Exploring & curating data
Dashboards, Looks, Visualizations— it all happens here
Hi all, I have the following problem: I want to schedule a Look on daily basis. It should be a visualisation (single value) which is based on a Yes/No variable per day, where the YES/NO variable is dependent on a measure. This measure is shown as a comparison value. Example of the data: Date | Completion Rate | Is there an issue? 05-01-2017 | 100%| NO 04-01-2017 | 70% | YES 03-01-2017 | 50% | YES 02-01-2017 | 90% | NO 01-01-2017 | 80% | NO Example of the viz: NO 90% completion rate What I want to do: I want to send out this Look (the single value) ONLY when Is there an issue? is equal to YES. When this is not the case, I don’t want anything to be sent out. What I tried: I tried only displaying the first row of the data (hiding all others) and then the viz only changes when a new day with YES is entered (hide “No’s” from viz is enabled). However, the data/results do change every day (even though the viz does not change on a daily basis), and hence, it will send out a new e-mail everyday
Data is powerful, but can be overwhelming. Especially if you don’t have a background in databases and/or analytics. This is why many self-service analytics projects become frustrating for everyone involved - users are unsure about what to do with all of the data, while analysts can’t understand why users aren’t digging into what they’ve built. As a pre-sales team at Looker, we’ve worked with many companies from a multitude of data backgrounds to get them up and running with Looker. From working with these customers and prospects, we’ve seen over time what makes (or breaks) an end user’s experience, both with Looker and with data in general. Here are some of our best practices. Start with Your User In order to provide real self-service analytics, you should consider your different users and their goals. Here are a few guiding questions to get you started: Some Guiding Questions Who will be using the data? How will they be using it? How can you make their job easier? Where would you l
Hi all, I have been looking around and have a feeling that there isnt a solve yet, but figured Id give posting a shot in case there is a looker wizard out there. I am looking to do a running total of orders across multiple columns. This works well, but the problem arises when you pivot by month, using the filter Sales Order Date: “this month, 12 months ago for 1 month” Basically I want a comparative running view of the 2 months side by side, but keep getting all months in between as nulls. I really just want the 2 and to exclude all months in between. I was hoping the “Hide ‘nos’ from visualization” would work but thats only for rows I believe. Anyone with ideas for a workaround besides a dashboard with 2 tables? Id like them to be together. Thanks!
We often report values for multiple different currencies within a single unified ‘Cost’ field, with the associated currency contained in a separate dimension. When we know we’re only reporting on a single currency, we might wish for the currency unit to be displayed in a table visualization. Ideally we’d like to do this by having users define value formats in the same way that they are defined for single value visualizations. So for example, I’d like to be able to define a value format of "£"#,##0.00 and, instead of seeing the unformatted numbers, see: Is this something that could be considered for a future update?
Hi! I created an master report dashboard and then created 3 copies of it to filter for different locations. In the when I update the filter in one the original copy gets it’s filters updated as well! Is there a way to diassociate the filters in the copy from the original versions?
I would like to plot table data such that only the pivoted months and corresponding max value for each month’s column is shown. So far I’ve only been able to plot the individual days along with max value for the whole month. Is hiding individual days (the rows) possible? It would be okay to either plot “2017-01”, “2017-02”, etc. or the actual date of each month’s max value (e.g. “2017-01-29”).
I want to create a visualization based on a calculation but each row is showing up in my visualization. Field “Item Name” | Calculation “Category” | Count Apple | Fruit | 2 Orange | Fruit | 4 Onion | Vegetable | 1 Banana | Fruit | 2 I would like to hide the Field “Item Name” and get Calculation “Category” | Count Fruit | 8 Vegetable | 1 Instead I am getting Calculation “Category” | Count Fruit | 2 Fruit | 4 Vegetable | 1 Fruit | 2 So far I have a Label calculation that sets up my category: if( row() = 1 , “Fruit”
I know there is a way to create a link so a user can hyperlink from a lookml dashboard out to an external spreadsheet but I can’t seem to find the details. For example, let’s say I have a table w/ a bunch of marketing cost information. Users can clearly download or explore on that table, but in some instances I want them to be able to link to an external excel spreadsheet that has data definitions or other details that are useful but not easily included in looker. I’ve seen other users create a link w/in the table to external excel sheets along these lines.
