their need. While others might not have that full list, different one. Because here, because here, because here, with prompt management, you can define those prompts that are available in a single project, in multiple ones, in all of the projects, and handle that in one nice piece of interface. Conant Creator got way more accessible over the past months. Here, a big release between Q1 and Q2. But as we see this as an ongoing task, more on that in a minute. This is, of course, ongoing work. The language-dependent release, piece of functionality. Here, in a granular fashion, you may want to make a change to that German translation of that page. That then, that then stays as in progress in your preview and content creator. While you also make a change to the French translation of that page, which then, you say, should go live. To have that split, depending on the language, is what this feature is all about. In Moody's app management, in its newest iteration, we are considering changes that were made in the target project. And are then communicated back in a new interface to the main project. For the editors in the main project, then to say, yes, we accept the change, or no. We will not consider the change when we do the next content distribution. AI and DQM. AI and DQM. That is, of course, the executive AI-powered summary that you have on the right-hand side, nicely available from your dashboard. That gives you, in a few sentences, a short paragraph, a wonderful high-level view on what is happening, what the key issues are. Collaboration. Collaboration. We will talk about that, and I will also demo that a bit more later in our session. The key idea here is you want to work in a team on content in content creator, and you have, I think, a wonderful way of doing so with that new collaboration piece of functionality. Closely tied to that are the AI agents. As you will see, as you will see, when we showcase collaboration. AI agents live in that feature, and are basically part of your team. How that looks like, I will also demo that. Let us zoom in on a few items. Because here, for OCM, for instance, you might want to take action. With OCM 2, we've, we've, of course, provided the Omnichannel Manager. And we will provide that up to a certain version. And then we will stop doing that. We will stop supporting it and providing it with the 2026.2 release of First Spirit. Why is that? Because we have a, as we think, much better new solution. That is on OCM.3.3.0. And if you've used it without customization, the change is absolutely minimal. You only need to go to one URL, change the two to a three. If you've done more than that, and have customized your Omnichannel Manager, do have a look at the documentation and migration path here on the right-hand side, because it will tell you what to do, if that is the case. And also, if you need an extension on the timeline, don't hesitate to reach out. There are ways that we can support and help you. Let's also talk about language-dependent release. Because the Open Beta has started. Here, the API-based approach has been in the product for a while. Now, we have the user interface, which we say is in beta, meaning, you can test it. It's not generally available yet. It might have a few fixes that need to be done, but nothing major. We are quite confident it is working. Do try it out. Also, give us feedback, as we will take that into consideration before making it generally available. Behind that QR code sits a page that explains to you in more detail than I'm doing here, what the functionality is all about, its potential limitations, its options, how you implement it, how you use it, much more on that. As I mentioned earlier, Content Creator got more accessible, which is a wonderful thing. It's more inclusive. It has screen reader compatibility, smart focus handling, and with your keyboard, you can control, as you see here on the screenshot on the right-hand side, Content Creator. Now, for the ones who are already familiar with that, you might have noticed that we are showing something new here on the screenshot in the middle, and that is what we call Navigation 8. Imagine an interface, a list that you open up, that helps you directly jump to that very section within your Content Creator. Here on the screenshot, you also see that we have search. So if you want to address that from within one interface that basically sits on top of whatever page or section within Content Creator you're in, it allows you to do so. It's not here yet, but it's going to come soon. Now, let us move over to Content Creator. And so what we're seeing here is, of course, our web-based user interface is a web-based user interface to create, edit, and optimize content. In this case, website content. And we have here crumpic.com in a demo environment. So we're not going to do anything to the actual crumpic.com page, but rather I will show you a few new pieces of functionality that the team either has released or is going to release soon. So let me scroll down a bit so you get a better idea as to what we are seeing here and doing here. So this is a, I think, straightforward page with a bit of text and you have a button here, a few images, a few elements. This is the end of it. Now I want to showcase our newest advances in terms of the AI suite. Let me go in. So usually what I would do if I was going to do is edit this text, these two sentences with the AI suite, I could, for instance, pick Content Assistant, which is what you see here now, and then say, yes, I want to expand on what was written here. And this is straightforward. And this is known for the ones who are seeing this for the first time. Let me briefly walk you through it. We have way longer demos on that. Do reach out if you're interested. For instance, a nice comparison between what was written by the marketing team, what was suggested then to be added by AI, you see it here, or even removed, you see it here. And if I'm not satisfied with the result, I can quickly create another one and another one. So this is, as