The helm chart supplied with Sisense allows global annotation of all service accounts within the sisense namespace. For example I can annotate an AWS role that is able to be assumed by pods: helm upgrade sisense ${sisense_chart} --namespace sisense \ --reuse-values \ --set global.serviceAccountAnnotations."eks\.amazonaws\.com/role-arn"=${pod_role_arn} While this covers most of the pods by updating all the sisense-* service accounts in kubernetes the ec-build pod (the one that actually builds the elasticube) only uses the default service account in the sisense namespace. A workaround is to annotate the default service account: kubectl annotate serviceaccount default eks.amazonaws.com/role-arn=${pod_role_arn} --namespace sisense But it seems odd that the pod would a) assume the default service account when all other pods have service accounts explicitly defined for them and b) despite having gone to the trouble of wiring up global service account annotations in helm for them not to be applied at the final step of elasticube building. This specific use case is for supplying web identity tokens to a JDBC driver so that you can interact with AWS by using a specific OIDC assumed pod role, rather than using an EC2 role or worse, static credentials in a file or connection string. Even more specifically using CData JDBC parquet driver with S3.   ... View more

Currently, the "Upsert" build behavior is only available for Build to destination. It would be great if we were able to utilize this feature in our Elasticube model where we are not building to destination. We have numerous tables containing created date as well as updated date that are attached to an id field but are unable to use the updated date field since this will bring in duplicate (or more) rows per id which we can not have. Ranking duplicates and filtering out is not a sustainable with the size of our datasets. The ability to look at the id and updated field would solve this issue without the need for a complex workaround in our data source (Big Query). Thank you! ... View more

Hi, Currently in the filters, we are unable to directly filter a comma separated list of entries. The options available are individually selecting the members, however that can get tedious if the list of entries we need to filter are large. For example, filter more than 100 entries. The other option is using the custom option in the filter. However, that too requires JAQL code which might be difficult for non technical users.  It can be useful to have the option to directly copy paste a comma separated list of entries in the filters instead of multiselecting or using the custom JAQL option.   ... View more

Currently, if we export the pivot as an excel/csv, there is no indication of the date range it was run for in the output file. It will be really helpful to have control on the exported output in excel or csv, such that we can include the date range of the result or other filters applied in the output. ... View more

Is there a way to add dynamic values inside of the text boxes that can be added to Dashboards? As far as I can tell, this is strictly static text. Maybe through scripting or something that is not obvious? If not -- this would be a useful feature to be able to link values to, for example, Dashboard filter values (records from "last 90 days" or "last 30 days") according to what has been configured. ... View more

The number of source of frustration and confusion amongst our users, when building Dashboards, is conflicts/inconsistencies between Dashboard and Widget filters. I have separately suggested that Sisense introduce a visual indicator of widget-level filters and it sounds like that is planned for the future, which is great. Another related feature that a user suggested was a "reset all widget filters" button which would be a single place at the Dashboard level. If the user clicks it, it forces all Widget filters to (a) be deleted (or maybe disabled?), and (b) puts the widget filter settings back in-sync with the dashboard (meaning if the user has selectively toggled off Dashboard filter inheritance at the widget level, that will revert back to the defaults). This is a really quick and easy method to put users back to square one if they have painted themselves into a corner, so to speak. ... View more

The list of our shared formulas can grow to be quite long.  We would like the ability to search within formulas, just as you can for attributes. ... View more

Currently, we are using below Rest API to generate a daily report to know the users, groups, roles etc.  /api/v1/users/get We need an additional field (last_password_change_date) for reporting that would tell us when the user changed the password. ... View more

When the sisense user is deleted the email of this user is still in the recipients list for a report in the Report Manager. If the report Manager is configured to send the reports to external customers, the user who is no longer with the company will be receiving the reports with dashboard data as the reports for the external users are generated using the owner or an additional user. Thanks. ... View more

Currently, ReportManager is only availbale on the system tenant, not on additionnal tenants. Could we have it for every tenants ?   Thank you, Franck ... View more

Scenario: embedded Sisense By default, when creating a new widget, Sisense puts the user into the widget "Wizard Mode", which then requires the use to click "Advanced Configuration" to view the full-feature widget editor. We don't want our users to use "Wizard Mode" and to always be placed directly into "Advanced Configuration". This should be a configurable setting at the Admin level. ... View more