I work with a lot of dirty user entered data. I would like to try to bucket all of my results into a cleaner report. Something like: Report all records where vendor_name contains ‘Agilent’ as ‘Agilent Technologies’. I can ask for all of the iterations, but I’m having trouble bucketing all of the results into an " as " statement. Is there way to build a customer filter or alter my report?
I looked in the Excel documentation as well as other Looker resources and didn’t find anything on how to format Y Axes values in visualizations with multiple series that are not combined. I tried “#,##0 | #%” for example, and a comma delimiter won’t work since that’s included in some of the formats. Is there a different delimiter to use between Excel style formats to make this work?
I have a Visualization that features three sets of data. The first set has values ranging from 0 to 10,000. The second and third have the same range, between 0 and 50. I know that under Visualization > Edit > Y > Orientation, I can put in left, right, right to show the first legend on the left, and the second two (identical) legends on the right. I am interested in hiding the third legend, since it is completely redundant with the second. Is there a way to do this? I was hoping for a hide option in the Orientation field, but that isn’t there.
I have a visualization that plots 3 data points: sum, weighted value, and count. When I plot all three, it creates three different axis to measure them. I would like to combine the sum and weighted value axis because they are both dollar values and have the count value separately. Below is a screenshot of the visualization I’m working with.
Sometimes, when I edit a look, the query keeps running (or so it seems). When I press Stop, nothing happens. So, what I do is, refresh the entire page, and lose all the changes I have made and re-apply the changes. Just wondering, is this a known bug? When can we expect the fix to be deployed? FYI, the database that I used is Google BigQuery. Thank you.
Anyone have any hacks to reduce the size of PDFs from downloaded dashboards? I am using a table visualization and most of my PDFs are coming out to 1.5MB - 2MB each. I am then combining multiple of these into a deck. So it quickly adds up to 30+MB. I have tried optimizing/shrinking using Adobe PDF and Nitro PDF but it doesn’t shrink it very much. Anyone else have any luck with anything else?
I’m trying to create the following table: Category | Dec 2016 | 3Mo Benchmark A | 200 | 230 B | 145 | 189 C | 235 | 300 My approach is the following - filter: 5 months ago for 4 months. Pivot above table by Month and got the following header: Month | Sep | Oct | Nov | Dec | BM (table calc) BM is equal to average of (pivot_index(events, 1) + pivot_index(events, 2) + pivot_index(events, 3) ) But I can’t hide Sep, Oct, Nov results so that only Dec and BM columns are showing. Is there a better way to get the top table? thanks.
We have created looks for the Users and , we need to set up a process to schedule and download the entire data. Is there a way possible,? I checked for the option, it has row limits but i want to explore options of downloading the entire data for a look which is scheduled. It would be fine just to download the look, even without emailing but in a scheduled way not like running and waiting for the query to get executed.
I’ve been trying out PDF exports for dashboards, and for some reason the resulting elements shrink. The results are the same if exporting from a space or a LookML dashboard, and by email or download. When I let four tiles sort and size themselves (as when exporting from LookML), they end up center-top aligned on the (landscape) PDF page, in a total area about 1/2 as wide as the page and 2/3 the height. Does anyone have any thought about why they’re not expanding to fill the page? When I look at the same dashboards via the web interface, they tile and size normally.
This is definitely something I’ve run into, it makes it hard to show people graphs by short time-spans over which there may be periods with no entries. But to be clear, this is not available yet? Only in Looker 4.0 when we have that? Definitely something I would like to see soon!
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.