On August 6, 2018, Adobe announced plans to end-of-life Ad Hoc Analysis, as well as the Discover point-product. Ad Hoc Analysis functionality and capabilities are being moved to Analysis Workspace. It is common practice for Adobe to announce plans to deprecate a technology at least one year in advance, to give our customers plenty of time to migrate projects, learn new workflows and ease into the transition. To ensure all users make a smooth transition, Adobe is targeting Q3 2019 to deprecate Ad Hoc Analysis. For more information, visit https://adobe.ly/discoverworkspace.
In support of this announcement, we are working toward improving 3 primary workflows in Analysis Workspace, giving you the ability to:
1) Build advanced queries, without having to wait for individual components to build
2) Download 50K rows from freeform tables
3) Remove repeat instances from Flow
A date for end-of-life/access will not be set until these workflows are improved.
This thread has been created to provide a channel for feedback about the announcement.
Thanks for this feedback. We agree that a table builder-like feature is important to have in Workspace before access to AHA ends. We are planning on building an advanced query building experience into Workspace, one that is similar to table builder in AHA but also much improved. This is something we are currently in development on and are targeting for release in the spring, ahead of AHA EOL.
I wanted to take the opportunity to chime in with some thoughts. I appreciate the transparency and the requests for feedback, and while I share the displeasure of many others in this thread, I understand Adobe's point of view - it's simply not cost effective to maintain multiple tool-sets that could potentially be rolled into one tool, ideally covering all user needs.
As someone who has (on multiple occasions) been a part of talks debating migrating an organization from Adobe Analytics to a competing tool, I can tell you that it's the 8-10% of power users in the organization that make that final call. The remaining 90% of occasional users in the org rely on the tools that the core analytics team ultimately selects to implement. These users tend to use the simplest, most user-friendly toolsets available. For the 8-10%, Ad-hoc analysis, and more generally the ability to perform advanced deep dive analysis, has always been the main sticking point in these conversations - Adobe, in the past, has been ahead of the competition in this area.
However, this is less and less the case. Knowing the advanced toolsets are being stripped out, and with little confidence in the proposed replacements, Adobe continues to lose ground in this area. Your main competitor has recently introduced their 'advanced analysis' toolset to compete with workspace analysis. But what's far more important and relevant than this, is they now allow for a one-click integration into an Enterprise level, cloud-based data warehouse. This solution allows customers to use toolsets of their selection for advanced analysis (See Looker, Tableau, R, etc.) in a way that doesn't require constant limited exports and extracts of Data, and jumping through Hoops to get out of Adobe's walled garden.
What the retirement of Ad-Hoc analysis truly does, is highlight true shortcomings of the product for advanced users. I'm obviously only one person, with knowledge of experiences and opinions from only a few colleagues at a few organizations. While I respect the use of data to drive your product roadmap forwards, I hope you have taken customer segmentation into account, and are listening to the needs of your clients that drive analytics roadmaps within their organizations.
For my team we're going to move to using more SQL based analysis for deeper analytics once Ad Hoc is sunset as the current tool set won't allow for the types of analytics we currently do. Once that happens Adobe will have lost a lot of what makes it desirable as an Analysis tool.
Thanks for sharing this feedback. I have a ton of thoughts, but I'll try to be concise.
First, clearly the ability to query data and get it out quickly and easily is more important to analysts than it was five years ago. I think it's already been publicly announced that we're hard at work on Adobe Experience Platform (AEP), which is a place to bring all relevant customer data together in one place to power your experience optimization and marketing. Part of AEP includes a SQL interface with export capabilities as well as built-in connectors to front-end tools like Power BI and Tableau. Of course, the proof should be in the pudding, but where some tools are mimicking (poorly, BTW) Analysis Workspace, we're also working on a solution to address the need to build custom queries and take data out more effectively than you can today with Data Feeds.
Second, as I think has been stated elsewhere on this thread, generally speaking the signal from our broader user base -- both quantitatively and qualitatively -- has been strong in favor of Analysis Workspace allowing deeper and more flexible advanced analysis than Ad Hoc Analysis allows you to do. Even just Attribution IQ alone should open up dozens of new use cases for analysis for every brand who has adopted Analysis Workspace. There certainly are some gaps between the two products, which Jen has described on this thread and on the migration page that we've posted, but those are being addressed on the product roadmap. And we've said repeatedly that we're keeping Ad Hoc Analysis around until you and we are satisfied that those gaps have been closed.
