Dear colleague,
In an earlier issue, I wrote about why advisory-grade knowledge should stay inside the organization. We built the Stratenity Library to give teams access to methods and assets they can put to work. The next question is what happens when those teams can apply that knowledge alongside their own evidence, decisions, and operating history.
Imagine a leadership team reviewing a drop in customer renewals. The monthly report shows where renewals fell. A general AI chat can summarize it, suggest possible causes, and draft questions for the team. But the decision turns on facts that may sit elsewhere. Did renewals decline before or after the pricing change? Were the customers who left the same ones who experienced onboarding delays? What did the company try last quarter, and what happened?
| A better prompt can help the team ask a better question. The organization still needs a reliable way to supply the context, examine the evidence, and retain what it learns. |
What the organization remembers
Organizational memory is the ability to retain and retrieve information from past decisions for use in current ones. James Walsh and Gerardo Ungson’s research on organizational memory describes how that information can reside across people, routines, and structures. When the rationale for a decision exists only in the recollection of two people or in a slide nobody can find, its value depends on who is still in the room. [1]
In Stratenity’s approach, the operating graph gives that memory structure. It connects a decision to the evidence behind it, the assumptions it depended on, the person responsible, and the outcome the team agreed to measure. In the renewal example, it could connect a pricing decision to the customer segments it affected and to the results of a later onboarding pilot. A new finding can then be examined in the context of the original decision.
Research from Microsoft on GraphRAG illustrates a narrower technical point. In the researchers’ evaluation, a graph-based approach helped answer questions that required synthesizing themes across a collection of documents. For an enterprise, there is a further design choice about which relationships to preserve. A leadership team needs to know who approved a decision and when it should be revisited, alongside what the documents say. [2]
Introducing OneMind Strata
OneMind Strata (OMS) is the intelligence layer of Stratenity. It applies five specialized models to different kinds of enterprise questions. Each model has its own method and review standards, and Stratenity offers deployment inside an organization’s own environment. [3] [4]
- Research investigates a question and traces material claims back to their sources.
- Analytics reconciles stakeholder interviews, AI-engine outputs, and benchmark evidence in one executive readout.
- Benchmarking compares performance against named peers using comparable measures.
- Brand Perception examines how audiences view the organization and where their perspectives diverge.
- AI Visibility measures how AI engines describe the organization, which sources they cite, and where their answers need correction. [4]
| The distinction matters in a renewal review. Research can test the evidence for competing explanations. Analytics can surface where customer accounts diverge from other evidence. Benchmarking can help determine whether the pattern is unusual among comparable peers. The team can then make a decision with a record of what supports it and where uncertainty remains. |
What this means for strategy, operations, and governance
Strategy decides where to commit resources. Should the business revise pricing, invest in onboarding, or focus its retention effort on one segment? OMS can help leadership test the assumptions behind each option using company evidence and external benchmarks. The operating graph can preserve the reason for the choice and the conditions under which it should be reconsidered. That history matters when the investment comes up for review six months later.
Operations makes the decision work in practice. If leadership funds an onboarding pilot, the operating team needs a defined cohort, a workflow owner, and a way to measure results. It can track time to first value, review customer feedback, and compare renewal outcomes against the baseline where the measurement period allows. When the results come in, they can be linked to the original decision. The next team can see what was tried and where it worked.
Operating governance determines how decisions are made, reviewed, and escalated. Who can approve a wider rollout? Which evidence can AI use? When do conflicting findings require review? In Stratenity’s approach, those rules belong in the workflow. An OMS readout can support a review, and the approval record can capture who made the call, the evidence they considered, and any exceptions they authorized. NIST’s generative AI risk profile reinforces the importance of documenting provenance, data quality, assumptions, and limitations. [5] [6]
Where OMS Snapshots and Reports fit
An OMS Strategic Snapshot is a sourced, dated, two-page response to a specific question. An Industry or Customer Snapshot can help leadership scope a decision. An Operating Model Snapshot can help identify which parts of the business to examine. An AI Governance & Risk Snapshot can frame questions for the people responsible for evaluating exposure. [7]
An OMS Report gives teams room for deeper analysis. For the renewal decision, an integrated analytics report could bring together interviews and benchmarks for an investment review. A later readout could examine the results of the pilot and the evidence for expanding it. Traditional reporting gives people a shared record at a point in time. Connecting that record to operating governance gives it a life beyond the meeting. The team can see which decision the report informed, what action followed, and which evidence should trigger another review. [3] [8] [9]
Preparing enterprise data for the next era of AI
As AI takes on recurring work, small data problems can shape repeated decisions. A renewal rate calculated differently by Finance and Customer Success can send two teams in different directions. An outdated policy can keep resurfacing if its replacement is unclear. A customer interview may be valuable evidence while still requiring tighter access than an aggregate report.
Data moderation establishes what AI may use, who can see it, which sources carry authority, and when evidence needs human review. Data optimization makes approved information easier to retrieve and compare. It can mean reconciling metric definitions, marking effective dates, removing duplicates, and linking records to the customers, workflows, or decisions they describe. [6]
Stratenity can help put this into practice around a recurring decision. Together with the people responsible for it, we can identify the source records, resolve conflicting definitions, establish review and access rules, and apply the relevant OMS models. The same work can improve what outside AI systems see. The AI Visibility Model can establish a baseline for how those systems describe the company and measure again after updates to public information. [5] [4]
OneMind Strata Think of a decision your leadership team makes every quarter. When it comes up again, can someone who missed the last meeting see what the team knew, why it acted, and whether the action worked? That is the enterprise capability we are building with OneMind Strata and Stratenity. |
Two sample reports
Each one-pager summarizes a full OneMind Strata report and shows what the full report contains. Request the full report, or have one produced for your organization.
Research Model Where should a hospital group invest in AI over the next three years? Build the data foundation first, fund operational AI on evidence, and defer agentic AI until it earns its place. Read the one-pager → | AI Visibility Model What AI engines say about Warby Parker, and what they leave out Three AI engines name the brand in nearly every answer but state its NYSE listing in only 5% of them. Read the one-pager → |
Sincerely,
Dodi
Dr. Dodi Mossafer, DBA
Founder
Stratenity - member of NVIDIA Inception
advisory@velorstrategy.com · stratenity.com · LinkedIn → · Book time with me →
- Walsh, J. P., & Ungson, G. R. (1991). Organizational memory. Academy of Management Review, 16(1), 57–91. journals.aom.org.
- Edge, D., et al. (2024). From local to global: A Graph RAG approach to query-focused summarization. Microsoft Research. Publication page.
- OneMind Strata. onemindstrata.com.
- OneMind Strata. Our models. onemindstrata.com/Our-Models.
- Stratenity. About us. stratenity.com/about-us.
- National Institute of Standards and Technology (2024). Artificial Intelligence Risk Management Framework: Generative Artificial Intelligence Profile (NIST AI 600-1). nvlpubs.nist.gov.
- OneMind Strata. Strategic Intelligence Snapshots. onemindstrata.com.
- OneMind Strata. Order reports. onemindstrata.com.
- Stratenity Resources. Decision log template. stratenity.com/resources.