What’s new with Amazon QuickSight

Neil Colley: Uh my name is Neil Colley. I'm a worldwide principal uh go to market leader uh for Quicksight. And so I've been with uh Amazon and Quicksight for going on six years. And um it's been a, a tremendous journey, Doug, it's nice to meet everybody.

Doug Berquist: I'm Doug Berquist. I'm a manager of solution architects for Quicksight.

Hi, everyone. My name is Bashi Meta. I'm a QuickSight solution architect.

All right. So by show of hands, uh how many folks in the room have interacted or used Quicksight before? Oh, yeah, love this, love this. This is awesome. All right.

So, you know, from a Quicksight perspective, let me just catch you up a little bit for those who, who haven't, you know, heard of Quicksight before. It's a unified BI tool uh suited for all your analytical needs. Um one of the things that we got right? You know, when we first created Quicksight was the serverless component. So you're gonna have that high performance and auto scaling built right in. Um and the thing that we're gonna obviously cover today and be excited about is the augmented analytics from a generative AI perspective. And then you know, one of the things that we also did at the beginning was, you know, a lower cost, right? And so that pay as you go uh model.

And so what we mean by unify from a BI perspective. So, you know, by doing our research, we found that, you know, most companies have anywhere from 4 to 7 different business tools, right? Or BI tools, they're using certain tools for reporting and that has a diverse population. They have a subset of tools for dashboarding. Um and that has aaa diverse population as well. Uh if you're, you know, creating intelligent revenue generating applications, you may be using a different types of insight tool to be embedded into that application. And then for search or natural language query, you could be doing a myriad of things, right? And so what that does is it creates silos uh from a BI, from a BI perspective. Um there's different UI, so you're gonna have to have those different learning curves and then from a commercial terms perspective, you know, it's a lot of stand alone contracts and commitments from that perspective as well.

And so when you think about Quicksight, you know, it's all baked in modern dashboard reporting, you know, comparable to just about any tool that you see in the market uh rep uh also reporting like so for legacy style reporting, pdf reporting, uh we had some phenomenal sessions early in the week on that, we also have baked in natural language query and mo insights that date back a couple of years ago. Uh and then what we're excited to talk about today um is the generative capabilities all baked into one, you know, performance based uh UI tool with lower cost in mind.

Now, the thing I'm excited about that, I've, you know, that I've experienced over the last five years is the pace of innovation. So i, I started with, with Quicksight very early. Um and since then, we've had 600 plus features in the last two years, we've been super busy, uh you know, creating a lot of innovation as well, right? So definitely started out with that cloud native hyperscale foundation. Of course, we've moved on to analytics for everyone, anytime and anywhere um as part of our embedding strategy. Um and now we moved on and incorporated things like generative BI as well. Uh so I'll pass it over to you Doug for the next session section I just was now.

Doug: So when we think about features in Quicksight, uh we think about four areas of investment. So we want to democratize data for business users. Uh we want faster and richer insights for analysts. Can you build a dashboard and get it out to your users quickly and more efficiently? Uh we focus on governance are the right users seeing the right data at the right time and then APIs for everything, right. Quicksight was born at a scale. Uh so APIs are really important for that. Can you manage the deployment um at scale at will?

So we're going to start with business users. So we've added quite a few visual experiences to make it easier for users to digest data. And so uh we focused on new mapping capabilities. Uh we've added a few new visuals um and we've made enhancements to pivots and tables.

So let's start with maps. We've added new base maps, both imagery uh as well as street maps. Uh we've changed the ways that you can digest the data. So marker clustering uh heat maps and we've added global geo coding. We've also added quite a few new visuals. Uh so text boxes uh which is something people have been asking for for quite a while. We're very excited that they're out. So, hello world. Uh we've also added radar charts. Um we've enhanced our scatter plots uh as well as we have small multiples, which is a great way to view data um where duplicate visuals across that values of a specific dimension. And we have enhanced our KPIs. So now you can add uh period over period change as well as a line chart at the bottom to track um you know, daily, weekly, monthly data.

So within pivots, uh obviously, pivots and tables are some of the most used visualizations within dashboards. Uh so we want to make it easier for our users to interact with them and consume that data. Uh so we've added data bars, we've changed um layout options to grab a hierarchy layout uh which allows you to hide and collapse columns. Uh we've also added contextual ro contextual row subtotals. Um this is really useful if you have a big pivot and you're looking at sub totals, you're not really sure where that aligns to it. Now actually has the value of that column in there. So you know exactly the total you're looking at. Uh and within tables two, we have freeze and unfreeze columns and different ways to set default widths so that you know that your end users are looking at the right data that you want to publish to them. Ok?

