Hi everyone, I’m looking for clarification regarding the AI Mentor’s Technical Debt analysis, specifically the metric “% High Tech Debt” shown in the dashboard (example: 45% High Tech Debt — image attached). I have a few questions: How risky is it when AI Mentor shows a high percentage of Technical Debt? What kind of impact can this have on application performance, maintainability, and future scalability?
What is considered a normal or acceptable Technical Debt percentage in typical OutSystems applications?
Is there any recommended threshold or benchmark from OutSystems?
At what point should I worry and take action to reduce it?
Any guidance or best practices from your experience would be really helpful.
Thanks!
Hello Ajit,
Hope you're doing well.
"How risky is it when AI Mentor shows a high percentage of Technical Debt? What kind of impact can this have on application performance, maintainability, and future scalability?"
A high percentage of technical debt does not mean automatically that your application is broken, but it means that you have issues. The real risk depends on the type of issues that you're facing (Does this technical debt affect the performance? Is it blocking upgrades and scalability? Is it slowing the development velocity? Is it impacting security or architecture?)
For example, like Sherif mentioned, missing a description on an element is not equivalent to having a circular or upward reference in terms of severity. However, missing a description will still be considered as technical debt. But it doesn't affect your app. On the other hand, having an architectural problem may lead into several challenges on deploys or scalability.
In your case, you should check exactly what types of findings is AI Mentor detecting, but a high percentage of technical debt clearly indicates that the app deserves your attention.
"What is considered a normal or acceptable Technical Debt percentage in typical OutSystems applications? Is there any recommended threshold or benchmark from OutSystems?"
There is no official recommended threshold or benchmark from OutSystems.
However the AI Mentor has an heat map that will help you to identify the critical areas of attention. Your goal should always be the green, which represents an healthy state.
"At what point should I worry and take action to reduce it?"
Great question. In my perspective, it all depends on the issues that you're facing.
As some examples, I would say that you should worry if:
Also I would like to suggest that you take a look into this documentation:
https://success.outsystems.com/documentation/11/monitoring_and_troubleshooting_apps/manage_technical_debt/handling_technical_debt_with_outsystems/
Hope that this helps you!
Kind regards,
Rui Barradas
I'd say it depends on the type of technical debt. For example, if there are missing descriptions in elements in Service Studio, AI Mentor will count it as technical debt that should be handled. Does it hurt performance? I don’t think so.
On the other hand, if we’re talking about issues like cyclic references or upward references, those could affect deployments and create unnecessary dependencies between apps. Similarly, aggregates or SQL queries inside loops could impact performance, and aggregates or SQL without a MaxRecords setting would also affect performance.
So, I’d suggest starting with the small things to reduce the percentage, such as adding descriptions and comments inside logic or setting max records number to sql or aggregates, and then move on to the bigger issues like architecture, which may require more effort.
Hi @Ajit Kurane ,
U add AI Mentor Studio, It will analyze and generate a report. It look like a checklist containing many questions about the techdebt. Please read it carefully and solve them.
Tech debt is categorized by level, and you need to deliver the product to the customer once the tech debt has been upgraded to green.
Hope this helps!
Dear Ajit, In AI Mentor Studio, If you navigate to the Maintenance section (a sample screenshot below) . You can find the list of code patterns as its impact level if they are not met. they are also categorized into 4 categories by default (Architecture, Performance, Security and Maintainability)Resolving the debts would automatically start to reflect on your apps health and as OutSystems recommends please try to keep the apps health in green.In addition to that it would be wise to go through all code patterns to check if they also meet your organization's best practices and guidelines and bump up the impact levels if neededHope this helps, Cheers, Praveen
Hello.
This is duplicated from https://www.outsystems.com/forums/discussion/105800/question-about-ai-mentor-studio-high-technical-debt/
Please close one of them.