After almost seven years in FP&A, I've stopped being surprised by any of this. But I remember when each one hit me for the first time. Someone will demand a 5-year forecast by end of day. You'll build it. They'll ignore it completely and make decisions based on board pressure. **Build the model anyway. The process of thinking through it is where the value lives, even if the output collects dust.** Sales will miss their forecast by 40%, and finance will get blamed for inaccurate projections. Even though you used their numbers. I've learned to **document everything. Every assumption, every source, every email where someone signed off on a number**. Not because you're building a case against people, but because memory is short when results are bad. Everyone wants to be "data-driven" until the data says something inconvenient. Then suddenly it's "let's look at the qualitative factors." I've seen this happen in almost every budget review I've sat in. **Keep the receipts.** Your meticulously built budget will be irrelevant by February. Leadership will still hold you accountable to it in December. This is genuinely the most disorienting part of the job when you're new. You get used to it. They'll reject your headcount model as too conservative. Then panic when cash gets tight and ask why you didn't flag it. You did. Three times. In writing. (This one still stings.) The same executives who say "we need to be more strategic" will ask you to explain a $200 variance in the monthly actuals. Pick your battles. I've gotten better at knowing which variances actually matter and which ones are just noise that someone noticed. You'll spend 80 hours building a board deck with scenario analyses and sensitivity tables. The board will spend three minutes on it and ask about a footnote. This is normal. I've [[lab/Variance Analysis Bot|automated parts of this process]] now, which helps with the sting. These aren't bugs in the system. They're features of working in finance. My advice after living through all of them: build credibility through consistency, not through being right. Document your recommendations. And remember that **your job isn't to control the decisions. It's to make sure the consequences can't be blamed on bad data.** ← [[Signal|Back to /signal]]