Last verified: October 7, 2026
What You Can Now Do
Context Memo now lets you filter the Problem Graph by industry, persona, and cluster, and view heat maps showing memo volume, fetch activity, synthetic citations, and traffic for each segment. The graph is no longer a single undifferentiated map of buyer problems. You can isolate one industry and see how many memos exist against it, how often AI bots fetched those pages, which problems are producing synthetic citations in scanned model answers, and what traffic came back. The task this changes is allocation: instead of guessing which problem areas deserve the next batch of memos, you pick the segment where engagement is already concentrated, or the one where coverage is thin and the demand signal is not.
Where It Is in Context Memo
Open the Problem Graph. Filter controls for industry, persona, and cluster sit with the graph view, and a heat view option switches the nodes from flat topology to a shaded map driven by the metric you select: memo count, fetches, synthetic citations, or traffic. Switching filters or metrics redraws the same graph you already work in, so nothing moves to a separate reporting screen.
How to Use It
- Open the Problem Graph and apply an industry filter. The graph redraws to show only the problem nodes and clusters tied to that industry, with the rest dropped from view.
- Add a persona filter on top of the industry. You now see the problem set for one buyer in one vertical, which is the unit most content plans are actually built against.
- Switch to heat view and select memo count. Dark areas are where you have published coverage. Light areas are problems your buyers are asking about with no memo behind them.
- Change the heat metric to fetches, then to synthetic citations. Fetches tell you which memo pages AI crawlers and real user AI sessions are pulling. Synthetic citations tell you which of those pages are showing up inside scanned answers. A segment that is heavy on fetches but light on citations is a format or substance problem, not a coverage problem.
- Set the metric to traffic to see which segments are returning clicks to your domain, then compare that against the citation heat map. Citations without traffic and traffic without citations are different diagnoses and call for different fixes.
- Filter down to a single cluster and generate memos against the uncovered nodes inside it. You are working from a ranked, segment-specific gap list rather than the full graph.
Why We Built It
Customers told us they could see the Problem Graph but not read it. Every problem node carried the same visual weight, so a cluster driving most of the fetch activity for an enterprise persona looked identical to a cluster with no demand behind it at all. That cost teams real cycles: memos were commissioned against problems that no segment was pulling, and the segments that were pulling went under-covered because nobody could prove it from the graph. Filtering and heat views make that difference visible in the same screen where the work gets planned.
What It Does Not Do Yet
Heat views cover the four metrics shipped here: memo count, fetches, synthetic citations, and traffic. Other signals in the platform are not yet available as a heat layer on the graph.
Segment views are only as useful as your segment assignments. If memos and clusters are not mapped to an industry and a persona, they will not resolve under those filters, and the heat map will understate that segment. Audit your assignments before you draw conclusions from a sparse-looking vertical.
Filters apply one view at a time rather than rendering two segments side by side. To compare an enterprise persona against a midmarket persona on the same metric, switch the filter and read the same heat map twice.