I said, straightforward. We've added something new. Let's go to tone of voice. For the ones who are familiar with the AI suite and content assistant, you might ask, well, tone of voice, what is new here? New is the following. As I click on that icon and open up my editing capabilities, because I'm now having a look at the prompt that lies behind that AI action, I can fully edit that, given the correct rights to do so, of course. You notice that the prompt itself isn't very specific as to what the tone of voice should be about. Rather, it is saying to fetch the writing guidelines from Confluence and then consider those. And also, we have a few more things on this AI action. This AI action leading them to a certain output and format to make it as efficient and focused as possible. For the ones who are familiar with MCP, you might have guessed that this is now our MCP compatibility that we are showcasing here. Because as I scroll down, you, of course, see I could put this in a certain folder. This is straightforward. But also, down here, we have our new functions area. And here, we have a few adapters, plugins, connectors, all of them about model context protocol. And bear with me, there's going to be more in terms of what this actually means and stands for and so on. Let's focus on the demo first. And this is the one, the Confluence connection, that this tone of voice action is referring to. So, let me close this. When we have this proposal generated. The rules for that are actually coming from this Confluence page. So, external tool, we are connecting with that through MCP. And here, we have our five rules. Zoom in a bit. One of them being that we have our five rules. One of them being that we have the trademark symbol that needs to be added whenever we see first spirit in the text. And as you might have noticed, this is exactly what happens. We have first spirit. And then my proposal is here. There are a few more proposals. But this is the relevant one, of course. Because let me show you what happens if I edit this. Like that. Hitting save up here. It's saved. Going back. Let me re-trigger this. There we go. So, you see in real time, imagine someone else in the team is making changes to your tone of voice. And tone of voice being one example, of course, of many. You see it reflected here right away. Let me just change that back. And let's go back to our demo. Going to give you a minute here if you have any questions or remarks. Please leave them in the chat. Just see if anything came up yet. Nothing yet. And of course, for the ones who are seeing this for the first time, we have many more options in terms of the AI and the AI. actions that are with content assistant. In the same manner as you just saw it. You can create them. You can duplicate them. You can edit those however you like and whatever is relevant in your project. But there is more to show. So, we touched on collaboration and we touched on the AI agents. On the right-hand side, you might have spotted this icon already. So, here, we have a what you would see is our collaboration functionality. And we see a long list of actions. Things that were done. Why is that? Because I'm seeing everything now. Also, the resolved ones. Let me just do that. And so, on this page, apparently, we don't have an issue. No comment. Nothing to collaborate on. Let's change that. So, I click on add comment. So, I click on add comment down here. And then, as I scroll through my page, you see the highlighting. Because this is highlighting the element that we now collaborate on. Let's pick this one again. And I'm going to say that's this. So, imagine we are working on one team. So, imagine we are working on one team. We are working on this page. And you see here on the right-hand side now, you should see in collaboration this item. Of course, I could reach out to someone from you. Or I could say, actually, I'm going to involve AI. So, we have the first bird AI here. And I'm going to ask it a simple question. How can we improve this? What this agent is now doing is it is having a look at what else is on the page. And you might remember, as I was scrolling through the page, there's quite some content on that page to work with. And it's considering all of that to then make a proposal in terms of, zoom in here a bit, in terms of how we can enhance our data. So, we can enhance this and clarify the existing intro text. So, it notices that this is an intro text for better engagement and flow. And you see here, we even got a second suggestion and a general advice. What we could do now is we could go in, copy-paste what we have here. Save that. Save that. Or even say, well, I want to do more things. Maybe I have my keywords created here or do another round of review or shorten. Because AI capabilities in First Bird are tightly integrated with one another. And basically, we have the handoff now. So, as I'm actually satisfied with what is written here. So, as I'm actually satisfied with what is written here, I could say, well, this one is going to be resolved. So, I hit that icon here. And it's going to disappear from my list. Let's imagine someone else from the team is also working on this element. And they might have forgotten. And they might have forgotten that they actually left a to-do. Now, what happens if we release this page with the to-do still in that sentence? What happens is that this element gets highlighted automatically. You see it here. And on the right-hand side. And on the right-hand side, I see a new action item. Something to collaborate on. And the AI is telling me, oh, look, you forgot something. And it's saying, please improve. So, I could ask the AI to do that. I could go in myself, as you see here, real quick. I think I'm capable of doing that. And then trigger the release once more. And as you see, we have a nice release workflow here where this collaboration element, you see it up here as well, got resolved right away. So, what have we seen? We have seen so far content assistant work on text, optimize, use MCP, influence, tone