Add Looker-on Ssn with airSlate SignNow

Add initial looker-on

[Music] hi this is Annika from liquor today I'm going to show you how to get started deploying liquor in your organization by connecting a new database generating and customizing a reusable data model exploring the data through that model and then sharing my new findings the first step to exploring our data in looker is going to be to connect database here I'm looking at the admin connections section where I can see a variety of different databases that have already been connected or out a new one lugar connects exclusively via JDBC and writes sequel queries to databases the complete list of dialects to which liquor can connect is available here you'll notice this includes both relational and non relational sources for example if we look at Google sheets or something like spark sequel this is an ever-expanding list as well once we've connected a database we're ready to generate our first model I'll go to develop and select manage locomote projects and from here I'm going to create a brand new project a project is something that's going to stick with us it's something that my co-workers and I are going to build business logic into that will collaborate on it's going to be version controlled and it's also going to serve as a single source of business truth for any end-users or technical users who might interact with liquor in the future so let's just call this e commerce since sneak Commerce data set I'll point it to our demo new events data set and I'm just going to generate all of the tables in here for now I can also choose to select a single table or select schemas what lookers doing is I'm generating this is scanning through all of the different tables and columns that exist on this database connection so if we click into order items for example we can see that it's found an ID column it's found an inventory items column it's also found some time stamps and is automatically breaking these out into a number of different time frames these are going to be very flexible for me to use and I'm not going to have to remember how to get a date or a quarter out of a time stamp anymore in addition to all the columns in the database innately it's also generated one aggregate for me account star if we look at the model file lookers also gone through and identified some relationships between tables this should save me a lot of time writing joins even if I need to specify some changes to some of these maybe users needs to be joined on a different key or additional conditions need to go into here everything we're looking at whether it's an individual tables or in the model file is completely customizable and then the logic I defined here will be shared across others who interact with the data model so before we start modifying anything here let's jump to the explore section to begin interacting with this data I'll click Explorer and select which table I want to start writing Riis from in this case let's talk about order items so I'll click that and it'll open up the explore page for me the explore page is where analysts and developers and anyone who's interacting with the data is going to be able to curate their own questions and interact in real time the fields that we're seeing appear on the left are just the initial generation we're gonna customize and curate what we see here but as we can recognize from the order items table is an inventory item ID ID sale price this count that we talked about as well as the different time frames that came through the time stamp so let's write a simple query let's start out just by selecting count this is an aggregate as we can see it's under the measures section instead of the dimension section which are not aggregates if I want to see the sequel that's about to be generated to my database I can click the sequel tab here this allows me to validate any reports that I might generate and looker against my existing reports I should be able to find any discrepancies and track them down to their source so let's click run and when I click run here lookers directly interacting with my database and running that query we can see it just returned one road taking zero point two seconds let's start to break this out by more things so let's group this out by date and let's jump back to the sequel tab what we can see is happening here is it's taking the raw order items created at timestamp it's running a convert time zone function on that which I would have specified on the database connection and then running a date function Lucar already knows how to generate different time frames for a number of database dialects you won't need to customize these and just build in the logic for everyone to reuse let's also add a filter so let's look at 30 days and give this a run switching back to the results tab I can click on any of these headers to sort the order that these results will appear in and then I can start popping open my first visualization keep in mind we haven't done anything to customize yet this is just what we're getting straight out of the database without needing to export any of my data I can quickly toggle through a number of different visualizations and if I so choose I can even import my own custom JavaScript visualizations to make available to all of my users next let's add a pivot pivoting is something commonly done in Excel but often really tricky to get directly out of a database in this case I want to see what status my different order items are in so I'll click pivot here and we'll see the distinct values of the status column appear across the top and then appear as nice series in my visualization let's stack these on top of each other to make them a little bit more readable now everything I'm seeing here is interactive so for example if I don't care about it cancelled orders I can just toggle that out of this visualization or I can quickly and easily add another filter to remove that from the query altogether so maybe I want to start interacting with this a little bit more for example if I'm wondering about some of these orders that shipped all the way back on the 2nd of February I can click on that and I can see what those individual orders are we'll also notice that I have an option to drill into a smaller time frame because I use this date time stamp by default smaller timeframes like time are being made available for me to drill into we can see that these are the 10 orders that haven't been made available yet and this isn't telling me a ton about why those aren't available so let's explore a little bit more why don't we look at the distribution center that these came from something automatically generated here and let's also add in the state that they're being shipped to this might tell us a little bit more about why those orders haven't been delivered yet from this table and we can see a lot of these are having to go all the way across the country and perhaps that's why they're taking longer to ship if I want to share this with somebody else who can then make a change to our operations around where orders get shipped from from distribution centers I can just grab this URL and send it to my coworker who's going to be able to then take this and make action upon it they'll jump right into my context they'll be able to alter statuses so for example if they want to see also processing orders that haven't been delivered yet they can alter the filters that I've jumped them into as we can see here because everything is coming straight out of the database I haven't lost any granularity and I can continue to add ad hoc questions that I might want to interact with here let's jump back to our visualization we created besides just sharing the URL I have a few other options to get this in front of other people if this is something that I just need to send to somebody as a one-off and I don't really care about them getting access to it again I can just download a text file a CSV Excel even a PNG of the image itself this allows me to just quickly get the data in front of somebody even if they don't have a look or login now if I want to do this on a regular basis I can also start to schedule