Neil: So gene BI is the most exciting piece uh that we've come out with this year. I was really excited to talk about it. Um it's really neat. I'm excited about it, but Neil stole it from me. So i do have to hand this to him right now. Um I'm gonna take it back when he's done though.

All right, thanks man. Really, really appreciate that. Um so, you know, finally, we hear from a generative BI perspective, i can tell you uh since Swami's keynote at New York City Summit, we showed a lot of good functionality, but there was always more to come and there'll still be more to come, but i'm excited about what we're about to show here from a demo perspective, as well as kind of double clicking into dinner to BI.

Um so we've made, you know, tremendous strides in our enhancements around AI powered author and experience, which enables business analysts to build faster. Whether that's calx, whether it's refined in certain visualizations. Um you can do that now with Amazon and uh Amazon Quicksight from a Q perspective now AI powered insights on demand, right? So simplifying the extraction uh of those insights for business users and we'll double click into what that means. And then AI assisted do data storytelling, right? The ability to tell a data driven story and do it with up to date and III and when i get to that part, this is gonna be a really cool area to show as well.

Now, when you think about the AI dashboard experience, right? You know, using that natural language uh to build and refine those dashboards uh being able to just simply uh text or write text to create advanced calcs and then refining those visuals. And then of course AI answers to questions on demand, right? Having those executive summaries where you can get those insights um and then easy, powerful Q and A experience for non technical users. Um whether that's, you know, multi visuals or suggested questions that's been verified that you should, you know, want to know about and ask or even vague question support as well.

Now, AI assisted data storytelling is the ability to generate those stories on demand and create that refined content. And of course, you know, the data is always gonna be up to date, right? We've all created dashboards reports uh for leadership, uh maybe your directs and sometimes that, you know, by the time you gather all that data and snip it into something like powerpoint, it becomes very stale, right? And so i think we, we have a different take on how you can go about that.

Now, let's, let's uh just double click here. Let's start out with a demo. So this demo is gonna be related to real estate data here. Uh so we'll just let that build for a second. So the first thing that we're gonna show is the ability in plain language uh to write a custom calc by just texting it. So there's no complex language that you have to learn, you know, like when you're utilizing certain tools or other tools. Um and then i can remember when Quicksight didn't have fl statements in it. So the ability for a non technical user to write this type of uh just natural language and have it generate a syntax is mind boggling in terms of the innovation just in a couple of years, right? Um so really cool functionality and then the demo user is just gonna add this to the dashboard. Uh so they're able to take advantage of that uh later.

And then the other component of this is being able to just type what you wanna build and what you wanna see, right? So in this case, you're gonna see a line chart, right? We're gonna add that to the analysis and then as that builds for a second, another thing that we're gonna be able to do is refine that visualization um by simply, you know, typing out change to stacked column chart and a market, right? So it's gonna simply do that now, you can just kinda adjust this visualizations size for a second. Um and then voila, it's there, right? So incredible functionality and Quicksight has always been about democratizing data so easily to publish this out, including the executive summary um as well as the data storytelling as well. Um so just let this build for a second.

The second thing that we're gonna talk about is AI powered insights on demand. Um and Doug thank you for creating this component because this is awesome right here. So we didn't talk about this at the New York City Summit and i can tell you for nine nda customers. I was, you know, bursting at the seams to really uh talk about this feature set, right? So we've all seen dashboards where you got 2030 different visualizations in it. You know, there was a lot of research that went into it, but you just want to get to the meat to the value of what, what's actually being stated, right? And so one of the things you can do with, with the Q functionality is also ask questions, right? So it's part of that Q and A experience. And then for in this case, it's gonna generate a set of visuals, three visuals um that you're gonna then be able to look and see if that, that question in Q and A part has been verified for accuracy. And then you can simply pin that to the dashboard for later usage when it comes time to do some storytelling, right? So we'll just let this build.

The other thing that you're gonna be able to do is filter sort uh based on what it created for you. So you can quickly refine that. Um and i can tell you we've actually slowed this demo down when Doug created it. It was just flying, right? Rapid pace. So, uh it's pretty awesome.

