Horizon Rules Bugs and Enhancements List (Last updated December 8th)

rakesh
Userlevel 6
Badge +1
  • rakesh
  • Lets put your data to work!
  • 824 replies

We sincerely appreciate all the valuable feedback you’ve all shared thus far on the launch of Horizon Rules Engine. We are working diligently to prioritize and address all bugs, issues, and parity gaps and plan to keep you up-to-date as we progress. 

As a result, the table below provides relevant information on each of the items. We will be updating this table, so feel free to reference this as a source of the most current information regarding known issues. If you have any questions or comments, please thread your responses and we are happy to assist. Please allow some time for a response. 

Note: We recently announced the extension of the Bionic Rules deprecation and Horizon Rules auto-migration dates. If you haven't had the chance to review the updated time frame, please see the post below

Thank you again for all your contributions!

Table last updated on December 8th. New table items are highlighted in green. 

*Please note the ETA listed is an estimate based on current resourcing and a basic understanding of the level of effort required to deliver. Should resourcing or level of effort to deliver change, we will update the table appropriately and consider any downstream impact to overall deprecation dates. Time frame for hot fixes is typically 2-3 weeks.

 

40 replies

alizee
Rank
Userlevel 6
Badge +9

Hi @rakesh That’s awesome, thank you!

Should we continue to add our enhancement requests in the open Community and can we link them below for reference or how do you prefer us to proceed for anything new?

If you don’t want to make mistakes, don’t do anything. If you want to learn, ask questions and make mistakes, except in construction because they’re the most expensive part of building.
W
Rank
Userlevel 5
Badge

Also, if there are items people have submitted tickets on should we post them in this thread also to get them added to the list?

 

rakesh
Userlevel 6
Badge +1

Hi all, thanks for asking this. 

I do have a Gainsight Dashboard where I can see all Horizon Rules issues (& their impact, status). So any support tickets, once accepted by engineering will be visible to us directly. If you face any issues, I recommend you submit the tickets and most of them should appear here once we identify them as bugs. 

 

Any enhancement requests that are parity gaps, I recommend you ping here. If its generic feature requests, open community is the best place to post to get broader customer feedback. Note: parity gaps and bugs will be prioritised higher than any new feature enhancements. 

bradley
Rank
Userlevel 7
Badge +6

We talked about how having two ways to build rules doesn’t really make sense during the beta (here), and now it seems like more and more discrepancies between the two methods are popping up like this: 

(can’t rename field names or see Path when you Select an Object), and evidently Select an Object rules are limited to 5 rule actions? (Haven’t encountered this myself but it came up in the slack community). 

Is having two ways to build a rule really solving a problem or is it one step forward and 2 backwards? Consistency in the product has been a major complaint from admins for years and while I get that there is some value on having a simple method it seems like it is just creating more complexity to maintain and disparity on function even within the same feature (rules).

rakesh
Userlevel 6
Badge +1

 

We talked about how having two ways to build rules doesn’t really make sense during the beta (here), and now it seems like more and more discrepancies between the two methods are popping up like this: 

(can’t rename field names or see Path when you Select an Object), and evidently Select an Object rules are limited to 5 rule actions? (Haven’t encountered this myself but it came up in the slack community). 

Is having two ways to build a rule really solving a problem or is it one step forward and 2 backwards? Consistency in the product has been a major complaint from admins for years and while I get that there is some value on having a simple method it seems like it is just creating more complexity to maintain and disparity on function even within the same feature (rules).


Thanks for the feedback Bradley, 

The single object flow is if data is already perfect and you just want to act on it. In this flow,

  • Can you make me understand why you would need field renaming at all? 
  • Showing the path in derived mapping is a miss and we will explore if there are similar gaps and will handle it.
  • The 5 actions limit is one thing we are observing and will work to increase this. 
  • [Roadmap] - Easy switch b/w single object flow to data prep flow 

More than that, we will continue to observe the usage and feedback here to inform our direction.

bradley
Rank
Userlevel 7
Badge +6

Hey @rakesh sorry, I tried to link to the specific example from the Slack community. Basically, when you get situations like this:

Maybe seeing the pathway would solve it, maybe it will help but still be annoying to have to mouseover every time to see which GSID is which, but that also kind of misses the point. The main point here, and with all of my examples, has been “why are there two functionally different classes of rules?”. That’s the question I’m still struggling to understand the answer to.