When we are building pivot tables, sometimes there is a need to sort things by the columns. If I want to show a pivot table with a Team Lead as the rows, Total Spend and Total Volume as the values, and Weeks as the the columns, I have no way to sort this to show the date in descending order and keep the table structure the way I want to see it. I think it would be helpful to have this option. ... View more

When adding a dependent filter, the only [available] lock option is a group lock. It is more useful having the ability to lock the primary filter but allow for dependent filtering. For example, if I lock an audit title, I want to retain the ability to filter specific question(s) within the audit. If both are locked as a group, I have to ensure end users know how to filter through the widget, which can be difficult if the full text of the question does not display or the user is not proficient. ... View more

When using dependent filters and selecting new filter values this automatically resets all filter values already chosen.  Whilst this is desirable from a security perspective, not all models have security applied.  The proposition here is to have a toggle or similar that the designer/owner of the dashboard can use to turn this behavior on/off.  Thus the reset reasoning behind this can be maintained if required for that model or use case and the behavior can be turned off if non resetting is optimal for the use case needed. ... View more

NOTE: This is based on a post I made in the "Help & How To" forum. It was recommended to me to make a post here, in the Product Feedback forum, about it. There should be a way for Sisense to display the currently-selected filter values directly on the dashboard. I know that users can see this by expanding the right-hand sidebar to see the Dashboard Filters, and it's also a display option when you export to PDF. However, these aren't always obvious or available to the intended audience for the dashboard. For example, if I have a dashboard that's setup with a dashboard-level filter for "Work Group", it would be nice if I could display - directly on the dashboard itself - what work group is currently being filtered down to. Also, some of the values can be unclear, especially for date filters that use colloquial language such as "This Year" or "Last Quarter". There should be some way of displaying the exact date ranges that are being filtered for. For example, if today is November 13th 2024, then the "Last Quarter" filter value would mean dates from July 1st 2024 through September 30th 2024. But after the new year, if I were to check the dashboard again on January 6th 2025, then the date range would instead be from October 1st 2024 through December 31st 2024. Currently, in both cases, it'll just still how as "Last Quarter" and it's up to the user to figure out what that means (which can be difficult if they're looking at a static printout from some unknown time in the past). Other time periods can also be ambiguous or prone to misunderstanding. For instance, it's common to see people assume "Last Week" means a rolling time period covering the last 7 days, rather than the prior calendar week. Or assuming "Last 2 months" means the same as "this and last month", rather than the prior two calendar months not including the current month.  ... View more

Hi, I have issue with setting two date dimensions in dashboard. The goal I want to achieve is to have a history of records, which are also have it's own date dimension. You can download model and dashboard. I would like to have difference in growth ower selected period for defined dates. For example 2 days ago I sold a car for 1000, and yesterday the proce have changed to 1500. The sale date is for today. So the growth over 2 days is 1500, and for 1 day 500. Additionally i would like to compare those values to previous year, and go back with both date dimensions. Tricky part is also, that if there is not sale for this year, PastYear() wont count prior year. I tried to play around with filtered measure plugin, however it doesn't give me exactly what i want. Do you have any suggestions how can I approach such data and 2 date dimensions? Thanks 🙂 ... View more

Allow widget designers to turn off the display of the Grand Totals calculation for certain columns in a pivot table. Situations exist for which it is highly beneficial to display the Grand Total of a select few columns in a pivot table, and not to show others as they can be misleading or distract from the primary purpose of the pivot table. I envision an implementation that enables the Grand Total to be toggled on/off for each column, with the default being 'on' for all columns. ... View more

Hi, We are facing some issues in the hour timezone related to builds. There's no way to change the UTC setting for the build schedule. It's very confusing and it's generating many issues in timezones when there are changes in the hour due to summer/winter time adjustments. We tried to change the timezone in the upgrade process, but the build schedule only works in UTC. This is a nightmare for data administrators. We need to be able to set the local timezone as the build schedule setting.  Best ... View more

It seems strange to me that each time I need to add another table to my model, I have to re-enter all the connection (or pick a recent connection). If i have 10 tables in my model, they all have their own individual connection information. If a server name ever gets renamed, it'll be a crazy headache for us. There should be a place where you define the distinct list of actual database connections (one per server or database), give it a name, apply security, etc. And then when you go to add a table to a data model, at that point you pick from the previously defined available list of connections. ... View more