The SCOPE Synthesis Framework

After fifteen months of refining how this team processes intelligence, one synthesis framework consistently produces the clearest output. It operates in five stages, each building on the previous one.

Stage one: Sort. Organize all findings by topic, not by source. This breaks the common trap of source-by-source summarization. When you organize by topic, findings from different sources naturally sit next to each other, making patterns visible. Stage two: Connect. Draw explicit links between findings in different topic clusters. A hiring pattern in one cluster and a patent filing in another may be causally related. These cross-cluster connections are where the highest-value synthesis insights emerge.

Stage three: Assess. For each connection and pattern you identified, assign a confidence level using the verification standards from DR-201b. Not every synthesis finding deserves the same weight. Stage four: Prioritize. Rank your synthesis findings by impact — which ones, if true, would most affect the decision being made? Research consumers are overwhelmed by volume. Prioritization ensures they see the most important findings first. Stage five: Articulate. Write each synthesis finding as a single declarative statement supported by evidence from multiple sources. "Company X is executing a healthcare market entry, assessed with high confidence based on convergent signals from patent filings (Q3), hiring patterns (Q4), and executive commentary (January earnings call)." That is a synthesis finding. It could not have come from any single source.

1. Sort by Topic Group findings by subject matter, breaking them free from their source containers. All findings about a company's AI strategy go together regardless of whether they came from an earnings call, a job posting, or a patent.
2. Connect Across Clusters Draw explicit links between topic groups. Does the talent acquisition cluster relate to the technology investment cluster? Do the partnership signals connect to the market expansion signals? Cross-cluster connections are the synthesis engine.
3. Assess Confidence Apply the triangulation and confidence standards from verification methodology. Each synthesis finding gets a confidence rating: high, medium, or low. This prevents synthesis from inflating certainty beyond what the evidence supports.
4. Prioritize by Impact Rank findings by their decision-making relevance. A high-confidence finding about a minor competitor matters less than a medium-confidence finding about a direct threat. Impact ranking ensures the reader encounters the most consequential findings first.
5. Articulate as Declarative Findings Write each synthesis output as a single sentence stating the conclusion, followed by the evidence chain. One finding, multiple sources, explicit confidence. This format forces clarity and prevents the fuzzy hedging that makes intelligence unusable.