Now, let's get to what I'm really excited about is the AI assisted data storytelling, right? Like again, we've all created powerpoints, the data becomes still or maybe we have data and we've derived it from a couple of different sources and we wanna quickly be able to tell a story, right? So in this case, we've pinned a couple of those visualizations that we wanted, that we want to drive the data story that we're telling it to do. And then once we've gathered that and add it, we simply tell it, build me a story about mortgages in California and AI is gonna be a little bit verbose when it comes up with the title, which is, is kind of comical. But um in a sense it's uh it's, it's really cool functionality. So in a matter of seconds, um it's building out, you know, a whole strategy around this, whether it's a blog or if you wanted to do powerpoint, uh like you could. And so it's generated navigating California's escalating mortgage landscape, right? So very bobo in that, but then you can refine that from a them in perspective. Um you can, you, you know, uh you know, type a different, have it do a different tone of the text if you wanna be more authoritative or less, less authoritative things of that nature, right? And then also, you know, just the ability to um add the different them that you wanna add to it. So it's just, it's amazing like this was like when we first saw this like seven months ago or whatever, it was just a phenomenal functionality, we really couldn't believe it. Um and again, that's part of a subset of our generative AI capabilities powered by Bedrock. Thank you.

You do give me this back. Ok. Thanks Neil.

Um all right, let's talk about analysts. Um so if you might have noticed uh we've changed the authoring experience came out, i think a week or two ago. So we've made it easier for authors to build visualizations more quickly and efficiently. We've also added customizations that you can add to individual dashboards.

So with the new file menu, it's easier to navigate to the specific functionality you're looking to use while you're building the dashboards itself. And if you don't know where it is, there's the universal search on the top right there, you can just type in what you're trying to do and it will jump directly to that.

The three pan layout is probably similar to what you're used to. Um we've moved a few things around so that you're not digging in for functionality trying to find what you're looking for. It's much easier flow from adding the data, uh adjusting the visualization and maybe changing the properties of visualization.

We've also added custom time zones and custom week starts per dashboard so that when you publish out a dashboard to users in a specific time zone, they no longer have to worry about is this UTC, is this east coast, whatever it may be. Uh so they know exactly what they're looking for and the same thing with week start. If you start with sunday, if you start with monday, uh they'll know exactly what they're viewing.

We've also add added unified field level coloring. Um so what this is, it gives a level of consistency across the entire dashboard you can see here uh a pj and a pie chart is the same in the column and it will be the same throughout the entire dashboard

So you're no longer confused when colors are different between metrics or within categories within a visualization. It brings a nice consistency throughout last year. At re invent, we announced paget reporting as neil mentioned, um the ability to burst reports out on pdf or cs v schedules. uh this year, we've added some functionality. So we now also support excel reports. So that's gonna take tables and pivots and keep the formatting that you've added to those. So whether it's group cells, uh total sub totals, um all the width and headers uh as well as we've added an api so that you can create these reports on demand. So you no longer have to be forced to use the scheduling. That's in quicksight. If you want to use this maybe at the end of an etl uh or something like that, you can trigger it whenever you need.

We've also enhanced some data preparation. uh so within spice, we've increased the query performance in a few different ways. um we've added a common sub expression elimination, essentially creating better queries on the fly. so they are faster response times from spice data sets. uh we've also improved the ingestion uh with parallel ingestion with large data sets. um we will actually drop the time by up to 75% for refresh schedules. uh we've also added the float data type for more precision. uh we've had two new api s um to manage the refresh schedule as well as incremental refresh configuration, right? when that incremental refresh happens uh as well as the configuration on the column. so admins um can start to do this at scale, right? that's the point of using quicksight, we can do everything at scale.

We've also added data connectors for starburst, reno and google bigquery. we've added data set parameters. uh so this is the ability to use a parameter from the dashboard and have it flow through back to the actual live query data set. um this helps for complex use cases as well as faster query results because you can put a wee clause within sub queries as you need to and populate that from choices on the dashboard itself.

We've also integrated uh amazon sage maker canvas. so sage maker canvas is the no code, low code uh machine learning platform. you can now take data sets from quicksight, bring them into canvas, uh create a predictive model and then use that predictive model back in quicksight for your data sets. i'm actually going to go through uh a quick demo. again, we're trying to uh show a lot of features in one hour. so i may go through quickly. so if you have any questions at the end, please come up and talk to me.

