Hi Marcin,

great comparison.

Honestly speaking I am still a fan of analyses and the "easy" way of just creating a nice looking dashboard in a very fast way but I totally understand your concerns and I am also concerned as we are also using more and more views (Starter Kits, several apps and own implemented views) within our environment.

Especially the question how do views behave after release of the new studio that are using components that are no longer supported should be addressed beforehand.

I cannot imagine that they haven't thought about this. On the other side would be a mess if views would now behave different based on its creation date like having a view 1.0 vs. view 2.0.

Lets see what will happen but this might be a complete showstopper.

Dear @marcin.konop13 ,

First of all, let me address the most important concern and assure you and others, we won’t break any in production running apps! Your existing content will not change in any way and you can (for a long time) decide which of the dashboarding tools you want to use, by switching back to the previous view editing or continue to create Analysis. 

Second, we are not ‘done’ yet with our development and we are aware of gaps. We are actively working on resolving those, so there will be many releases following after May 13th. More below.

Third, thank you very much for the detailed feedback, this helps us to make our communication clearer.

Let me address the topics you raised 1 by 1:

1. Why did we invest in this? - Even though we are humorous people working in Product and Engineering, no, of course this is no joke We took our user feedback from the past 3 years about the difficult view editing experience very seriously.
   1. As announced at Celopshere in November 2023 we invested (and will continue to invest) heavily in lowering the entry barrier for our users to build UIs and dashboards and enabling them to leverage the latest platform capabilities.
   2. The first version of Views was and is hard to adopt and to be leveraged to create value by our users due to the steep learning curve for building and missing process mining capabilities compared to Analysis.
   3. This is why our key priority is to deliver a solution that is intuitive and easy to use, in many aspects more similar to Analysis (like you also pointed out at the end of your post), while incorporating all the latest platform capabilities, such as Augmentation, OCDM process analysis or linking views (formerly known as profile views) for better user flows, to just name a few.
   4. We have run an extensive Limited Availability program with over 89 customers who provided us graciously feedback to define a feature set and bring the usability to a level that first productive dashboards can be built successfully. While we are not able to cover everything yet, it will reduce the learning curve and bring many users to the latest platform capabilities. And we will continue to close the gaps as fast as possible to enable also more advanced set ups, like you already did in legacy Views.
   5. What will be released on Monday, May 13th is just the beginning. Please also see the pre-release notes, outlining the changes (also additional ones that are coming with the general availability vs. what is currently on the training cluster) and explaining that your work will not be impacted. 
2. How will older assets work? -The same as today!
   1. When you create a view, we’ll use the new view editing experience by default. You can switch to the old editing experience if you want to go back to the way it used to work. Simply click on the button on the top right to switch to the old editing experience. We recommend you to use the new editing experience to enjoy the simpler flows and latest features, when already possible. However, we do still have some gaps to close as you noticed and rightfully pointed out. To be as transparent as we can be, we added a Studio Feature Availability Matrix to give you an overview where we currently stand and what we plan to address with urgency.
   2. Studio Analysis hasn’t changed, so you can continue to use that in the same way as before.
   3. All your existing (already created) assets will be unchanged and when you go to edit you will edit them as of today. 
   4. We will further offer support and maintain legacy Views and Analysis until we are sure to cover all needs and ensure a smooth migration of existing content - this is far in the future and there is no rush!
3. Layout changes - Scroll is coming with GA in May!
   1. The drag and drop layout is indeed one of the core changes we opted in for to allow for an easy drag-and-drop experience to place, resize and rearrange components. We are glad you like it!
   2. You will be able to change the layout from a ‘Scale-to-Fit’ to a ‘Custom Height’ layout. The ‘Scale-to-Fit’ is what you are currently seeing in your training team. The benefit is that you will always see the same no matter what screen sizes you are opening the View. The ‘Custom Height’ on the other hand will allow you to define a height and this will add a Scroll if that cannot be displayed at once on the screen.
   3. By the way, in Analysis you are also not limited to the Fullscreen option, but in the Sheet settings you can change to ‘Fixed Width’ to define the height of the page and add more components.
4. Availability of components
   1. We are aware of our gaps and I can assure you that it is our utmost priority to drive the development of the remaining components. In the meantime, we want to be as transparent as possible to give you a heads up what's already there and what not, please see our Studio Feature Availability Matrix for details
   2. And most importantly, we do not want to get in the way of any implementation that is in need of capabilities that we are not able to cover yet. This is why you always have the choice to switch back to the old view editing experience or create an Analysis until we are able to address your needs!
5. Tabs within Views coming soon (Q3)
   1. The new Views come with in-built tabs, similar to how Sheets worked in Analysis, the difference they are on top vs. bottom of the page.
   2. We do have it on our short term roadmap to also offer so called ‘Container Tabs’, which enable you to place containers on your view with tab-able content.
6. Profile Views set up is now simplified by just linking Views - already there!
   1. To open a view providing more details when clicking on a table cell you can now simply link any view, without the need for a definition of a Record ID. The the flow will allow for more flexibility that before required rather complex workarounds. You can find it described in our documentation about Linked Views.
7. Filter sidebar - coming with GA in May!
   1. Not all changes are already live in the training teams. The filter sidebar, as well as making views scrollable or further improvements on variables will only be released on May 13th.
   2. As you correctly saw in the video the filter sidebar will be an in-built capability that will no longer require YAML nor javascript skills to achieve a toggleable sidebar for placing filters.
   3. After careful consideration with our design team, we decided to for now not offer vertical collapsible content because we do not consider it a design best practice. Based on this we will overhaul the design of the Celonis provided Marketplace apps.
8. To close, I want to iterate again on the fact that 
   1. It is optional. When creating a view you can choose between the two view versions. Or create an Analysis.
   2. It won’t break any running apps and dashboards.
   3. And to address @dennis.pflug comment - we are also already actively reworking the Marketplace content provided by Celonis so you can start new implementations in the simplified editing experience. Once we closed more feature gaps, we will also provide migration tooling if you decide to move your own view implementations to the new experience.

