Exploring & curating data
Dashboards, Looks, Visualizations— it all happens here
Hi, I have a dashboard(d2) which has several measures and that gets called in from drill down menu from another dashboard(d1). Depending on what measure user clicks on d1, I am able to pass those values to d2 and the dashboard d2 displays required data. My problem is that I want to sort d2 rows by the measure that user has clicked on d1. can someone please share some way to get this done. Thanks
Hi all. My company users Looker to manage messaging and push notifications data for our app. Please forgive me for lack of clarity/detail in explaining the issue - I’m a new user with no training, and only a few hours of self-guided learning/discovery per week. I’m trying to create a report that shows how many users have opted out of push notifications on each day within a specified time frame. Right now I’m filtering on the Dimension - Tag Event Notification Opt-In is equal to FALSE and also the Dimension - Tag Event Tag Action is equal to ADD In the results I’m seeing, some users are being repeated every day, which I understand because every session, they’re reporting their opt-in status… but how do I get my report to return only the day when their Tag Event Notification Opt-In went from TRUE to FALSE? Thanks much
Hello, I am designing a dynamic reporting where I compare several geographic locations and how they evolve through time (e.g. every month for the last 6 months) I would like to create a column with the CAGR (compound growth rate) in the end to understand the trend easily and compare it with the target (easy via conditional formatting) and identify areas to prioritize Is there a way to do that numerically and not by the trend visualization tool? I am putting all in a data table because with lots of data it becomes hard to understand with a graph Ideally the CAGR function would recognize directly the number of weeks/months/years (N) and calculate automatically the growth rate and display a number This need is recurrent as I always try to understand the trend in many different situations and with often a heavy dataset. I am sure plenty of people have similar needs *do not need to be in years, N as time periods (day/week/month/year/semester etc.) would be great so it can be adapted to eve
When creating a dashboard it is important to understand how dashboard global filters override filters that you have saved on your Looks. When applying filters to Looks in the Explore Section, they will be overridden by global dashboard filters as long as they are on the EXACT SAME field. For instance: In this example I have applied a filter to Orders Created Date for the past 30 days in the Explore Section. In this first dashboard example, we will apply a listener for the dashboard element on Orders Created Date. In this case you can see that applying the SAME filter has overridden the previous “Orders Created Date in the past 30 days” filter and is now scoped to view the “Orders Created Date in the past 120 days.” The Orders Count has returned “4,332”, which is the count for the past 120 days. If instead we globally filtered on Orders Created Month, this would not override the Look filter of “Orders Created Date in the past 30 days.” For Example: In this example you can see
I have created a look, with a calculation that extracts a certain amount of characters (VIN numbers). I need to take those VINs and put them into another look. Is there any possibly way to automate using the results of the calculation as a filter for the other look?
I have this following table as a result of a look: |Server Domain||aternity.com|azure.com|box.net|clicktale.net|google.com|homedepot.com|live.com|microsoft.com|pubsub.googleapis.com| |---|---|---|---|---|---|---|---|---|---|---| |Time Date|Total Hits|Total Hits|Total Hits|Total Hits|Total Hits|Total Hits|Total Hits|Total Hits|Total Hits|Total Hits| |8/20/2018 20:00|42948|11946|4507|14426|1194|4198|18761|885|1215|12299| |8/20/2018 20:01|36749|11651|4509|14216|1421|4714|9612|1200|1180|6175| |8/20/2018 20:02|36032|11593|5373|14217|1691|4247|13781|1592|1302|7283| |8/20/2018 20:03|37670|11491|5706|14328|980|3947|10709|985|1126|5242| |8/20/2018 20:04|35879|12027|5416|14181|1280|5463|8511|871|1166|7456| |8/20/2018 20:05|40239|11097|5089|14282|1304|3983|9396|1063|1145|7349| |8/20/2018 20:06|36231|11956|5207|14535|1097|3939|8105|935|1186|13474| |8/20/2018 20:07|36692|11210|4435|14413|883|4329|11084|1095|1290|5290| |8/20/2018 20:08|49245|11609|5664|14364|911|3829|11824|884|1277|9763| |8/20/2018
I need to create aggregated measures that compare a measure’s value within a given cell against a larger contextual aggregation. The simplest example is a percent_of_total_sales measure that just takes the sales_sum measure for a given cell, and divides it against the overall total sales_sum for the entire table, so that I could go from values like this: to Obviously, that’s not all I’m just looking for – what I really need is the equivalent of the DAX ALL()-related functions, that allow you to selectively ignore any given dimension used in an explore. From what I’ve been able to find so far, you can do something like this with ‘positional functions’ in table calculations, but that falls short on at least two important fronts: I need to be able to use these measures freely in multiple explorers and dashboards, so adding and maintaining the same table calculations into every single one isn’t r
I have a dashboard where it makes sense for some charts to be connected to some filters and another group of charts to be connected to a different set of filters. However, because all of the filters are at the top of the dashboard, it’s not clear to users which filters affect which charts (unless you dig in to the settings each time). Is there any way to make it clear which filters will affect which tiles, and even better, can they be grouped together with the tiles themselves so that it’s visually easier for users to comprehend (like is possible in other tools like Chartio)?