So here we have a dashboard of customers or telephone customers. and so maybe we want to predict the churn, right? we have this data, we want to create an ml model to predict who is actually going to churn within my customer base. so the scenario here would be i am a quicks site author. i've been shared this dashboard and i then now want to make a prediction. so within the dashboard, i can find my data set, i want to use, i can save this as an analysis so i can start to make changes to the dashboard and i can start to clean up the data right. there may be data in there. i don't want to use for my predictive model. so i can filter, i can change fields. i can bring fields in and out. uh basically prepare the data to send to sage maker. you can see it's taking out some dummy data there. once i'm ready, i'll go hit the option here. i'll build a predictive model. it will take that export of that table and send it to sage maker. so they can then start the model building process.

Once it's ready, i can launch stage maker, i can select that data set. i can start to build a model. you can see you can create predictive analysis, uh image analysis or maybe even text analysis, you can do sentiment analysis or something like that. i'm going to do a prediction when i'm ready. i can pick the field i want to predict here there's a churn, a historical churn field. i can preview the model which will run uh one of the models. so it can give me a preview of how accurate it's gonna be. and as well as which fields are gonna be most impactful on my model. and you can see i can go through and take a look. so there might be fields i don't want to include, i can go on the left hand side and start to take those out, maybe identifiers of that customer. um maybe region doesn't really matter for my predictive model. i could start to remove those and once i'm ready, i can run the build and this will generally take anywhere from 45 minutes to a few hours. it will run through a few different models to figure out which one is going to be the most accurate for my prediction and choose that one.

For me, you could see the accuracy is going to be at 97%. i can again see how impactful each individual field is as well as i can actually see which model it used, right? it can show me all the models that it checked and show me the percentages. now, normally i would use the predict and sage maker canvas and do batch predictions and export this data to cs v. um or i would deploy this as an endpoint in sage maker studio. but with this new integration, i can send it back to quicksight and i can use this model within a data set. i'll download the schema here and once i'm ready, i'll jump back into quicksight.

So now you can see my churn data set, which is an athena data set. so my production data set that continually gets refreshed. i will add a predictive field and i'll select my model that i just built, i'll add my schema files. so the data set matches the fields that are coming from uh the schema itself. you'll see that they match here and you'll see that the name was matched for the churn field. so for the new field that i'm adding on to my data set, i'll name this churn prediction and when i'm ready, i'll refresh the data set. and so now every time that refreshes that new field will get updated with the canvas model, i can start to use that field in visuals.

So maybe i want to break out um the prediction of customers and the number of customers who will churn in an in a specific time frame. as well as the mo most importantly, how much money i might lose with that churn model and i can publish this out to my users and they now have this field ready for their use.

All right, i'm going to pass on the bashi for governance. thank you. um so while both my co co presenters may believe they snagged the shiny and like exciting pieces. um when it comes to the updates of quicksight, uh little do they know that we're about to rule quicksight and that is with administration and governance. so, um with that, um this year, we've been pretty busy um granted, we've been saying this quite a lot throughout this entire week. but um i think we've made a lot of really big strides when it comes to the governance and managing permissions of your assets. and so, uh this ranges from, for example, introducing shared folders, um or shared restricted folders uh to being able to do nested conditions and then as well as being able to find different ways of authorizing connections uh to quicksight.

And so as some of you guys are already familiar with um personal folders and shared folders existing in quicksight. so being able to add that hierarchical structure when it comes to your asset management, but now we're able to do shared restricted folders. so how is this different? um it's a new type of shared folder. it allows you to restrict assets from being removed. and so for an example, if i'm a product owner uh for um this manage inventory management dashboard, and i wanna give one of my teammates the ability to make different edits or changes. um i'm able to give them the contributor rule and what this contributor role does is that it allows that that team member to be able to make different modifications. um so for example, maybe add a new visual um without having to share and move the analysis outside of the folder, um fine grain access granularity.

This has been something a part of quick site for quite a bit of time as you've seen with like multi-tenant structures, our row level and callum level security. but now we've introduced two new updates. and so the first one is run as im for um amazon s3 and amazon athena. and so for this, what it allows us to do is that we can provide an individual user um access to um a certain um s3 or athena data source. and from there, they're able to um be able to access this information, it's able to override the current permissions that are existing and it just makes it really easy in order to give that permission just to that single user versus having to provide it um company wide or just to everybody for that entire s3 bucket or for the entire um uh athena data source.