By making some of the rules easier to build most of the time it just seems like the net complexity of the rules engine overall has increased. 

I’m just thinking back to my early admin days where I was trying to understand rules engine and there was one way to make a rule - I can’t imagine walking into it now where instead of learning one rules engine, I’m learning two rules engines.

“For this use case use option A because it’s quicker to build off a single object unless you want to do ABC, oh or ZYX, or unless you think you might need to add steps later”. “Or, you could use option B for every use case but it’s marginally less efficient if you’re just pulling from a single object”.

And now, to close the gaps between the two there’s additional work for your team to complete which also leads me to believe that the tech debt and ongoing maintenance for two workflows has also increased as well.

alizee
Rank
Userlevel 6
Badge +9

Hi @rakesh May I ask you to check out this post I just made, as editing actions has been driving mad: Improve action editing in Horizon Rules Engine | Gainsight Community

If you don’t want to make mistakes, don’t do anything. If you want to learn, ask questions and make mistakes, except in construction because they’re the most expensive part of building.
bradley
Rank
Userlevel 7
Badge +6

is this considered a bug or enhancement? 

 

shambhawi
Rank
Badge


Thanks for the feedback Bradley, 

  • The 5 actions limit is one thing we are observing and will work to increase this. 

 

 

Hi @rakesh @bradley, In the Single Object Flow, we have limitation on the branch level but not on Action level, as in we can create only 5 criteria/branch but one rule can have up to 50 Actions.

@bradley Based on your recommendations, we also had this limitation updated in our FAQ document. 
 

christian_auner
Rank
Badge

Hi @rakesh,

thanks for this post and the special group related to Horizon Rules. I still have some concern about the auto migration. I’m definitly a fan of Horizon Rules but for me there are still some feature gaps that would prevent us from migrating existing rules to Horizon.

Please checkout that article: 

 

For us it is important to define the execution order for actions and this is not possible with Horizon rules so far. In a lot of cases it is mantatory that e.g. a reset action needs to run before adding new data point.

How to be sure that such feature will be implemented before having an auto-migration?

 

Best

Christian

darkknight
Rank
Userlevel 7
Badge +3

@rakesh  I reported an issue on Aug 1 that was accepted as a bug that I don’t see on this list but needs to be addressed. 

Frequently when I open actions in a Horizon Rule that contains SF fields and was already built and running, it claims that fields have been deleted from the source or data set, but in fact they have not been deleted from the source or the data set.

The “workaround” is that If I refresh Salesforce metadata it resolves the issue in that moment, but invariably it happens again and again. 

As I told support: Before Horizon Rules, I **never** had to refresh metadata except for when a new field was added in Salesforce that I wanted to appear in Gainsight.

Now I am frequently having to refresh metadata. It's frustrating and frankly not acceptable.

 

 


 

Jeff Kirkpatrick
tmorgan
Rank
Userlevel 1
Badge +4

@rakesh I also have an outstanding item that was noted as a bug that is not showing on this list. It was logged on August 17 ID#268822. The Horizon Rules are not allowing you to use the function to Subtract N Days if the number is larger than 365. This was not an issue in Bionic rules.

sshroff
Rank
Userlevel 6
Badge +3

@christian_auner This is being considered and potentially targeting October release, but the release date is very much subject to change. We will update the table accordingly. 

Sonia Shroff, CS Ops Program Manager
W
Rank
Userlevel 5
Badge

 

rakesh
Userlevel 6
Badge +1

@rakesh I also have an outstanding item that was noted as a bug that is not showing on this list. It was logged on August 17 ID#268822. The Horizon Rules are not allowing you to use the function to Subtract N Days if the number is larger than 365. This was not an issue in Bionic rules.

Hi @tmorgan 

We have fixed this issue and it will be in production next week. 

heather_hansen
Rank
Userlevel 7
Badge +10

@rakesh Posted this one a while back re: Update CTA Action not including Linked Objects.  I get it’s not a bug or parity issue, but does seem like a miss.  Happy to share our use case if it would help.
 

 

W
Rank
Userlevel 5
Badge

Found a parity gap.

 

Within bionic rules on the add date/subtract date function you are able to both use a numeric field from your source data set or type in an actual value.

 

However, in HRE, as well as DD but I’ve been told that is expected, you can not do this. HRE behaves the same way as DD does and only allows you to type in an actual value.

