Oct 29, 2026
How to Close the Year Without Burning Out in December: A Checklist for Leaders
Seven decisions to make in the next few weeks, each with a one-minute test question. Save this one. Decisions made in December set the budgets, owners, and capacity of the first quarter. Decisions left open get made anyway, usually under deadline pressure, by whoever happens to be available. Closing these areas in the coming weeks spreads the work out, instead of leaving all of it for the last two weeks of the yea…
Seven decisions to make in the next few weeks, each with a one-minute test question. Save this one.

Decisions made in December set the budgets, owners, and capacity of the first quarter. Decisions left open get made anyway, usually under deadline pressure, by whoever happens to be available. Closing these areas in the coming weeks spreads the work out, instead of leaving all of it for the last two weeks of the year.
Below are seven areas worth closing before December 31. Each has a test question you can answer in about a minute. A confident answer means the area is covered. “Not sure” means it belongs on the December task list.
1. The technology budget for next year
List every line: licenses, infrastructure, API and token costs of AI tools, external teams. Give each line a named owner and an expected figure per quarter.
AI tooling needs particular attention because much of it is billed by usage. A flat annual license is easy to plan, while consumption grows with every new use case and every new person who adopts the tool.
Test question: how much does the company’s AI tooling cost per month, and who can answer in a minute?
2. The roadmap for the first two quarters
Write down what you definitely do, what you deliberately postpone, and which number will show that each priority worked. Three to five priorities get delivered more reliably than ten, because every extra priority takes capacity from the others.
Test question: name one thing the team is not doing in Q1 and say why. If everything is a priority, nobody has decided yet.
3. Rules for working with AI
Agree in writing which tools are allowed, which data may be passed to models, who reviews the result before it reaches production, and how the company talks to clients about its use of AI. For products on the EU market, check with your legal advisers which GDPR and AI Act obligations apply to your product and when.
Without a written version, every team interprets the rules on its own, and the differences surface at the worst moment: during a client audit or an investor’s due diligence.
Test question: can a new employee find these rules in ten minutes?
4. Key people, knowledge, and holiday weeks
For every critical system, at least two people should understand how it works, and documentation should let a new specialist get oriented without weeks of explanation. Documentation that exists only in one person’s head is a single point of failure with a name and a vacation schedule.
December adds a practical layer: who is reachable during the holidays, how an incident gets escalated, and when releases are frozen.
Test question: if the one person who knows a critical system is unreachable for a week, what happens?
5. Access, security, and contractors
Review roles and revoke access for everyone who has finished working with you, including contractors and former employees. Rotate secrets. Confirm that a backup can actually be restored. That last item is the one most often skipped: until a restore has been tested, nobody knows whether the backup works.
Then go through the contracts. Mark which agreements end on December 31, what you renew, and what you revisit, and check the clauses on replacing a specialist, transferring knowledge, and code ownership while they can be negotiated calmly. Repositories, credentials, and documentation should sit in accounts you control, regardless of who builds the product.
Test questions: when was the last time a backup was restored, not merely created? If a contractor stopped working tomorrow, who would hold the access, the credentials, and the knowledge?
6. The team for next year
Separate permanent needs from needs tied to a project or phase. Expertise in implementing an AI agent or building a critical integration often has the second shape: intensive for a period, much lighter afterwards. Workday’s October 2026 report found that demand for hands-on AI skills, such as building tools and automating workflows, rose 51% between September 2025 and July 2026, while the time to fill a role stayed near 60 days, so a position opened in January may be filled in March. For each planned hire, compare the full cost, including recruiting, onboarding, and time to full productivity, with the alternative of bringing expertise in for the period you need it.
Test question: for each open position, is the need permanent, or tied to a phase of work?
7. The report for investors and the board
Prepare the metrics, the risks, and the answers to the questions that come up in due diligence: who pays, how much, how long they stay, and what the architecture looks like. These answers take days to assemble properly. In December there is time to do it. In the middle of a funding round there is not.
Test question: if an investor asked for a technical risk summary this week, how long would it take to produce one?
How to use the list
Block one session with the people who own these areas. Go through the seven areas, and for every “not sure” assign an owner and a date before the holidays. Anything still open on December 31 becomes a first-quarter problem, and the first quarter has its own priorities.
Wamisoftware has been building software products since 2014. Our senior engineers work together with AI agents on projects where architecture, security, and clear ownership decide how the product grows.