And so from there, um we've also introduced um supporting complex row level security when it comes to embedded dashboards. so when you use row level security tags, now you can have and conditions as well as or conditions. um so, and conditions always existed, but we've now allowed for you to be able to nest these pieces together. and so it makes it just a lot more easier when it comes to those multi-tenant structures as well as for your anonymous embedding. so if like, for example, um i have multiple roles within um of being able to access certain information. and so it just makes it a lot more easier to handle those very complex security uh data scenarios and being able to just have those multi d dimensional uh views.

And so this is my favorite announcement um of when it came to um one of the features, i think it just makes everything a lot more seamless. and that's specifically uh being able to configure quicksight access with im identity center. and so all quicksight users um so new users can now subscribe to quicksight um using im identity center to enable their users to log in um to their existing identity sources that they have. and this doesn't require additional single sign on capability um or for having to do those configurations in quicksight.

Um another feature that got introduced that is a part of im identity center is trusted identity propagation. so i i'm sure some of you may have heard about this yesterday if you went to the session with camille and ramon. but to give a little bit of um a summary of this, um what this allows you to do is be able to pass a user's identity from direct query red shift into quicksight. and so this just makes that seamless end to end um access for no need to do re aic and then simplifies that role level security um when it comes to that management. and so you're able to, for example, if you have certain role level and column level security enabled in red shift, that is able to translate all the way over into quicksight. and so just removes duplication efforts that you've typically had to do in the past. um and it makes for that consistent um authentication experience.

Um this is something that's also for um aws uh la formation um into redshift spectrum and direct query redshift into quicksight and will soon be introduced for some other data sources. now, last year we had the large announcement about um amazon datazone. and so this is a new data management service um that makes it a lot more faster and easier when it comes to customers being able to catalog or discover and share and be able to discover and data stored. and so today, through private preview, you're able to um do that integration with quicksight. so being able to see the type of data that maybe different um people within your organization have access to and subscribe to it. so we're gonna go through a demo of this to walk through um each of those pieces and see how this integration works together.

So starting off as you can see in this architecture diagram, we have two producers. so we have our sales and then we have our product development team. and so both of these teams have a data leak that they catalog data into their data zone from there. we have our consumer within operations where they're able to take this data and visualize it into quicksight. so the first step of what we're gonna do is being able to publish that data into datazone, sorry, taking a look.

Yeah, we're logging in. um so as you can see, we are a product uh development team right now owner and we're looking at our data. so i'm looking at my product data and so you can see the glossary terms i have below uh being able to identify it. so i want to share this across my aws organization. so i'm gonna go ahead and publish this asset. so by clicking on it, i'm able to publish my asset and make this available across different organizations. so now or sorry within my aws or organization.

Now, um from the contributors standpoint, we're able to search for this catalog and request access to it. so now i'm logged in as my operations. and so here, what i'm able to do is i'm just gonna go ahead and look for my asset

So yeah, I'm gonna click on and now I can see I have one asset available right now. So let's click on it, see what we have. So we have sales data. The thing is I need some detailed information about product information regarding those sales. So, what I'm gonna do is I'm gonna search into my data warehouse and so I'm typing in warehouse and I see there's product data available.

So I'm gonna go ahead and click on that and then request access. And so I'm just gonna be like, please give me approval. I need access to this. Um and so I'm gonna go and ask for um that subscription. And so now we're gonna go back again as a producer. And so, and here we have to go ahead and improve this access. And so how we go about that is we're gonna go into our, yeah, product development again.

Um and we can see that we have a notification and it's asking us to basically um approve this. And so I'm gonna go ahead and all caps be like I approve. Um and then here I'm gonna click to approve and now we've made this available uh for our consumer.

And so here you can see um this is the structure of now, it looks, looks a little different now for a consumer. So for our operations team, we can see we have these restricted folders if you guys remember me mention this earlier. Um and in it, we have our production development. Um so we have our sales data and our pro uh product, product um data.

And then in here we're now able to create different visualizations. So our a our different analysis and dashboards. So next, we're gonna go into building out these pieces. And so here, um, so we're gonna go quickly check, um, seeing if this is available from our operation standpoint.

So we can see there's now two data sources. And so we see we have our product data, um, but it's still loading, it's still, we're still waiting on it. So we're just gonna redirect to our environment. And when we go into this environment, uh we're gonna just go directly segue into uh quick site.