Third, looking at the landscape for the customer intelligence BI market, the idea of Analysis Workspace -- with its speed and flexibility --running on top of AEP, with all of your relevant customer data across channels flowing into it, stitched together however you want, and with its connections to points of action like Adobe Target, Adobe Audience Manager, Adobe Campaign, and Adobe Media Optimizer -- and even embedded within those tools -- are really compelling, we believe. It may not be quite the same path that others are pursuing, but that's intentional. If that isn't enough for an analyst to get excited about the possibilities, then I don't know what to tell you. From where I sit, talking to hundreds of brands every year about their analytics goals, the future is exciting indeed. But we only get there as quickly as everyone wants if we focus our efforts on our primary analytics environment, which is Analysis Workspace.
Group Product Manager
How many users were aware of Ad Hoc or were using Ad Hoc that said they preferred Workspace?
I have plenty of user I recommending Workspace to as it is a vast improvement over the SC interface. It is going to give them more self service options and ultimately make what they do easier. However these are not our power users. Those like myself that spend a significant amount of time in Ad Hoc are not looking forward to this forced change and feel that were are losing more than we gain. Those things that are gained could have been developed for Ad Hoc but were instead developed for a lesser tool.
ashert33772002 This is one of my earlier posts...
"Adobe's been trying unsuccessfully to end the life of Ad Hoc for several years now, and replace it with Workspace."
"In the recent announcements they quote a low number of Ad Hoc users. Somewhere in these comments, someone said that's because most nubee's aren't aware of it. I would concur. When I work with younger analysts, and turn them on to Ad Hoc, I usually get two comments:
"Internally at my company there has been a lot of griping about this among analysts (not just report jockeys)."
I have been concentrating in setting some templates with customized roll-over dates and intelligent alerts in Workspace in the past weeks.
As an AdHoc deep-dive user, I more and more feel like:
1. AdHoc is a fantastic tool to do the ANALYSIS JOB in various of dimensions, by using the combination of different report suit
2. Workspace is a good "data visualization" tool, to my Director, to Marketing Consultant, as well as a good tool to set scheduled reporting. More "visual-friendly" types of charts give business users better feeling.
3. I tried to create complex table, calculated metrics in workspace directly, the always "not responding" by the browser is really not tolerated. I need to wait longer and longer time to responding along with the complex of the tables.
It's the 2nd time I raise the opinions, and as I replied before, I don't think the performance issue can be resolved as workspace is Browser & Internet based.
Again, even the voice will be listened but not be heard, I do not wish Adhoc sunset.
I agree 100%. Even if Workspace eventually gets all of the functionality of Ad Hoc, which is wishful thinking, it still will cause huge delays for analysts because it is browser-based.
Real question, Jen/Adobe: what will it take to get you to reverse this decision? I honestly can't envision my day-to-day without ad hoc. If it's true that only 8% of the users utilize the tool can you also look at how much time is spent logged in to the adobe platform by those 8% verses the 92%? I bet it's close to half if not more. You're hurting your power users (likely your loudest advocates as well) and that seems like a mistake from a business perspective. Maybe it's a money saver but at what "true cost"??
Yes, to disregard the 8% of users and use that as an excuse to end a product is mystifying especially for a company that prides itself on analytics. On our site 3% of users account for 40% of total page views. For us to discount this 3% because it is small number is unfathomable.
Good morning. I don't know if it has been mentioned, but another key feature that needs to go in workspace if this is to replace ad hoc the ability to download 20K rows of information. Currently you can only pull what is displayed.
It was nice of Adobe to provide a forum here to discuss and log our discontent with this decision. However, after reading their responses, in my opinion, it seems they are not seriously entertaining the group's general consensus to reverse the sunset decision.
I have created a simple petition to show them how many of us there are. My initial goal is to gather as many signatures as possible and get this into the hands of the right decision makers in Adobe Senior Mgmt. My ultimate goal is to change their minds!
I hope you'll sign it and also share it with your various colleagues, groups, circles, etc.
We still have a lot of time. Together and with the help of the awesome Adobe Analytics community I honestly believe we can accomplish the goal of retaining Ad Hoc Analysis as a supported product.
I welcome anyone's feedback.
All the best,
First, let me start off by saying that I think Workspace is a great tool. I've been very impressed by the progress made on it in the last 2 years. It started out only having a few simple visualization options, etc. and now it's a really rich tool. Little things like doing side-by-side date comparisons in the same table are great features recently added and much appreciated.