Thank you again for trying out the new product right away and raising your questions with us. So we had the chance to address them. I do hope I could create some clarity and relieve your concerns. 

If you have more questions, please do not hesitate to continue in this thread - and I am also more than happy for a 1:1 feedback session!

All the best

Sabeth, on behalf of the Studio Product Team

Hi Dennis.

Oh by all means. I love analyses as well. They are perfect for what they are. Especially good for new users and for citizen developers. I wouldn't want to get rid of them in any way.

What I am concerned however is the changes to views. I like analyses for what they are just as well as I like the views for what they are. A robust framework for creating more detailed, proffesional apps that can enable really advanced analytics.

In my opinion there is space for both. And if this new enhanced views was a seperate component or a replacement for analyses, I would be by all means supportive of that. Because in my opinion both serve the purpose of simple (not mistake with simplistic) and easy way to create dashboards. An inclusion of YAML layer additionally allowing for better archiving and developing.

But replacing the advanced and really detailed asset of views with something more akin to analyses (at least in current version) is something that I am very concerned with and would really like to avoid.

Since I started working with Celonis back in 2018 I must admit all the changes until now were of very successful and positive impact. Even if early implementation of views were a bit problematic. This is however something I see in its current form as an unnecessary step back.

Like you mentioned I am hopeful, but that is also why I wrote this post. Not because I want this to fail or take pleasure in criticizing others, but because I am deeply concerned about the state of it.

Kindest

Marcin

I also want to weigh in this discussion as a participant of Limited Availability phase who has made first hands on experiences and is not a Celonis Employee.

The important thing which Sabeth clearly stated: everything is optional. In our environment we have Analyses, Old Views and the new Views all next to each other. And obviously the existing assets are still working fine and can be maintained further. When I need to create something new I can freely choose between all three.

