Hello everyone. I have some news. Good news. If you are in a hurry, have a look at the end of the post. In recent months I had some time to think calmly, a rare thing, and decide 'what I want to do when I grow up'. And it was too much...I might never be able to answer that question. So I focused on a simpler question: 'what do I want to do here and now?'. It is simpler and allows me to test my feelings immediately. Some discoveries bubbled up from the depth of my mind. One of the first ones is that **I still love FinOps**. More than I anticipated. The mix of technical, financial, diplomat and novelty still attracts me. I want to help FinOps find its marks and expand. I see the work of the FinOps Foundation as a series of experiments based on Practitioners' hypothesis and I want to be part of that. This also leaves some unexplored parts of the FinOps jungle to explore. There is so much to do in FinOps. Another discovery is that the most complex topics in FinOps (Hi forecast) seem frustratingly blocked in their initial, theoretical, state. There is not much progress in the How to do stuff. At the first FinOps X in Austin, I did a short 10 minutes presentation on Forecasting, and now years later that talk is still relevant, instead of being old. The number of tools is increasing and the State of FinOps shows that 'Investment and tooling' is up and growing, that is proof of a need. Is the need because FinOps is expanding in new areas, or is it due to the inability of existing tools to deliver the quality needed? Not sure. My most recent frustration is about dashboards. I love dashboards. Everyone does dashboards. We all do dashboards because they are easy to build in a standardized format, sell as cool both to customers and investors. Dashboards represent facts and their content comes from the underlying data, not the needs of the readers. Reports, on the other side, are more complicated, they are specific to the recipient and hard to standardize at scale. Yet, as a user I want to understand, and sometimes act, without the need to grasp all the facts in a dashboard. As a VP of engineering, I want a daily message that tells me that everything is all right (or not) without me having to dive deep into the data. And that 'all right' message has different context on a normal day or on a big release day or black Friday. I want my context to be included. And that is hard, non-standard, not good for VC funding. Context is what makes things solve our problems. Context is why we all take the data from a tool and put it into Excel: we make the data fit for our context to solve our problem. Generic tools solve generic problems, specialised tools solve specialised problems. Context is everything and everywhere. There is a context for the industry, one specific about your company, the day of the month (end of quarters), and so much more. This is why we love GenAI, it is adapted to our context. I started looking at how this is done in Enterprises. What tools do you use in supply chain when not one chain is the same as the other, or in sales where no deal is like the previous one. This is when you go into Enterprise Software, the SAP, Salesforce or Oracle of this world. How do they integrate context? The answer seems to be a strong yet flexible standard (CRM, ERP, DB). Based on the standard is a standard software (it is not a typo) implementing that standard. Then, there is a flexible layer allowing customization. But I am no expert and not planning to become one. So, I went back to things I know: the shell (zsh, bash and all the others). They have a strong core of specialised tools (ls, grep, find ...) and allow building incredibly context specific tools. The magic ingredient is the pipe, it provides a way to connect one specialised command to the other creating a context specific result. I am of the belief that if context is everything, then we need specialised software that can be connected easily. AWS and the cloud are also like this, and this is why I am also back being enthusiastic about the technical capabilities of Cloud (less so on the current pricing trends). The approach has strong business repercussions too. If you are that specialised company, you will fight and keep fighting to be the best in your speciality, but then, when you are the best, focus on growing the pies that use your product. Be an evangelist before being a crusader. To not only talk, but do, I am starting, during my free time, my set of tools that you can use on terminal and connect with pipes. https://github.com/fcontrepois/finops-toolkit, it is open source and I hope that my initial specialised tools will be vastly improved by volunteers. I am not the expert that can create the best-in-class tool, but I can try to kick-start it. But the most important news is that after some amazing interviews and offers (thank you for the opportunity and I hope to help in the future), enjoying being more at home to get the kids, and participate in the sucess of my wife's business, I decided to build **Coblan Ltd**. A new company, based on the 'Context is Everything' idea. The company's ethos is innovative, transparent, and opinionated. Coblan has a first customer: *Stacklet*, a company helping transform high level governance (very context specific) from words into actions. We are building a FinOps specific set of governance policies and books to help. Ask me about it.