I read today in Knack’s release notes that they will be implementing a task limit of 10 active tasks per table and only allowed to run any task 10 times per day, and we have 2 weeks before this restriction starts to come into place. With no warning unless you read their release notices. My view:
(1) It is unreasonable to impose this change without giving a reasonable amount of notice to folks - measured in MONTHS. This may need people to rearchitect their app. I can see a whole bunch of broken apps in a couple of weeks.
(2) 10 is a very miserly limit, in both cases. It is not hard to see some simple use cases that will exceed this quickly. And there is no workaround other than taking out an ENTERPRISE license, which will make Knack unaffordable for many.
(3) Putting that limit on MANUAL execution (i.e. no more than 10 times per day) is crazy. What if I’m developing a task and need to test it? 10 tests per day and then what - go to the beach and wait until tomorrow?
(4) Where is the proactive comms? For most Knack customers the environment is “set and forget”. they don’t read release notes. They don’t even log into the builder. The first they will know of this is when some of their tasks stop running and their app breaks or their data integrity falls to pieces. Not very “customer centric” Knack.
I would like Knack to reconsider this change and maybe take a more customer aware approach. We seem to be getting a lot “less for more” from Knack recently - prices increasing by 50%, with functionality being removed.
And I also wanted to raise this to everyone’s attention, given Knack quietly buried it at the back of their latest release notice … it might give you an opportunity to do something about it.
From Knacks Release Notes:
UPCOMING CHANGE:
Task limits Starting January 28th 2025, Knack will limit each task to run a maximum of 10 times per day. Starting March 4th 2025, each table can have a maximum of 10 active tasks. For more details on Tasks limits, please visit our Knowledge Base.
This does seem like anyone who is using Knack for anything business critical is being forced into an enterprise payment situation. Which for many that simply will make Knack no longer the great solution it used to be.
Yep - its becoming a deal breaker for me. I currently have an app that runs 14 tasks on one table every day - these are mostly simple tasks to email out reminders when a sales lead hasn’t been updated in a certain number of days: 7, 15, 30, 45, 60, etc. As tasks provide no way to do conditional stuff within the task (e.g. if 15 days send this else if 30 days send this else … like you can in record rules for example) the ONLY way to achieve this spread of emails is using a discrete task for each time period. I see NO WAY to re-engineer this function to less than 10 tasks without creating a “clone table” that I move some of the tasks to, which is sort of stupid and likely to cause me all sorts of grief in future. And I’ll guarantee the company wont move to an enterprise license - the app isn’t worth that much to them. It has been running for 8 years now. I doubt it will see another one the way things are going.
If anyone has another idea on how to solve this issue I’m all ears.
And Knack, if you read this, please explain to me what YOU think I should do (and every other longstanding loyal customer in my situation). And don’t say “buy an enterprise license”. That isn’t an option.
And don’t suggest I “use flows instead”. I am a SHARED BUILDER for this business, which means I don’t have any access to use flows. Maybe fix this constraint before you start taking away the alternatives.
Have you looked at using make ? - I can only highly recommend @CarlHolmes for any assistance you may need. His YouTube videos are a massive help and he may have some insight into how this could be achieved using a 3rd party integration.
I am in agreement that they should have made sure that all users had access to flows before pulling the rug out from under Knack’s users. Plus 2 weeks notice is not be a long time to find work arounds.
I have to agree with @NickKnack Make is the easiest solution to running scheduled and/or triggered tasks. I don’t believe you can schedule tasks yet in Flows but I haven’t been able to access flows as I am also shared builder on most of my apps. I think Make is reasonably priced but still more expensive than tasks. I know users who have a lot of tasks this is going to cause havoc with.
Thank you all on this thread for the important feedback! I am glad we getting input from our builders on these intended changes… I will continue monitoring this thread for feedback over the next few days, then reply back with comments. I am open to altering planned limits on Tasks based on the feedback from the Community
@LeighBerre96926 - I was informed of this change by @CSWinnall last night. It’s highly likely that such a significant update may have simply slipped past me. Like many others, I have a number of clients who use Tasks. Most of them only use a couple, but I do have some edge cases, which I shared with @JohnKaras today, that exceed the limit of 10.
