We analyzed over 20,000 strategic plans. We looked at how they track risk. Here's why risk frameworks fail — and how to tie risk to the strategy it protects.
- Start from strategic objectives, then ask what could stop each one. A risk not attached to an objective is just anxiety in a spreadsheet.
- Give every documented risk a named owner. Initiatives with a named owner complete at 2.5x the rate (21.5% vs. 8.7%) in ClearPoint data, and ClearPoint shows which risks still need an owner.
- Score impact and likelihood over one-year and five-year horizons. Slow-building risks that look unlikely next quarter stay visible.
- Pair each KPI with a Key Risk Indicator and a threshold. A rising supplier defect rate can warn you before on-time delivery slips.
- Rebuild your annual exercise with the annual risk assessment template. Then review each risk on your KPI cadence so it never goes stale.
Somewhere in your building, there is a risk register. It might be a spreadsheet. It might be a binder from last year’s offsite. It lists the things that could sink you, and almost no one has opened it since the meeting that made it.
You’re not alone. We analyzed over 20,000 strategic plans for our 2026 Strategic Planning Report and counted how many tracked a risk.
Eighty-six percent contained none.
The median plan we studied tracked six KPIs, one project, and one objective. Risks: zero. Organizations measure what they want to achieve in fine detail. What could stop them gets filed away, out of sight.
Your framework isn’t the problem. NIST, ISO 31000, and COSO are well-built and freely available. The problem is that risk lives in a separate document, cut off from the strategy it’s supposed to protect.
This guide covers the main frameworks, when to use each, and how to run one. It starts with what the data says you’re probably doing wrong.
Risk is the rarest thing in a strategic plan
Share this data:in ShareX PostGet the 2026 Report →
Organizations without risk data are flying blind
Of the 20,000+ plans in our dataset, 11,289 held real content: objectives, KPIs, projects. Among those, 13.8% included a documented risk. ClearPoint shows which plans have documented a risk and which still need to.
What the median strategic plan tracks
Organizations measure their ambitions in detail and their risks not at all.
Risk was the rarest element in a strategic plan. Teams logged measures, milestones, initiatives, and even routine action items far more often.
Once a risk is written down, it needs an owner. In our data, initiatives with a named owner complete at 2.5x the rate (21.5% vs. 8.7%). Risk is the item that most needs a named owner, and ClearPoint shows every documented risk still waiting for one.
When risk is logged, give it an owner
This isn’t a ClearPoint quirk. North Carolina State University’s 2025 State of Risk Oversight surveyed hundreds of organizations. Only 11% said their risk process gave them a strategic edge. Both datasets point the same way: risk belongs in the same room as strategy.
What a risk management framework actually does
A risk management framework does three things. It helps you spot what could go wrong, decide what to do about it, and keep watching so a small problem doesn’t become a headline.
The rest is structure, and structure is where the major frameworks differ. Three matter for most organizations.
COSO ERM — built for strategy
COSO is the one to know first, because of what it did in 2017. It renamed its framework “Enterprise Risk Management — Integrating with Strategy and Performance.” The most influential body in risk management put strategy in the headline, not controls or compliance. COSO’s argument: risk had been treated as a side activity, judged only after the strategy was set. COSO wanted it built in from the first decision.
ISO 31000 — principles over checklists
ISO 31000:2018 is the international standard. It’s principles-based: guidelines, not a controls catalog. Its first principle is one word: “Integrated.” Risk management, ISO says, is part of every organizational activity, not a department or an annual event.
NIST RMF — built for security and compliance
The NIST Risk Management Framework (SP 800-37, Revision 2) runs seven steps: Prepare, Categorize, Select, Implement, Assess, Authorize, Monitor. It’s rigorous, and it’s the backbone of U.S. federal cybersecurity. It’s also the most controls-first of the three, designed to protect information systems rather than set company strategy. If you’re a federal agency or you handle sensitive data, you’ll likely use it. Just know what it’s built for.
| Framework | Best for | Structure | Strategy-integrated? |
|---|---|---|---|
| COSO ERM | Enterprises & boards tying risk to strategy | 5 components, 20 principles | Yes — by design (2017) |
| ISO 31000:2018 | Flexible, any size or sector | Principles + framework + process | Yes — principle #1 Integrated |
| NIST RMF | Federal, security & compliance | 7 steps (Prepare to Monitor) | Indirect — controls-first |
| NIST AI RMF | Organizations deploying AI | 4 functions: Govern, Map, Measure, Manage | Via governance function |
One more, if you’re deploying AI
If you’re putting AI to work, as many in government now are, there’s a fourth worth knowing. NIST released its AI Risk Management Framework in 2023, organized around four functions: Govern, Map, Measure, Manage. In 2024 it added a Generative AI Profile naming twelve risks specific to generative systems, from hallucination to data poisoning. AI risk is its own discipline now. We cover it in depth in our AI Governance guide.
COSO’s 2017 rewrite and ISO’s first principle say the same thing: integrate risk with strategy. The guidance has been clear for nearly a decade. The execution hasn’t followed.
Most teams build their framework backwards
Most teams adopt a framework by starting with its controls. They download the NIST 800-53 catalog. They build a risk register in a tab that links to nothing. They populate it, score it, and feel safer. Then the register and the strategy never speak again.
That’s backwards. A control protects something. A risk threatens something. If you don’t start from what you’re trying to achieve (your strategic objectives), you’re cataloguing hazards in the abstract. A risk that isn’t attached to an objective is just anxiety in a spreadsheet.
Risk tracking tends to follow habit, not danger. Where a regulator forces the discipline, it happens. Where it’s left to good intentions, it doesn’t. The framework was never the real issue.
What 15 years of Balanced Scorecard work taught me about risk
A note from Ted Jackson, co-founder of ClearPoint.
I’ve spent more than fifteen years putting the Balanced Scorecard to work inside hundreds of organizations: owners on every objective, targets on every measure, and a review rhythm that kept it all honest.
The Scorecard works because it makes the upside a system. Goals stop being wishes. They get names attached and dates that matter.
Again and again, I watched the same teams that ran their objectives with that discipline handle risk completely differently. Once a year, in a conference room, someone would lead a brainstorm. The list went into a document. The document went into a drawer. The next time anyone looked at it was eleven months later, in the same room.
We treat the upside as a system and the downside as an event.
The Balanced Scorecard always implied risk. You can’t chase an objective without managing what threatens it. But almost no one extends the Scorecard’s discipline of owners, cadence, and accountability to the things that could derail the plan. A documented risk with no owner is the same gap I watched in those rooms.
How we rebuilt our own risk process
For years, we made the exact mistake the data describes.
Our annual risk assessment was a one-day exercise. The whole company in a room, brainstorming what could go wrong. We’d build a list and assign a few action items. Then the list sat there. We had no follow-up to report on those actions. Worse, it wasn’t linked to our strategy in any meaningful way. It was a separate, standalone effort. The full-room format had another flaw: only managers and senior leaders ever spoke up.
So we rebuilt it. This is the process we run now.
First, each department brainstorms its own risks, tied to the company’s strategic objectives, not in the abstract. A short questionnaire guides the conversation. Customer success looks at what threatens its goals. Product looks at its own. The people closest to the work name the risk.
Second, department heads meet with leadership to clean the list. Cut duplicates. Group by theme. Confirm each risk maps to a real objective. Some risks touch several.
Third, we score each one. We skip the vague high/medium/low. We rate impact on a 1–5 scale, then estimate probability over two horizons: one year and five years. A risk that’s unlikely next quarter but likely over five years gets seen, not buried. The averaged score points to a response: avoid it, reduce it, mitigate it, transfer it, or accept it on purpose.
Fourth, and most important, every risk action item links to a strategic objective inside ClearPoint. We report on them quarterly, in the same reviews where we report on our KPIs.

