Loading...
Most strategy teams can say what happened last quarter, but not why a decision made two years ago was right. Score your team across five dimensions of strategic memory erosion.
The Strategic Memory Erosion Index (SMEI)
Most strategy teams can tell you what happened last quarter. Few can tell you why a decision made two years ago was right — or wrong — and what they learned from it.
That's not a personnel problem. It's a structural one.
Data overload is silently eroding your team's ability to retain, retrieve, and apply strategic knowledge. Every new tool, dashboard, and alert feed adds to the noise. The signal — the hard-won judgment your team built over years — gets buried under it.
Information overload costs the global economy an estimated $1 trillion annually in lost productivity, degraded decision-making, and increased errors. Inside your strategy function, the cost is measured differently: in repeated mistakes, recycled debates, and decisions that don't stick.
This diagnostic measures whether data overload is eroding your strategic memory — and gives you a concrete score you can act on.
Rate your team on each of the five dimensions below. Score 1 (never true) to 5 (always true). Be honest — this works best when you're uncomfortable.
| Score | Can your team answer: "Why did we decide that?" |
|---|---|
| 1 | Decisions are made in meetings with no formal record beyond scattered notes. Nobody can trace the "why" behind a strategic choice made more than 6 months ago. |
| 2 | Some decisions are documented, but rationale is thin — a bullet point, not a reasoning chain. |
| 3 | Major decisions include written rationale, but it's inconsistently stored across tools and people. |
| 4 | Decisions are documented with context, tradeoffs considered, and alternatives rejected. Most team members can find and reference them. |
| 5 | Every strategic decision is recorded with full reasoning, decision rights, dissent logged, and a scheduled retrospective. Your team could reconstruct the logic of any decision from the last 3 years. |
Your score for Dimension 1, out of 5.
| Score | When a strategy fails or succeeds, does the learning survive? |
|---|---|
| 1 | Lessons live in individual memories. When someone leaves, the lesson leaves with them. |
| 2 | Post-mortems happen but are rarely referenced after the fact. Same mistakes recur across teams. |
| 3 | Key learnings are documented in a shared space, but nobody systematically reviews them before starting new work. |
| 4 | A formal lessons library exists and is actively consulted during strategy kickoffs. Patterns across projects are visible. |
| 5 | Learning is institutionalized: every strategic cycle begins with a review of prior decisions and their outcomes. The team has a documented record of what worked, what didn't, and why — and uses it. |
Your score for Dimension 2, out of 5.
| Score | Of the intelligence your team receives weekly, how much drives a decision? |
|---|---|
| 1 | Your team consumes dozens of alerts, reports, and dashboards. Less than 10% ever leads to a strategic action. |
| 2 | Some feeds are useful, but the team spends more time filtering than deciding. |
| 3 | Key intelligence sources are identified, but synthesis is manual and inconsistent. |
| 4 | Intelligence is curated and tied to specific decision cycles. Noise is actively managed. |
| 5 | Every intelligence input has a named decision it feeds. The team can explain, in one sentence, what each data source is supposed to change. Everything else is cut. |
Your score for Dimension 3, out of 5.
| Score | How fast does raw data become a decision-ready recommendation? |
|---|---|
| 1 | Weeks. Data sits in dashboards until someone manually pulls, cleans, and interprets it for a specific request. |
| 2 | Regular reports are produced, but they describe rather than recommend. The "so what" is left to the reader. |
| 3 | Synthesis happens on a predictable cadence, but recommendations still require significant manual effort. |
| 4 | Most intelligence is synthesized into decision memos within days, with clear recommended actions. |
| 5 | Raw signals are converted into decision-ready recommendations in hours. The bottleneck is the decision, never the synthesis. |
Your score for Dimension 4, out of 5.
| Score | Who owns the team's strategic knowledge — and keeps it alive? |
|---|---|
| 1 | Nobody. Knowledge lives in Slack threads, email chains, and departing employees' heads. |
| 2 | One or two people informally act as "the historians," but there's no system behind them. |
| 3 | A shared drive or wiki exists, but it's inconsistently updated and rarely trusted as the single source of truth. |
| 4 | Strategic memory has a named owner, a defined structure, and regular maintenance. The team trusts it. |
| 5 | Strategic memory is a governed asset. Decisions, lessons, and rationale are systematically captured, retrievable by anyone, and actively used to shape future work. It compounds instead of decaying. |
Your score for Dimension 5, out of 5.
Add up your five dimension scores. Maximum: 25.
| SMEI Score | Diagnosis |
|---|---|
| 20–25 | Strategic Memory Fortified. Your team retains and applies institutional knowledge. Protect this. |
| 15–19 | Memory Under Pressure. You have some systems in place, but they're fragile. One key departure or tool migration could erase years of judgment. |
| 10–14 | Memory at Risk. Your team is losing strategic knowledge faster than it captures it. Decisions are being remade from scratch. |
| 5–9 | Strategic Amnesia. Most of your team's strategic knowledge is walking out the door every evening. You're paying for insights you've already bought. |
The data environment is getting worse, not better. Deloitte's 2026 State of AI report found that 74% of organizations plan to deploy agentic AI within two years — but only 21% have a mature governance model for autonomous agents. McKinsey research shows 88% of organizations now use AI in at least one business function, yet only 6% qualify as high performers with measurable EBIT impact from AI.
The gap isn't technology access. It's the ability to synthesize, remember, and act.
Meanwhile, the MIT Technology Review and Microsoft's 2026 Agent Confidence Index surveyed 300 tech leaders across 101 tasks and found an average confidence score of just 64 out of 100 — with the lowest scores concentrated in the judgment-intensive work that defines strategic decision-making.
More data. More tools. More agents. Less memory. Less judgment. Less follow-through.
That's the equation your SMEI score helps you solve.
If you scored 15–25: Your foundation exists. The next move is formalizing what's already working — appoint a memory owner, schedule quarterly decision retrospectives, and cut intelligence inputs that don't feed a named decision.
If you scored 10–14: You have scattered practices but no system. Start with Dimension 5 (Governance of Memory). Name an owner. Build one canonical decision log. Then tackle Dimension 3 (Signal-to-Noise) — cut half your intelligence feeds that don't drive action.
If you scored 5–9: Your strategic knowledge is hemorrhaging. Don't try to fix all five dimensions at once. Start with a single, high-stakes decision. Document its full reasoning, tradeoffs, and alternatives. Schedule a 90-day retrospective. Use that template for the next decision. Build momentum from there.
The teams that win over the next five years won't be the ones with the most data or the most AI. They'll be the ones that remember what they've already learned — and act on it.
The MIT Technology Review Insights and Microsoft 2026 Agent Confidence Index (June 2026), covering 300 tech leaders and 101 tasks with an average confidence score of 64 out of 100, is cited via Autostrat competitive intelligence monitoring.
Book a 30-minute demo. Bring a live question and watch the answer get built.