Hi, What’s the best way to set default, initial value for a dashboard filter in embedded mode? Regular dashboard filter seem to allow setting static values only. I am wondering if there’s a way that embedded dashboard can pre-populate initial filter values with something that’s either received as URL parameter or looked up dynamically from the database. For example, each SSO user may want to open dashboard with his own preferred currency . Thanks, Matt
Hello, I might have missed something but besides using the matches to filter, “2018,2017” I do not see how you can compare to previous year easily. I get the full year of 2017 and YTD 2018 which are not comparable. I would need to be very specific entering exact dates. What would be the most efficient way to compare YTD N to YTD -N ? I found a thread on how to implement the YTD but I am stuck on how to apply it to a LY analysis? Month to Date and Year to Date Analysis Analytical Patterns Month-to-Date (MTD) and Year-to-Date (YTD) analyses are useful when conducting timeframe-based comparisons. For example, while you are part of the way through a month, you may want to compare your progress with earlier months. However, it would not be useful to compare part of the current month with the entirety of earlier months. Instead, it would make more sense to only look at the days in earlier months that have already transpired in the current month. The best way
Hey, I have a measure of date time which I would like to format as a time within a chart. Here you see my measure 🙂 I have been referencing the material on time formatting for charts https://docs.looker.com/exploring-data/visualizing-query-results/time-formatting-for-charts. From what I understand I should be able to use “%H:%M:%S” to return only HH:MM:SS of my date-time measure, however when I try this I receive the below message: 'Invalid Format String Format strings follow standard spreadsheet format using ECMA-376 spreadsheet format codes.’ Any pointers? Cheers Will
Hey guys, I’m new to Looker and trying to learn a lot! I wanted to query one of my Explore data sets and be able to search for all IP Address connections within a CIDR range. Is this possible to do or will I need to convert my CIDR range to all of the individual decimal IP addresses that make it up? Thanks!
Hi! I’m trying to figure out what would be the best visualisation option in Looker for this specific use case (budget and invoicing for projects). In a single viz, I’d like to see: Target budget Value of work performed Actual amount invoiced In a different dashboard tool I’ve done this using a bar chart (for the value of work performed) with an overlay of shapes to show the target budget and amount invoiced, but Looker does not support this type of chart, unfortunately. The goal is, in one chart, to view where if we’re on track on the budgeted amount, and whether the actual invoicing is on track with the invoicing that has been performed, and therefore easily spot projects that are over budget or under-invoiced. The amounts could be individual dimensions (single row), or measures for aggregation. Any creative ideas? (This can involve a custom vizualisation if you have a code example).
I checked whether or not [suggestable] was working on a filter in the [explore] option, then exported that chart to a dahsboard. In the resulting chart, I cannot use [suggestable] in filters, which is exactly what I did want to do. As I understand it in the explore, when “equals” is used as the option it should run a sql DISTINCT on that dimension in the database and use these to populate suggestions. Sadly, it doesn’t. Any suggestions?
Hi All, I currently have data of users and their associated segment at specific points of time. Data looks something like this (sorry I can’t put more than 1 pic) User ID; Time; Category 1; July; Segment A 1; August; Segment A 2; July; Segment A 2; August; Segment B For example, User 2 can change his/her status from Segment A to Segment B from July to August. What I am trying to achieve is the percentages of users from each Segment converted to other Segments by time. It should look like this. So far I haven’t found a way to build something like that. Would really appreciate helps and inputs from you guys. Thanks!
When a user places their cursor in a field filter in a dashboard, a loading dotted line (similar to a loading bar, but it’s a line) appears under the input box and eventually a drop down of values appears. Most of our users do not know that they are going to get a list of options with that filter. I was wondering if we could add a drop down arrow to the right of field filters so users know they can select from a list of options.
Dear Looker Support, Since I have not been able to find a different place to submit bugs, I will place it here. When selecting a Stacked visualisation, changing the ordering in the pivot dimension and show the ‘Totals Labels’, and then changing the Series Order to (in my case) Ascending, makes the Total Value to show up above the bottom bar. I guess the total value is applied before applying the Series Order, which should be the other way around? Kind regards, Daan
Already have an account? Login
Login to the community
No account yet? Create an account
Enter your username or e-mail address. We'll send you an e-mail with instructions to reset your password.