As a product manager for an existing product, how do you choose between launching a new feature and fixing an existing bug?

Clarifying questions

1. is the product app or web based? app
2. Is the bug app or web based? app
3. Is it global geo or a particular market? global
4. Assume the bug is not critical/can cause outage/urgent? correct
5. Is the feature an upgrade that has been deprioritized multiple times? no


I could ask more questions around the nature of the bug or feature but for now I will assume that the bug would remove a customer irritant (as opposed to an internal impacting bug like measurement etc) and the feature would be a net positive enhancement.

Restate
As a pm for a product there is a potential new feature that would enhance the app and a bug that is not critical but would provide more ease of use. How would you choose between launching the feature or fixing the bug?

Generally I will need to do some sort of impact vs effort analysis but I will also need to consider risks. It is worth noting that the impact of even a seeming insignificant irritant on a customer can be outsized in comparison to the positive impact of an enhancement so the impact is not apples to apples, making this decision that much more nuanced.

First:
Risks
1. how many users impacted (bug/feature)?
2. How many sessions expected to be impacted (bug/feature)?
3. Is the feature a nice to have or considered necessary for strategic/competitive or other reasons?
4. Is there a risk of downstream impacts of not fixing the bug - ie. will it potentially be amplified with other updates

Second:
Classify impact of both bug fix and feature as N/L/M/H on the following list of key customer functions which drive engagement and contribute to experience and revenue within the app:

1. ability to order
2. ability to onboard new customers
3. ability to search and retrieve relevant products
4. ability to view product information
5. ability to make a return
6. ability to track order and return status
7. ability to update account information

Third:
investigate the overall effort of building the feature vs fixing the bug. 

Effort for feature
1. is the data available?
2. are advanced techniques ie. GenAI needed?
3. does the feature require a net new tool to be built?
4. Is there a significant engineering lift?
5. Is there significant compute power and infrastructure changes needed?

Effort for bug fix
1. assume the bug cannot crash the system or cause an outage since not critical
2. is there a high liklihood that fixing the bug could lead to new bugs?
3. Is the bug appearing in a complex code base
4. Is there a major update coming where this bug could be in scope (so no need for off cycle priority)
5. Is there a bug history of similar bugs that can give insight into the fix?
6. Is the bug impacting a feature that will be deprecated (this isn't directly effort but should be considered)
7. Can the bug be reproduced/tested?

Fourth:
Impact of bug on systems:
1. number of systems
2. potential for system failure
3. impact on UI
4. Impact on downstream systems
5. etc

Utilize infor above around risks/impacts to assign a category to the bug fix based on a) severity (system impact), b) impact (customer functionality impact), c) response (urgency or can wait to future scheduled update):
1. Drop everything and fix (we already stated we are not here)
2. high priority
3. important but not urgent
4. fix when possible
5. lowest priority
There is a well used hierarchy of bug classification - assume we employ this based on our qualitative assessments above.

Fifth: 
1. count the number out of 7 customer impact activities that are M/H for the a) feature and the same for the b) bug.
2. assess the bug fix category
3. If bug fix category < 3 bug must be fixed first
4. If bug fix >= 3 and a>=b implement the feature, a

Essentially if the feature can make a higher or same impact than bug fix and bug fix is of a non urgent classification implement the feature.