of voice. Then we've had a look at the collaboration. And you saw me add that action item involve AI to improve and also the workflow that picked up on the to-do and flagged that. And I want to showcase one more thing today, which is GEO capabilities of existing AI suite assistants. GEO being generative engine optimization. More on that in a minute. So, here, I have my analyze assistant. And it has a few options I can pick from. I'm actually going to do a GEO recommendations analysis. So, I click on that. I hit the analyze button. And as you see, it goes through the whole page. And it gives me quite a few results. Took a few seconds. Took a few seconds, right? And then as I go through it and click on that icon here. Let me zoom in a bit to show you. Click on that icon here. I see on the left-hand side the text. And again, we have full transparency on what has been suggested and changed. Here on the left-hand side the current text. And here on the left-hand side, we see the suggestion. If I'm not satisfied with a suggestion, I can also say, well, don't consider that. And I'm also not satisfied with that. And you see a change here. For full transparency, what lies behind that GEO recommendation for the analyze assistant. Let me open up the prompt is this. So, here, we have nicely written a prompt on optimization for AI, for large language models, to be found and adequately represented by LLMs. And do take a screenshot. Do copy-paste that prompt if you want to use it. Going to give you another minute. Do you want to ask questions or remarks? If so, please use the chat functionality. And with that, we are going back to the deck. So, let's recap. What did we see? We saw the collaboration and commenting functionality. And, of course, this scales nicely across big teams. You saw me work on that. But imagine you have 10, 15, 50, and more people who work with that. You also saw how the collaboration piece opened the doors to involve AI agents. And here you saw the first agent at FirstPirit AI that was used to make recommendations on how we can optimize text. We spoke about GEO. Again, again, the idea here is that you want to optimize your content, your textual content. So, it is found by LLMs that are crawling your page as a user is performing a web search or as the training data is built. And you also want to do this in an adequate manner, as in want to have a better position as your competition. And why is this? And why is this even relevant? Why should you even care? Bear with me for a minute because you will see a bit of data here, which is backing up that claim. If you look at the traditional Google search, you see a certain ratio between how often pages were crawled and how many referrals came out of that. Referral here being the one who was looking for a CMS than actually visiting one of those 12 pages. So, ratio 12 to 1. Microsoft's Bing. Here the ratio is 43 to 1. Now, let's head into AI territory, even more so than with Google and Microsoft. And I'm fully aware they have an AI summary as well. Perplexity, more than 200 to 1. And the leader in this market, OpenAI's chat GPT, has a ratio of more than 1,200 pages crawled with one referral. And if you want to look up those numbers, you see the URL here at the bottom. I highly recommend having a look at that dashboard or those dashboards actually, quite a few by Cloudflare. So, what is the message here? The message is that traffic is shifting from traditional search, web search, to LMS, large language models. And you, as someone who is working with websites, you want to still make sure that your content is found, that users have a look at your page, ads, overall traffic engagement on your page, and you want to secure that. And if that is not the case anymore, you still want to secure, of course, that you're adequately described within, for instance, chat GPT. And to do so, let me walk you through a few best practices that today are recommended to optimize for Gentile engines. The first one being to give your content structure. The first one being to give your content structure. You can do so by giving focus with key points. Having a format that is close to a questions and answer format, because it actually matches the large language model queries. Meaning, you're using chat GPT, you ask it a question, and you get an answer. Ideally, the user gets an answer that's the LLM, but the LLM got from your website. And informative chunks, as opposed to long, winded, fluffy paragraphs is preferred. Gaining trust, that's a big one, of course, with original content and verifiable facts, links to sources and also credentials. If you're reading this, and you ask yourself, well, isn't this SEO, search engine optimization? Yes. Sure. Sure. It's pretty close. To speak clearly, using natural language with meaningful variations, having great synonyms, for instance, and also connecting your informative chunks to related subjects, that is going to do well. Last but not least, machine readability. Enable the AI, the LLM, to find the relevant content, you're doing this, you're doing this by markup, best practices. Here, of course, an FAQ is an FAQ. You want to name it like that, structure it like that. The same applies to the how-tos and the articles and the products and so on. And yes, another SEO point, do a great SEO job, and GEO will follow. Since we also touched on the other big trending topic in the AI domain, MCP, let's speak about that for a second, too. Here, the idea is to connect to those tools out there in the AI ecosystem that are also MCP compatible. Why do that? Because they might provide services that are not natively available in the CMS. But still, as a user in the CMS, you might benefit from those. Let me make this a bit more tangible. Here on the left-hand side, we have an MCP compatible CMS. First bird, becoming one. You saw it in today's demo. We do not have a closed or an open beta yet, but of course, we're aiming for that. And do reach out if you're already interested. On the right-hand side, you have that AI, that MCP