Roast my idea
I have been thinking about this Idea for a while, would like to get your thoughts and decide whether I should proceed with a Landing/further validation steps 🚀
The product is a cloud cost dashboard to enable collaboration of non-technical people (management) and technical people (devs/infra/ops). As I work on the technical side of things and deal with cloud costs and cost optimisation, I find it challenging to draw the big picture and properly present cloud cost to the management. Cloud provider tools seem to be too detailed and require technical knowledge. Non-technical people are almost never able to break down the costs into buckets without the support of developers.
To add problems on top, modern startups use lots of other third-party apps in addition to cloud provider cost.
So the USP is a cloud cost monitoring tool focused towards non-tech people, where they can do most of their job without the need to involve developers.
Would be great if there is any tech management out here to comment on this 😄
Having build such a thing in the past (get some data, put in Excel, pivot, make fancy, copy/paste into slide deck for reporting), why do you need developers for this? Are you targeting a very specific target group that needs lots of detail?
Usually license & saas cost tracking is done by someone in finance (with or without help from someone below the CTO) and usually the costs are only broken down by supplier, maybe further down to e.g. per seat (depends on the company).
What is your skill set and experience in this area? Who's the target group? What specific problem does your solution solve? And what is solving this problem worth to your potential customers?
That is exactly my thought that [Get data + Excel + pivot + make fancy] sounds like a good automation candidate. I was tasked to do some part of it as a technical person, when the management wanted more details, like how much of our cloud budget did we spend on product X, or (in ideal world) on feature X. So this is a completely different dimension, rather than breaking down cost by supplier - one can understand the profitability of software products and teams. This understanding can help to prioritise work across projects and improve profitability/efficiency (as i see it...).
Another potential type of questions is to cover different aspects of running and developing software, like "how much do we spend on the recoverability and resilience", or "how much do we spend on security infrastructure"?
The benefits of being able to answer these questions is visibility and better control over software operations.
I am kind of struggling to understand the right scope, this is less interesting for small 1-product companies, there should be multiple projects and teams. But then again, I am looking at this from bottom-up - I do not know if this kind of thing is valuable for the management. Maybe slicing the money spent by approximate percentage (like 60% of our cloud spend goes to product A, 20% to product B, and 20% is shared) - is just enough to make decisions.
BTW Thank you for asking good questions, they help me to think about the problem rather than the solution 😅
This is a pretty advanced topic, not sure if IH is the right place for this, as most of us are working indeed on 1-product-companies.
How I have seen implemented this is with tags, e.g. in Azure or AWS. Just tag everything by at least environment, cost center, and ideally product name and ressource owner and you can do some nice analysis already in their billing dashboards. It'd be great if people would use tags in detail (e.g. this is for increased resilience), but I have never seen a tag implementation with more than 3-4 tags per ressource (but then, only seen implementations in companies with < 1k staff either).
Tags enable controlling to check against budgets and P&L line items (e.g. production env cost is part of COGS and all other envs are R&D costs) and can be regularly checked by all people involved (pull the report, email it). If you are so large that you have multiple subscriptions, environments, countries, departments and legal entities this totally makes sense, and for sure is already done in some capacity by company group controlling or company group IT management (depends on who owns IT).
The billing dashboards and API are already pretty good. Maybe some sort of integration with e.g. SAP or Netsuite would be an idea. Not sure.
As you said, it all depends on scope. Before you define that, better make sure this really is a problem that really needs solving and your time invested.