Ranking cybersecurity spend across the NIST CSF 2.0 functions
The NIST framework gives boards a shared language for security spend but does not rank investments for you. Here is a five-pillar method, with tests, and where it breaks.
Most security budgets are set by last year's budget. Lines grow, tools overlap, and a board is left asking which risk the extra money has actually reduced. The NIST Cybersecurity Framework 2.0 gives a leader a shared way to talk about that question. It does not answer it. Ranking investments is a judgement the leader has to make, and the framework is useful mainly because it forces that judgement into the open. This piece sets out five pillars for making the ranking, with a test for each. It is written for the chief executive, finance lead or board member who must sign off the spend, not for the security team that runs it.
Chapter one: Name the losses you are paying to avoid
Start with a list of outcomes you would not accept. A payments outage lasting a day, a release of customer data, a production line stopped for a week, a ransom demand you cannot meet in time. Write six to ten of them in plain language, each with a rough cost and the people who would have to explain it to a regulator, a customer or a board. The list is your risk profile in working form. Without it, every security proposal sounds equally urgent, because each one promises protection against something bad.
The test is simple. For each proposed purchase, ask which loss on the list it reduces and by how much. If the answer is 'all of them', ask again. In our experience, a proposal that claims to help everywhere usually helps nowhere in particular, and the finance lead will find that out at the first quarterly review. NIST's templates for creating and using CSF profiles are a sensible place to structure the list, and they are the first resource to read before inventing a format. NIST says its resources overlap and share themes, so expect to use two or three of them together rather than one.
Chapter two: Score every function against that list
Once the losses are written down, look at the framework's functions as the set of things your organisation does about them. Score each function on two questions. How exposed are we today? How much of that exposure would further spending actually close? The answers will not be precise. The point is to make the gaps visible, so that a function with high exposure and little spend sits above one that is well funded and already comfortable.
Keep the scoring crude. Use three levels, high, medium and low, with a one-line reason written beside each. Do it with the people who run the systems, not in a room of executives who have seen only the summary. Where a score depends on a fact nobody can confirm, such as how long a backup takes to restore, mark it as unknown. An unknown on a high-exposure row is itself a finding, and usually the first item for the next budget.
Take a hypothetical manufacturer with three plants and one finance lead. Its scoring shows high exposure on the plant control systems and almost no spend there, while its email filtering is well funded and rated low on exposure. A board that sees only the total security budget would not notice the mismatch. A board that sees the two rows side by side will ask why the money went where it did, which is the question the exercise is designed to raise.
Chapter three: Check what you already have before you buy
Most organisations own more security capability than they use. Licences renew automatically, tools were bought for a project that has ended, and two products quietly do the same job. Before any new proposal goes to the board, the security lead should list what is already in place against each row of the scoring. The list is dull work, and it is the cheapest saving available.
The test is whether the new spend closes a gap the current tools cannot close, and whether someone is accountable for using what is bought. A new product without an owner, a training budget and a measure of use is a cost, not a control. Ask the internal team what would be switched off if the purchase went ahead. If nobody can name anything, the proposal is probably an addition to a pile rather than a reduction in risk.
Licences deserve particular attention. A renewal that passes without review is a decision made by a vendor's calendar. Put every security renewal on the same list as the proposals, with its date and its owner, and treat a renewal as a purchase that needs the same justification as a new one.
Chapter four: Price recovery as its own line
Spending is often concentrated on prevention and detection, while recovery is treated as an IT cost rather than a security decision. Recovery is where a bad day becomes a survivable one. A leader should ask what it would cost to restore the most important service, and how long that would take. If the honest answer is that nobody has tested it, that gap belongs on the same list as the others.
The test is a restore exercise. Pick the service whose loss you named as most expensive, run a timed restoration from backup in a test environment, and record how long it took and what was missing. Budget for the gap that exercise reveals before buying another detection tool. This is judgement rather than a rule. Check it with whoever owns business continuity, because their definition of an acceptable outage may differ from yours.
Insurance and recovery are sometimes confused. Insurance may pay for some losses; it does not restore a service or a customer's trust. Whether a given policy covers a given loss is a question for the insurer and the company's own lawyers, not for the security team. Put it to them in writing.
Chapter five: Review on a fixed rhythm
A ranking made in January is out of date by the summer. Threats change, systems are replaced, and staff leave carrying the knowledge of how a control works. Set a fixed review, quarterly if the losses list is short and the spend large, and treat the scores as provisional until they have been tested. The aim is not a perfect ranking. It is a ranking that everyone can see changing, and can see changing for a reason.
The test for a review is whether each score moved for a reason someone can write down. A row that changed because a system was retired is healthy. A row that changed because a vendor pitched a new product is not. Keep the scoring sheet in the board pack, with the date and the names of the people who scored it, so that the next chair can see what was decided and why.
Before the next security budget goes to the board
Answer each question yes or no. Any no is a gap to close before the spend is approved.
- Have we written down six to ten specific losses we want to avoid?
- Does each proposed purchase name the loss it reduces?
- Has each framework function been scored for exposure and for the effect of further spend?
- Have we listed what we already own against each row of the scoring?
- Does every new tool have a named owner and a measure of use?
- Have we tested the restoration of our most critical service in the last twelve months?
- Is the ranking reviewed on a fixed date, with the reasons for each change recorded?
Outgrace checklist
Where it breaks
The framework is a vocabulary, not a budget. NIST's pages say what it is meant to do and point to templates and resources. On the NIST page we reviewed, there is no ranking method, no rule for matching spend to a profile and no budget guidance. Anyone who says the functions produce a percentage split of spend has added something the sources do not say. The implementation guidance may go further, but it was not in the material we checked, and it should be read before any organisation adopts this method under regulatory scrutiny.
The second break is ownership. A ranking that sits with the security team alone will be read as a request for money, and the finance lead will discount it. A ranking owned by the chief executive, with the security lead presenting it, is more likely to survive the budget round. Expect overlap between NIST's resources and do not treat any one of them as complete.
For practical material, the Software Engineering Institute names the Department of War, government agencies and private industry as partners on its homepage. It says most of its research outputs and educational materials are free of charge, that more than 100 of its software and security tools are freely available on its GitHub site, and that it runs training on topics including insider threat. These are useful places to look. Check each one against your own losses list before you rely on it, and do not assume a resource covers your sector or your regulator.
On Monday, do one thing: write the losses list. Six to ten lines, each with a plain-language cost and a named owner. Send it to the finance lead and the security lead with a request to score the existing tools against it before any new proposal is discussed. Put the list on the next board agenda as a single item, not a slide deck. If the list is hard to write, that difficulty is the first finding, and it will tell you more about the next budget than any vendor's demonstration.