ecosystem with all those tools. Now, let's assume that we want to optimize our content against certain trends out there in the market. Within the CMS, we don't have a trend analyzer. But out there in the AI ecosystem, there are those tools that will provide you with market information. That is then used in the same manner as Confluence was used when we looked at the tone of voice use case to optimize your content. On the right-hand side, it might be that you're not working within First Spirit, but outside of it. And still, you want to create a new website. Usually, you would go back to First Spirit, create a new website. But let's imagine for a second that you're using Claude or JGBT or any other of those. And you're using Claude. I'm going to leave this here in case you have any questions or remarks. OK. And again, let me just emphasize the MCP part that you saw today that is being worked on. And we are excited to make it available to you and form most likely of a closed beta first as soon as possible. So we had a look at the AI suite. And actually, we only looked at parts of it because the AI suite is much bigger. And what is available today is not only content assistant and collaborative AI agents, but we also have image assistant, code assistant, and underlying the assistant API. So think of use cases where you want to quickly analyze an image in terms of its content and generate all text based on that at scale. image assistant is here for you. Image assistant is here for you. Code assistant, as I touched on earlier when I spoke about the releases of the past nine months, is also part of the AI suite. See what is coming up. MCP support. And we saw it in a demo to AI-powered workflows. And if you ask yourself, well, how do I get my hands on the AI suite? We have you covered. We fully support SAS and on-prem. We fully support you bringing your own AI API key or BrownPeak, providing one for you. If you have your own model that you want to connect to our AI suite, you may very well do so. natively, natively, natively, natively, we integrate with open AIs, Azure AIs, and AWS Bedrock solutions. And also, what we offer is a trial phase. We align on a time frame, few weeks. It is free of charge for you. We have clear goals and regular check-ins to make this a success. Now, let's have a look at what the future holds. And we think of the roadmap as balanced. Because while we saw lots and lots of AI today, and we are working towards an autonomous content management solution for you, we also want to balance those AI features with what is definitely not AI-powered and what shouldn't be AI-powered. Because we think that this is still tremendously important and very much relevant. To give you a few examples, what we focus on, if you want to think about buckets, are these four. So yes, we spoke plenty about AI. You see it here on the left. But also, when we think about the developer experience, we offer, for instance, Git-based development. And we have CAS and GraphQL. Plenty of APIs for you to work with, with the REST API for First Spirit being worked on at the moment. That is going to come. And definitely, you will learn more about it during the next or the part of this or after that. When you think of project work. When you think of project work. You have content that you need to distribute to 70 plus projects. And you want to use more design management for that. Surely, this can be enhanced by AI. But at its core, not yet, at least, we have here a very deterministic feature. And we're going to keep it that way. Remote data, same applies. And yes, the self-service SaaS configuration. That is something that we always aim to improve on. And in terms of quality of life, you heard me talk about accessibility. There is more to it. The UX UI, we always want to improve that and make your work with content creator more efficient. If you're interested in details, we have, I think, a great roadmap tool that you can access by going through that to that URL. You will find there are a few sections under consideration. That is what we want your feedback on. What is it that we build next? Then you'll find planned. So that is planned for, for instance, the next sprint or the next quarter, depending on the size and complexity. You will find the beta section. And this is where the close and the open betas are placed. And if you click on any of those elements, you will find ways how to contact us. If you want to participate. And then launched, we'll tell you about what was released this year, previous year, and so on. Now, let us wrap up today's session with information on events. Here we have the big one in terms of digital marketing and tech. That is DMX Go. It's going to happen on the 17th and 18th of September in Cologne. Plenty of crumpy colleagues will be present. We have a booth. We have a booth for those two days. Additionally, we have several partners that are there to help you to have conversations with. They also present. They have great decks prepared for you. And we have a master class. So think of the big stage and ground peak plus UDG, one of our partners and a customer. We will talk about how one stays agile despite being placed in a rather strict environment or industry and how one can still find a path to AI with best practices and more information on what we touched on today. Geo, MCP, for instance. The other event is the DEFCON. This one is for the developers, for the partners, and everybody who is a domain expert in a certain field. In tech who wants to join in tech who wants to join this event. Think of hackathons. Think about super nerdy deep conversations about project solutions and implementations. Very intense, very fruitful. Those two days in November. The exact date will be communicated soon by email, most likely. And with that, I want to say thank you to everybody who joined today, giving you one last chance to write your questions, leave your remarks in the chat. And as they say, silence is golden. So thanks again for joining. See you next time. Have a good day. Bye. Bye.