it out so let's save this piece of content I'm clicking Save As look we'll call this borders by status and not to save this into my own personal workspace having saved this piece of content this look has its own page on here I can alter filters that have been placed I can see the underlying data or even the SQL that's under this overall query or I can also start scheduling it out so let's click create schedule I'll select that I want this to run once a week and let's have this go out on Mondays to my executive team this email will now appear in their inbox and I can select whether I want this to go out as a table as a CSV as a JSON as an inline visualization and I can also determine I don't want this to go as an email but instead I want to push this out as a web hook or dump this to Amazon s3 at a regular basis by scheduling this out I can quickly and easily get data in front of people and their existing workflows without there having to necessarily log into the looker platform now if they do have a login they'll be able to jump back into this context and start interacting and drilling with the data just like we saw before one of my favorite ways to use the scheduling feature is actually as an alert I can set this up to run really often as often as every five minutes and only between business hours or all the time and I can have it only return results if there actually are results and if they've changed since the last run so for example for this e-commerce business if I wanted to track inventory or if I wanted to track new user signups and we fell below or above a certain threshold that I wanted to pay attention to I could schedule this out to only send me an email if certain conditions are being met and have that come directly to me or even pushed a text to my phone directly now that we've spent some time exploring and sharing this data as our end users might let's start customizing the model and see the benefits of building in or this logic and having it become usable for an end user so we'll go to the develop menu and select the project we've been working on keep in mind that look Amell is not something that your end-users are going to have to work in or even know exists this is just what writes the sequel on their behalf and allows them to have a really flexible efficient experience exploring data within looker as our first custom field why don't we start thinking a little bit about our users we have a first name field and we have a last name field but realistically we're going to want to see these things done together so we'll create a new dimension and just call it a name now the definition of this field is not just going to point to an underlying field like some of the ones we've seen before it's actually going to draw from logic of two subfields so for example here we're going to take first name and we're going to concatenate that with a space and last name in this equal bar here I'm really just writing sequel that my database responds to in this case we're writing to Amazon redshift so I can use this can count operator and other dialects on my trekking cat or a different function this allows you to completely utilize the features that exist in your database without losing those and being dependent upon an engine that we would provide so I'll click Save I'm gonna run a validator that's going to make sure that all my references look good it looks like we're okay now why did I use this substitution operator instead of the ones we have here referencing these raw underlying fields in this case if I were to start customizing the definition of first-name for example if I only wanted to return the first letter or if I wanted to just capitalize the first letter of the name I would apply that function here and now when I go through and reference that in redshift it's going to recycle that logic and I won't have to write it again this is a really simple example but you can imagine that as you start to build upon this model you're gonna have changes that build on each other and you're not going to want to have to keep going back through and rewriting a knit cap every where you reference name let's do another example let's actually add a new measure which is going to be the total of sales price that we've sold so I'll give us a name again and we'll say I want to have total sales price be offered as a new metric this is going to be a type sum and I'll define this as pointing to the underlying sale price field I have available the last thing I'll specify is drill fields we saw drill fields before where we clicked into data and saw things underlying it this was a randomly generated list of things that we saw before let's actually specify what we want to see when we click into total sales price in this case I'm going to want to see the ID of the inventory item I'm going to want to see the ID of the order I want to see user ID and let's also reference that new field we created users name let's jump back into our explore section now when I expand these will see that the new fields I've created exist within this field picker so if I select a name for example that's available here indistinguishable from any other field and we can see that the definition of that is taking a knit cap of users first name concatenated that placing a space between the referencing users last name so we still have visibility into exactly what liquor is generating behind the scenes let's see this broken down by the total sales price these individuals have sold so we can see total sales price appears here under order items we can also see that liquor is making this necessary join out to users in order to write this query so let's give it a run now by sorting on this column we can see the highest sales price for a given users name is John Smith perhaps we have a couple John Smith's here so let's actually count these out so now if I want to drill in and utilize that drill path that we defined in the look of Mel section all I have to do is click on this value and I'll be able to see the orders and the details about them that I specified of Edward Hausman let's create a query that tracks our top 10 users for the past 30 days so we'll say only show me orders that were created in the past 30 days I'll add user ID so that we're deduping our users but I'll hide that from the query so it doesn't necessarily appear when I'm looking at it and we can compare total sales price to users count perfect let Sun combine these axes so we can see the different values now that I'm happy with this visualization let's actually add it to a dashboard so similar to before we'll save this new query and say top users by sale price save it into my personal workspace once again and let's actually add this onto a dashboard to add this to a dashboard I can either you select an existing one or create a brand new one let's go ahead ahead and make a new one called a new dashboard and click Add to jump there I'll just click the link and I'll be able to start building up something that others will be able to interact with let's add one other look orders by status and now I can simply drag around the different tiles on my dashboard to make them fit nicely I can also quickly and easily add a filter to this dashboard in both cases we're filtering down to the past 30 days but if I want to make this dynamic I'll just add a new filter called date I'll set the default to be 30 and I'll tell it what to put in the where clause of each of these having added that I'll now have this filter appear at the top of my dashboard and if I just want to look at the past 14 days I can quickly interact with that this dashboard just like anything else in looker can quickly and easily be shared with other users by grabbing the URL or I can also schedule this out as a PDF or as an inline set of visualizations if I want to get this in front of somebody who doesn't have a looker login thanks for watching our demo I hope you saw the fresh new approach we're taking to analytics there's a lot more we'd love to show you so send us an email or give us a call and we'd be delighted to set you up with a more in-depth demo or set you up with a free trial in your own [Music] you [Music]