The change wasn’t the framework. We didn’t adopt COSO or rewrite our charter. The change was location. Risk moved out of the annual binder and into the system we already used every quarter to run the company. More people own it now. It doesn’t go stale.

The strategy-first way to run any framework
You don’t need our exact process. You need the principle underneath it: risk belongs in the same system, on the same cadence, as your strategy. Whatever framework you adopt (COSO, ISO, NIST), here’s how to wire it in.
1. Start from objectives, not hazards. List what you’re trying to achieve first. Then ask, for each one, what could stop it. You’ll end up with risks that matter, not a generic catalog.
2. Source risks from the people closest to the work. Department-up beats top-down. The frontline sees the failure modes leadership can’t.
3. Score on two horizons. Impact times likelihood is standard. Add a time dimension, near-term and long-term, so slow-building risks don’t hide.
The risk matrix: impact × likelihood
It sorts risk. It doesn’t manage it. That’s the job of an owner and a cadence.
Score each risk on impact (1–5) and likelihood across two horizons (1-year and 5-year), then assign a response.
A quick word on the risk matrix, since it’s the tool most people picture. It plots impact against likelihood, usually on a 5×5 grid. High impact and high likelihood: act now. Low and low: watch and wait. It’s useful, but many teams stop there, and that is the mistake.
4. Give every risk an owner. This one step closes the ownership gap. A risk without a name is a risk no one is watching.
5. Choose a response and a control. Avoid, reduce, mitigate, transfer, or accept. Write it down. Assign it.
6. Pair KRIs with your KPIs. A KPI tells you if you’re winning. A Key Risk Indicator warns you before you start losing. If your KPI is on-time delivery, a matching KRI might be supplier defect rate. Set the threshold from your risk appetite, the level of risk you’ve decided you can live with. When the KRI crosses it, you act before the KPI ever moves.
Quick definitions, because the terms get muddled. Risk appetite is qualitative: “we won’t tolerate a compliance breach.” Risk tolerance is the number: “no more than 1.5% credit loss.” Appetite is the principle. Tolerance is the line.

