Knack Performance and Reliability

I see a great deal of frustration about app performance and reliability being voiced in this and other forums, and I am hoping you’ll spend a moment with me to read this. Before going any further though, I would like to thank each one of you who has been vocal, and acknowledge that there are
more Knack users who have been silently frustrated. It is a signal of how passionate you are about Knack, to the point where for some of you, it’s your entire business model and source of income.

I would also like to call out collective appreciation for @Bill and @Marek and everyone on their team for their tireless efforts in trying to pinpoint specific issues and relay them back to the Knack engineering team.

While the majority of Knack users have not had a degraded experience since the change in database technology between August and October, there are those who obviously have.
The database change is one of many that has taken place recently at Knack and this post an attempt to give everyone transparency and visibility into these changes and how Knack is addressing them.

Starting with performance and reliability:

Performance Examples: this relates to (but is not limited to):

  • Form submissions
  • Pivot table rendering
  • Returning results on grouped reports
  • Knack Spinners

Reliability examples: this relates to (but is not limited to):

  • 4xx, 5xx, and network errors
  • Task Failures

There is not one person at Knack who does not deeply care about making performance and reliability better. I feel very strongly that this is a true statement.

There has been a lack of communication from Knack to users regarding visibility into what we are already doing and plan on doing to address this.
I hope that we are able to regain some of your trust, starting today.

We have had our entire back-end engineering staff working on performance and reliability improvements.

With the recent benefit of the Resurgens investment, Knack can double our engineering team and quadruple our QA team in the next 6 months.
We will be able to hire additional Support staff so your tickets get attention sooner.
We have been able to invest in tooling for benchmark testing so we have actual proof of improvements, and if not, quickly pivot and figure out what’s not working.
We have brought in industry experts that are providing guidance on maximizing database queries. We have been unraveling the causes of 5xx errors and correcting those as fast as possible.
Each Engineering sprint has a specific focus on the performance AND reliability side- improving report speed, improving form submissions reliability, resolving x number of 5xx errors.

It will take time for some of you to see benefits.

We have confidence that in 3-5 months every user will be back at parity with the pre-SQL experience.
It will take approximately 8-9 months to excessively surpass expectations such that performance and reliability become such a non-issue that hopefully this post ends up at the bottom of a dusty long-lost thread.

There are some complex Live Apps that are slow to load pages or render a table or a half-dozen other areas. That is a different project being worked on separately than the performance and reliability issues mentioned above. We are in the very early stages of mapping out a plan that offers the most beautiful, functional, and modern (but non-breaking) Live App experience.
I do not have a timeline for this yet.

I have read comments that rightfully so, ask us not to release any new features and only focus on performance. I deeply understand that point of view, and would like to reassure you that any new feature development is absolutely not taking away from the laser-focus on performance and reliability work.
That is solely a product of me joining Knack in mid-October and being able to partner with fairly new front-end (think user-interface/user experience) engineers who have a different skill set and background than the back-end engineers. It didn’t take long for us to create a backlog of ‘friction points’ that we could quickly start knocking out, in conjunction with planning the future state of Knack.

If you have read this far, I thank you. As someone fairly new to Knack, one of the reasons I joined Knack and am excited to come to work every day (from my living room) is because I saw how deeply
Knack users and Knacksters care about Knack.

I’m not on the Board, I’m not an executive, I’m just the new product team. And I sincerely give a shit about what pains you and delights you. I have learned so much from those of you I have already met and I am excited to learn more.
I would love to open my calendar up to everyone, but I am also trying to build out this team, starting with a Product Designer :smiling_face:
I am very proud and humbled that I was able to join Knack. They are super smart, dedicated, and care deeply about each Knack user. From my end, I promise more transparency from Knack from this point on - you’ll get sick of me quickly, I’m pretty certain.

If you want to chat, please email

Thank you,

Hi Kara,

Thanks for your candid reply, it’s very reassuring.

I would imagine most of your users understand the challenges of running a platform such as knack, but being left in the dark, not knowing if there are even any staff left, or that you are even aware of care about the faults is unforgivable… what do we tell our clients or users… i’m sorry your application no longer works, or even worse is giving false results… this is the first time anyone has acknowledged the larger problem, and been open about time scales. I’m reassured now you guys are on this, but it has to be fed back that you as a company need to communicate, even if it’s bad news… especially when you continue to take money from us and our clients for a platform that in some instances is not working. I also feel if things will be sub par for up to 6 months or more, then you do need to think about your charges too, although i would be far happier to have a commitment of communication - than a reduction in charges.

keep up the good work - we used to love the platform, and will do so again over the coming weeks / months and look forward to seeing all the changes, not just in terms of performance but in front end design too.