That being said, as a "power user," it still isn't at the point where I can do in-depth analysis. It is mainly a performance issue, which I know has been brought up many times by other folks. However, if I cannot even get a fairly simple Workspace to load the first panel and let me interact with the first visualization, I cannot even start my work while waiting for the entire workspace to load. This has plagued Workspace since the very beginning and I have not noticed any improvement in performance wrt load times and being able to interact with the workspace while the entire thing is loading.
This past Cyber Monday, Nov 26, I was contacted by a former director asking if I was able to get any data from Adobe. Her team and our BI team in charge of automated reporting were not even able to log in, let alone get any info. I opened up Workspace and after a minute of waiting for it to load, I decided to open Ad Hoc in parallel. Ad Hoc loaded as typical and I was easily and quickly able to get hourly visits for this year's Cyber Monday and last year's. I switched over to Workspace and it still had not even loaded the first panel or first visualization. It was a very simple project with maybe 2 tables for daily visits and UVs. I think I had 3 segments applied but the timeframe was only set for the last week so it should not have taken very long to load such a basic project. Perhaps that was a rare occurrence due to Cyber Monday traffic highs.
Yesterday I was working in Workspace. Again, it was a very simple project, but it had 9 graphs for the last 90 days and 9 corresponding hidden tables. I was graphing visits, UVs, and PVs with basic dimensions like day, channel, english vs. spanish, etc. At first Workspace was working fine, but as I kept tweaking it, the entire project would freeze. This went on for the entire hour that I was trying to work on this dashboard. I was not even trying to do analysis, just simple setting up a dashboard for a team to get daily updates. I do not know if anyone else experiences these issues, but I run into them on a regular basis using very simple workspaces. I've found that even if I only look at 7 days with day as the dimension, if I have more than 3-4 tables (not even graphs!), it freezes up.
An average project of mine in Discover has 2-4 workspaces and 4-6 reports within each. The reports range from simple tables of visits to flow reports. The reports span 3 years and have different time granularities as well as some use page url dimension, etc. While one report is loading, I can easily switch back and forth to other reports in the project and I am not delayed at all. This kind of parallel work is NOT possible in Adobe Workspace right now for the type of work that I do on a daily basis.
Please please please do not discontinue Ad Hoc. Minimally, please maintain it so that it remains relevant and working on Windows 10 and beyond.
I agree with what Lindsay writes - WS is often very slow too load and may fully freeze or crash if I try to do things too quickly. It seems that especially freeform tables with (multiple) breakdowns and segments are vulnerable to this.
Adobe team - Any update to the state of workspace? Trying to come to terms and migrate my workflows, but it seems a lot of capability is still missing, and I keep running into issues like the attached. Would love updated timeframes on some of this stuff.
I've seen, on the Ad-Hoc product/download screen, that there is now a version that can be used with Java 11. Will this version also be killed off or will it continue you work when you end of life Ad hoc?
Most people replying to this thread seem to have similar ideas...that workspace is the much needed upgrade to the standard reports. Instead of trying to replicate ad-hoc analysis into workspace, Adobe should be looking to ensure all functionality available in the standard reports is available in workspace (e.g. Calendar Events). And then leave ad-hoc analysis for the analysts to be able to quickly drill down further.
It is undoubtedly much quicker than workspace and most importantly you can export 50k rows. I know this functionality is 'in the pipeline' for Adobe...but that means until then we have the limited capabilities of data warehouse or spend weeks downloading 400 rows at a time...it's just not feasible. Our solution has had to be that we set up a clickstream feed and force our business outside of the tool...which means we become supplier agnostic.
One thing that has tripped us up is that panel level segments are not applied to the fallout report in the same way as Ad Hoc Analysis.
To recreate the same affect you have to add the segment to every touchpoint which is a pain.
I was wondering if Adobe might consider aligning this? (or at least explaining why they are different)
Our biggest problem with this behavior is the following scenario where we have an application process that users can apply for several products through.
We wanted to have a fallout report for the different steps with a panel level segment drop down filter for the different products over the top of this. However, if a user applies for more than one product in a visit data from one products shows in the fallout for the other which is a real pain
It means you potentially have to build out 20 fallout reports where you could have had 1 fallout report with 20 drop down options. This feels like a very common use case that can't be properly accommodated.