Value/Field

 

Only Value here.

 

W
Rank
Userlevel 5
Badge

Found something else a little weird in HRE.

 

When exporting it asks you for the file path twice.

 

 

 

Do both work? I would assume not and only the second File Path would.

bradley
Rank
Userlevel 7
Badge +6

Found something else a little weird in HRE.

 

When exporting it asks you for the file path twice.

 

 

 

Do both work? I would assume not and only the second File Path would.

It appears that the top one is really just the File Name, and the File Path below is populated via browsing, and ends up being the path itself.

W
Rank
Userlevel 5
Badge

Found something else a little weird in HRE.

 

When exporting it asks you for the file path twice.

 

Do both work? I would assume not and only the second File Path would.

It appears that the top one is really just the File Name, and the File Path below is populated via browsing, and ends up being the path itself.

 

I would have assumed that. But it’d be nice for them to clean up the text so it is clear.

 

 

Another issue, but is not parity issue, I have is when I go to export it defaults to single quote as escape character. Can we get that changed to backslash? Since backslash is the global standard and single quotes are often used in text/rich text when CSMs are typing.

 

If you wanted to get really cool, it’d be nice if you could let us type our own text qualifier in. As sometime backsslash is also used when end users typed in/paste file paths or URLs. Allowing us to alter or change to what our standards are would be amazing. I feel like we had this at one point. Or maybe it was just ingesting data a long time ago. 

rakesh
Userlevel 6
Badge +1

Hi Wayne and Bradley, 

Thanks for bringing this up. 

On the formula field, you can enter a static value, but the copy or UX doesnt indicate so.
We will work on cleaning up the copy a bit and the default options in both Export to S3 and formula builder. 

alizee
Rank
Userlevel 6
Badge +9

Hi @rakesh 

I want to add this to the enhancement list: authorizing user pools in the DEFAULT OWNER field of the Call To Action 2.0 action. 

WHY: my CTAs will be assigned to say - CSM, but when I don’t have one I’d like the default owner to be a group instead of having to create multiple branches to filter on Region + Tech Touch. I think that’s be very useful and that’s really there area where one would like a group: as a fallback. 

Thank you!

Alizée

If you don’t want to make mistakes, don’t do anything. If you want to learn, ask questions and make mistakes, except in construction because they’re the most expensive part of building.
alizee
Rank
Userlevel 6
Badge +9

Hi @rakesh 

I want to add this to the enhancement list: authorizing user pools in the DEFAULT OWNER field of the Call To Action 2.0 action. 

WHY: my CTAs will be assigned to say - CSM, but when I don’t have one I’d like the default owner to be a group instead of having to create multiple branches to filter on Region + Tech Touch. I think that’s be very useful and that’s really there area where one would like a group: as a fallback. 

Thank you!

Alizée

 

Adding to my request: this would also make maintenance so much easier when looking at process CTAs such as “new account assignments”, where people come and go. The user pool would be automatically kept up to date with criteria and this would create less maintenance work for us. It would also help us update the default owner for a bunch of rules in one step, from the user pool. That would be some proper admin QOL.

If you don’t want to make mistakes, don’t do anything. If you want to learn, ask questions and make mistakes, except in construction because they’re the most expensive part of building.
spencer_engel
Rank
Userlevel 7
Badge +2

Hi Rakesh - 

 

Just now diving into Horizon rules. I did not see this issue pointed out anywhere else, but apologies if it’s a duplicate. The first thing I noticed is that Execution History is different. First of all, it seems I have to click into an individual rule to see its Execution Logs - is this correct?

 

Also, is there any way to fix the format of the numbers that show in the “pass” and “fail”? I’d like to see the exact number like the old execution history showed (see below). 

I like this format

Seeing a rounded “23k” or, worse, “0.30M” is useless to me. I’d like to see the exact number at a glance.

 

This format is no good

 

spencer_engel
Rank
Userlevel 7
Badge +2

One more Execution History-related question. I’m doing test runs for both the bionic and horizon versions of the same rule in our sandbox. Is this expected that I would see the Execution history for both bionic and horizon in the horizon rule’s execution history? But in bionic, it looks like the same concept does not apply - i.e. we only see the bionic rule’s execution history?

 

Reply


Terms & ConditionsCookie settings