However, already today where feature parity is not yet reached my first choice would be the improved "new View". The biggest benefit for me is that I can create ad hoc PQLs as in Analyes but can also decide to store it centrally in a knowledge Model. For me it gives the flexibility and allows "quick and dirty" as Analyses do but once you mature with it it is much easier to maintain knowledge centrally in a KM and not have stored formulas and variables in every analysis like the Views do. To me it feels like the best of both worlds and is perspectively able to cater both ends of the spectrum between analysis and View.

Since years I was honestly puzzled how annoything the Development Experience was for the old Views. A lot of thos eare being solved now.

There is an argument to be made how much the naming of this component is fitting as this is definitely a different asset type than the "old" View. But I guess in general in IT and especially with Celonis we all know the product names are kind of confusing and everchanging. I would guess that they did not want to introduce yet another name for another "Frontend/Dashboard" Component again.

Also obviously as of right now the feature parity is not fully reached and not all features they want to implement are implemented working perfectly. E.g. the scrollable View that you whished for and Sabeth mentioned will come soon. In general I am fairly optimistic Celonis will improve till it becomes GA and then in rest of year as well.

So in summary:

• I do consider it a rather positive impact.
• Rest asured that it is your choice if, when and how to switch.

IMHO no need to be feel unrest about the Launch of it with GA. Just test the water once it is there.

Hi Sabeth

Thanks a lot for your quick response! Hope my criticism wasn’t taken personally.

I have been working with Celonis since waaay back in 2018 and I must admit it was the first time I was really concerned about the state/development of the product.

I had engaged with what I saw available in the training environment and my concerns were (as I assumed) that it was already the end result which will replace the views (as I didn’t see any option of switching to old views). It is good to hear that things are still in development and the new views are an optional thing.

Let me as well respond to your points:

1.      By all means I understand and support the approach to lower the entry bar into development. I think it was quite a large hurdle for many to overcome after introduction of views way back when. And taking into consideration things like citizen developer programs it is a great direction. My concern was that it would completely replace what has already been developed and as you could’ve read from my post I saw it as a huge limitation. In my opinion while lowering the entry bar is a good idea, I would be concerned if at the same time it would lower the opportunity ceiling, disabling or limiting development possibilities.

2.      That is wonderful to hear. I was afraid that some of the celonis marketplace assets would no longer be applicable, not to mention assets we created ourselves.

3.      Brilliant, then that point is null and void if its coming in an update.

4.      Yes. That was also a large concern for me. Especially seeing as not all components were fully developed in current (old) views and many of them could only be edited in YAML, now adding to that the fact that the syntax is changing for many of them (if not all) it is a bit concerning how will that impact the documentation. Additionally some limitations are in my opinion a bit too strict (like only 4 categories for pie charts, analyses are working fine with no limits)

5.      Null and void like 3 in that case.

6.      Oh perfect. Was concerned as there were no training materials available for me to look it up in. I just noticed that in the new setup views are either regular or live, with no profile views option

7.      That is good to hear. Regarding 7.3 I understand your concern, albeit I saw the collapsable sidebars and content as a rather elegant and interesting part of the structure. A bit difficult to set up, but impactful nonetheless. Sidebar filters should fulfil most of the needs those components were bringing to the table, but I think in some niche scenarios they might be useful.

8.      And that is the most important thing that had me quite scared. That it would not be optional like the switch from analyses to views (maybe disregarding advanced features like multi eventlog which are not cooperating with analyses so well).

Thank you kindly for your response. I will certainly come back after playing around a bit more with it, and experiencing the training materials as they are released (or were released yesterday) especially with our own data and comparing results to old setups. I think developer enablement is great direction to go. It is something that is also very high in our priorities as we focus currently on our citizen developers and including more people in creation of assets within the company.

Thank you once again for your response and all your hard work. I am really happy to see that Celonis is developing and going forwards.

All the best to you and the team!

Marcin