So we're logging into quick site and we're able to now see here that we're able to access our shared folders. And so um as we go into, we're gonna click into our sales data and within our sales data, I'm gonna go ahead and make different, just check out the data that we have within our data set.

So, looking at the title that we have, um and I'm gonna go back and check if we've confirmed that our product data is ready, which looks like it's ready. So now we can go ahead and uh create a join across our two data sources so that um i can show more insights related to my data.

And so here I'm gonna go ahead and add a new data set, um I'm sorry, data source. And then here we can see our datazone connection um being able to access it. And so I'm gonna go ahead and select our product table and here create a join. So I need to make sure to define my uh my key i ds.

And so I'm gonna go ahead and select product um and then do select product again for that commonality and then do a right join from here. Now, I can go ahead and save and publish this data and I've gone ahead uh before and built out this um analysis that's already available.

So it's our sales report data. So you see all detailed information about sales reports, but we're gonna dive into our sales report with our product data. And so this is a combination of both of our different data sources that we have available.

So you can see we have our map, we have this visual. And so justice does a great job of showing how you can provide those permissions for your different users and be able to govern that access.

And so with that, um so with being able to um with re invent last year, we launched the expansion of um different api s. And so this was being able to treat your b i dashboards as code. And so that helps accelerate your dev ops strategies.

Um and so being able to enable import and export packages of these different asset sets into a bundle. And so what we've now done is we've been able to automate this process. And so that is with our new amazon eventbridge.

And so with amazon eventbridge, developers can respond automatically to events in quicksight, such as a new dashboard creation or making a different update and you're able to send emails or um trigger different types of functionality. So whatever works best of what you need to kind of enable to happen.

Um it's possible with eventbridge, another update is within our cost, um our cost allocation tags. And so for our aws cost allocation dashboards, we now allow you to view all licenses as well as your cost center um information where you're able to see your reader session charges of what's occurring um at the user uh perspective.

And you can even do it to the multi-tenant structure with having that um having your data now enhanced and being able to give that complete view. And so this makes it just a lot more easier to understand different monetization strategies that you can do as well as picking the right pricing model um based off of your usage patterns.

And so speaking of visualizing those different cost and usage data, um we're very excited to announce that now we have a native integration with aws cost and usage data with quicksight. And so now um you can launch these dashboards directly from your aws billing and cost management console.

And so what this looks like is um you can see the export page up there and what you're able to do is just click it and be able to deploy these prebuilt um templates of being able to within minutes and so what this does is it allows you to no longer have to worry about kind of tracking that data flow um as well as you can share these dashboards now with other users who may not have access to your aws console.

And then you can even safely embed this into your own website. And so being able to increase different like allow your users to see um like the increased cost or as well as being able to keep accountability across your organization.

And so we also no longer need to manage. Um so automation is always great um as we've seen multiple different scenarios and one of those is gonna be we no longer need to track how much spice data we're purchasing. And so a lot of times you're running into ingestion failures and because you don't have enough spice capacity and so now you can just flip a toggle and automatically increase that capacity for yourself.

Alright, now we're on our last but definitely not our least pillar. And so this is being able to provide api s for everything for our developers. And so whenever we think about api s, of course, we are thinking about the embedding experience and our embedding use cases that we have.

And so with that, we've done so much that like this is probably the greatest area where we've done so much development. And to the point where we actually changed, we added a name called two point. Oh, because so much uh changes with our javascript library when it comes to the integration as well as functionality.

And so one of those is for an example, being able to increase different productivity. So having that support for typescript, um we're also helping with just that building out that analytics experience quickly. And we've done this through, for example, being able to provide pre loaders as well as um having air handling configurations of being able to give infos and warnings, uh being able to manage your redo and undo buttons, uh being able to do a lot of different rewrites when it comes to uh the, for example, like status type checkings, being able to enhance run times.

Um and many more, especially when it comes to our different call patterns. Um something that's really great um is also being able to have that state persistence and bookmark um within um the embedded experience. So now you're able to, for an example, if you as a user, um you have certain views that you wanna have to stay persisted, that's possible with the embedding experience.

And so you no longer need to. So if you have this preference and you're presenting and wanting to show pieces, you don't have to rely on having to always add on those filters on the fly while you're presenting instead, you can just have those already in pre and so, um and lastly on this slide is um we have being able to synchronize and so having our runtime them and filtering and so what this does, it just makes it a lot more easier for on the fly, being able to make different changes.