John and I have been discussing this on Slack, and I am awaiting an update on the situation. I have shared this thread with him directly, so I expect he will join the conversation soon.
I hope there will be some flexibility with this new change, as it impacts many users.
I understand the transition to Flows, but currently, you can’t send a “Knack” email, which comes from SendGrid, meaning integration with Gmail or Outlook would be necessary.
While I was typing, I noticed that John has posted a reply, so I will sign off here. I would appreciate any flexibility in this decision.
I think Knack need to acknowledge that there are certain use cases where you have no choice but to run multiple tasks to complete things that can be done natively in other platforms. Also the comms on this issue has been bad to just give a short notice period and not allow builders sufficient time to either create alternative third party integrations or migrate to another platform that does not have such limits. This is especially irksome for some users due to recent price changes etc.
The issue here is the blunt instrument you are using to reduce task workload on servers. Limiting number of active tasks per table has huge impacts on APP design.
If that is the only way you can achieve this then my suggestion as to reasonable limits are:
Limit on executions should allow 1 run per hour. There are many use cases where this might be required So 25 times per day max.
Limit on active tasks is harder to provide logic for, but 10 isn’t enough. 20 would solve all my concerns but maybe not others. Surely you folks can query what is actually out there right now to get some feel for a threshold that isn’t going to undermine existing users.
BUT LET ME OFFER ANOTHER APPROACH…
It appears the outcome you are looking for is to reduce strain on task servers.
Perhaps you should be targeting the overall number of task executions per table rather than setting limits on active / runs per table. As in 100 task executions per table per day (10 x 10). Same result but different approach.
This would be a far more equitable approach for customers and for Knack. I could easily work within 100 task executions per table per day.
OR EVEN BETTER …
I (and I expect most builders) would be happy to go further and work within an overall APP limit of executions per day … rather than per table. This is the most equitable approach and allows builders to design with the most flexibility.
As a guide, maybe 500 task executions per APP per day. This is actually way more restrictive overall than 10 active 10 times a day per table. But way more flexible and easier to accommodate.
Thanks for the suggestion, but I’d like to keep things within the Knack ecosystem. The customer isn’t going to be thrilled with having to add yet another supplier to its “shadow IT” ecosystem.
And its ironic that this type of Knack policy change potentially drives us AWAY from using Knack functionality - especially at a time when they are pushing Flows so heavily which is effectively their play to STOP us needing 3rd party bolt-ons like Make.
It’s clear the proposed change of “10 active tasks per table” limit will be too restrictive for many of our customers using the valuable Tasks feature… As a result, we are rescinding plans to have a limit on “# of active tasks per table”. You will see broader announcements today in our Release Notes (and the Jan 13th Release Note has been amended to account for this). Please note, the limit of 10 runs per day per task will continue to be implemented as planned on January 28th.
We’ve also heard that customers want to test and leverage Flows, but are not able to do so easily with the current amount of Transactions each plan has included by default… As a result of this great feedback, we will be increasing the # of Transactions each plan receives by default, as well as the Add-On pack amounts and prices. More details will be announced by the end of January.
I hope these changes help our customers. And as always, please keep the feedback coming! Your input helps guide Knack.
Whoa! I very much hope this is NOT the case and they roll this back. Sum filters DO NOT WORK and running tasks to update fields to get around this is the only workaround. If they then limit the tasks this breaks the whole app for most of my customers! Come on knack, don’t be stupid!
Just adding an off-topic note:
New AI tools like https://www.perplexity.ai/search/ help me get instant code to copy/past in the Knack API, even JavaScript. You give it the scene, view, fields, and with enough details it generates the code.
Over the long searches inside the manuals, etc.
So more than ever, Knack as a NO-CODE plataform is the way to go.