7. Review risk on the same cadence as strategy. Monthly or quarterly, whatever rhythm you already use for KPIs. The goal is simple: risk should never be more than one review cycle away from attention.
Which framework should you actually use?
It matters less than you think. Wiring risk into strategy matters more than the logo on the framework. Here is a simple way to choose.
- Choose COSO ERM if you’re an enterprise, or a board wants risk tied to strategy and performance. It’s the most strategy-native of the three.
- Choose ISO 31000 if you want flexible, principles-based guidance you can adapt to any size or sector. It won’t hand you a checklist. That’s the point.
- Choose NIST RMF if you’re a federal agency, a contractor, or you manage sensitive systems. It’s built for security and compliance, and it’s often required.
- Add the NIST AI RMF if you’re deploying AI. The other three weren’t written with model risk in mind.
Pick one. Then spend your energy on execution, the part that actually fails, not on the choice itself.
The four ways risk frameworks die
Frameworks don’t fail on paper. They fail in practice, in a handful of predictable ways.
The write-only register. Risks get logged and never closed. The register fills up, nothing leaves, and it becomes a museum of worries. ClearPoint shows which documented risks are still open, so the register can be reviewed and closed out.
The orphaned risk. A risk with no owner. Everyone can see it, no one is responsible for it, and it will surprise you.
The annual event. The once-a-year offsite that produces a list and no follow-through. If risk gets attention in one meeting a year, it isn’t managed. It’s commemorated.
The controls-first inversion. Starting with a catalog of controls instead of the objectives you’re protecting. You end up busy and unprotected at the same time.
Put risk where you’ll actually see it
Risk management is the discipline of looking at what could stop you with the same seriousness you bring to what you’re chasing. It isn’t a binder, an annual meeting, or a tab in a spreadsheet. The frameworks and standards already say this. What’s missing is putting risk where you’ll actually see it.
A strategy that only plans for success is a wish. The organizations that last write down what could go wrong and keep looking.
Related Articles
Common questions
Where do we start if our strategic plan has no documented risks?
Start from your objectives, not a hazard catalog. For each objective, ask the people closest to the work what could stop it. Then have department heads and leadership remove duplicates and confirm each risk maps to a real objective. Score impact and likelihood over one-year and five-year horizons, assign an owner, and choose a response.
Who should own a risk?
One named person who can act on it, not a department or committee. Record the owner with the response you chose (avoid, reduce, mitigate, transfer or accept) and the control that carries it out. An unowned risk is one nobody is watching, and ClearPoint shows which risks still need an assigned owner.
How do KRIs and KPIs work together?
A KPI tracks progress toward a goal; a KRI is an early warning that the goal is under threat. Pair them: if your KPI is on-time delivery, a matching KRI might be supplier defect rate. Set the KRI threshold from your risk appetite, and act when it is crossed, before the KPI moves.
How often should we review our risks?
On the same cadence as your strategy, monthly or quarterly, in the same meeting where you review KPIs. The common failure is the annual offsite that produces a list nobody revisits. Risk should never be more than one review cycle away from attention, and each review should check every risk’s owner, score and response.
When should a risk be closed or taken off the register?
Close a risk when its response is complete or the objective it threatened has left the plan, and record why. A register that only grows stops being useful. At each review, confirm every open risk still maps to a live objective and has an active response.

%252520Risk%252520Management%252520Framework.png)






%2520Strategy%2520vs%2520Tactics%2520Differences%252C%2520Examples%2520%2526%2520How%2520to%2520Track%2520Both.jpg)



















.png)




.png)


















.webp)















%2520OKRs%2520vs%2520KPIs%2520Breaking%2520Down%2520the%2520Difference.jpg)