Hacker News Reader

Comment by advaita | Hacker News Reader

Slightly off-topic, but I have been raising this point (after learning the lesson hard way) within my larger engineering org and seem to be getting nowhere towards convincing people that the #1 reason for degraded quality of delivery is too many big ticket changes going on at same time.

What's the playbook for this? Any recommended reading etc would be a huge favor.

PS: Not necessarily a question to OP but anyone passing by.

View on HN

The best advice I've seen on this is, whenever someone new comes up, ask "what should be deprioritised to accomodate this new thing?".

And the complimentary - which feature do you want delivered first?

It should be a single backlog for the whole company (obv divided up by product or component of whatever for ease of managing) but a single list of features in order.

This will of course mean that the decision makers will need to sort out their politics and divisions to agree an order. This is usually hidden by giving vague and conflicting directions to the rest of the business and hoping it will all sort out.

By solving the politics at the top the organisation will benefit greatly from clarity and resource allocation.

Chances are high that even asking for this will reveal enough problems that you can decide if staying is worth it.

Worth looking into SRE concepts such as Error budget and SLO/SLIs if you want to consider one way of tackling this at the org level. It does however need senior buy-in (like most things) to be effective.

Reply on news.ycombinator.com