So for an example, when it comes to themes, you can with quickly within a button, be able to turn it into dark mode or light mode, you're also able to just have different api s to trigger this functionality as well as for filtration, having it within your home application.

And so being able to make those changes on your dashboard within your host application, we've taken it even further. So did even further enhancements when it comes to your embedded experience. So we've done deeper integrations in the sense of being able to do custom action callbacks.

And so with custom action callbacks, you're able to build custom workflows, be able to have business logic so that your users can um in an instance, be able to take action from the application.

So for example, if a dell developer is registering a data point callback on a spe specific slice of a of a pie chart. And so when a user maybe clicks on that, it's able to pass information about that data point.

So whether you want it to, for example, be able to like send an email or you want it to just capture and process that information and be able to update um that status within your database.

So if it's related to, for example, if you click on this, it, it adds um when you click on it, you're adding another value to it. And so you're increasing that pie chart now or being able to track those usage patterns within your users. So you're able to do that with those callback actions.

And for last year, we launched the quick site community. So how many people are members of our quicks site community? Oh, that's so little. Ok. We have to change that. More of you need to join because we've now launched the um quicks site arena.

So with this, um it makes it we have this provided this embedded instance where it allows you right within our community to be able to publish different dashboards. So show off your skills that you've done with building all these different dashboards that you have.

Um also if you have questions that you run into like the point of the community is to help you with being able to ask different questions, being able to learn different how to videos as well as just gaining more knowledge on quick site.

And so with that, you're able to put up a sample dashboard and be able to log the issue that you have. And um from there, be able to post into the community and you can have people be able to see that sample data and be able to understand it a bit better versus when you feel like this calculation is not working.

They're like, i don't really understand your data, you're able to instead provide an instance where it's available and it makes it so much easier for being able to receive that help.

And so i'm about to go through quite a few qr codes now. So get your phones ready because there's quite a few.

Um and so for one of those, um if you're curious exploring more about jared b i pieces that neil covered, which was the most exciting, of course, um that we've experienced this week. Um i recommend scanning this qr code, this will have all of our latest updates um and all the exciting news related to it.

Um then afterwards, if you want to check out demo central. So for an example, um doug was presenting all these richer visualizations that we've added within this last year. And so if you want to see some amazing dashboards that our team has built out that incorporate some of those visualizations, i highly recommend clicking on our um scan this qr code and going to our demo central website and you'll see some of those there um for signing up for a quicksight community.

So this is probably not the last um qr code you're gonna see all of today, but at least it's the last one in the slide deck. And so um i think this one's definitely the most important, especially which i saw how little of hands that were up that have signed up for the quicksight community.

So here is a great place to interact with other customers. You can hear about upcoming events, different um if like clicks sites in town and we're doing like a local no code event, which is um we've done a few of those this past year. And so, and being able to hear like we have these learning series and so just a lot of new updates and i just, i really recommend going ahead and um signing up to be a part of this community where you can ask questions, be able to learn together and get the latest updates on quick site and now um our quicksight community.

So another piece of being able to learn within our community um is our developer corner. And so we just released this, this, this year. Um and at the second half of this year, um and it offers a centralized access of being able to learn about the quick site embedding analytics capabilities.

So a lot of the pieces that i covered today, you could be able to look at different examples of those, be able to learn about different videos. Um i think all the blog posts related to you can access. And so it's just a really helpful place of just knowing where to get started when you want to embed your applications.

And then lastly, um i want to take a moment to celebrate our top 10 quicksight community experts of 2023. Um they've been gone above and beyond in order to help with a lot of answering questions. And i know we have one community member um here, i don't know if he's here right now in the room, but um we're really grateful for all the work that you've done, especially just answering questions, creating content and just sharing your expertise with others.

And just really excited to see the rest of you join along as well and become experts yourselves.

And so with that, here's a last few, couple of things. Um so we have two remaining events um from the quicksight perspective. So we have a lightning talk on generative b i um i think it starts immediately after this. So just start running um um as for an explorer, we have a repeated workshop. So being able to build dashboards and reports um exploring generative b i um yourself. And so that's our last two sessions that we have.

And then lastly, um thank you so much everyone for joining for our session. Um it's been a great year for quicksight and i'm really excited to see where we chart to next.