The Collaboration lens turns review text into a network view of the firm. Where Reviews > Collab Points uses explicit recognition allocations, this page is built from what reviewers say about how people work together.
What this page answers
Which departments collaborate most? Where are the silos? What types of collaboration are happening — and who’s doing the work?
What’s on the page
Three headline KPIs
Total Evidence: count of distinct collaboration evidence entries Compass found across all extracted reviews
Avg per Person: average count of evidence entries per person with evidence
Collaboration Types: count of distinct collaboration types Compass identified (e.g., cross-practice work, mentoring, knowledge sharing)
Strongest connections
A ranked list of the top 15 department-to-department pairs by the number of completed cross-department 360 reviews between them (a contributor in one department reviewing someone in another). The bigger the number, the more collaboration is happening between those departments. Requires 360 feedback data.
Potential silos
A list of departments sorted by how connected they are, expressed as a percentage of all other departments they have 360 review connections with:
Green (≥80%): well connected
Blue (50–80%): reasonably connected
Amber (25–50%): lightly connected
Red (<25%): isolated
Red entries are the silos worth investigating.
Strongest collaboration network
An interactive network graph where nodes are departments and links are cross-department 360 review connections between them. Only links above a threshold (max of the median weight, or 2) are drawn — so the graph is the backbone of the network, not every connection.
Interactive: drag nodes, hover for detail.
Collaboration types
A donut chart breaking down the types of collaboration Compass found across the cohort. Common types include cross-practice work, mentoring, client teamwork, internal knowledge sharing — the exact set depends on what’s in your reviews.
Collaboration by department
A bar chart showing which departments have the most collaboration evidence. A high number here means this department’s people have the most collaboration evidence in their reviews.
Collaboration evidence
A paginated table (10 per page) of the actual evidence Compass extracted:
Name: the employee the evidence relates to (clickable)
Department
Type: the collaboration type (filterable)
Evidence: the free-text evidence snippet from the review
Where the data comes from
The KPIs, types, department breakdown, and evidence table are built from the collaboration evidence on each review extraction. Compass reads the review text and extracts a list of collaboration-related signals — what the person collaborated on, with whom, and what type.
The Strongest connections, Potential silos, and network graph are built from 360 feedback contributor data — they map actual cross-department review flows between contributors. If the cycle has no 360 data, those three widgets are replaced by a notice; the evidence-based widgets still work.
How to read it
Read silos as “where we should invest in cross-pollination”. Isolated departments often produce isolated thinking.
The network graph is best as a conversation starter. Show it to leadership — the backbone view makes “we’re not really one firm” arguments much easier.
Collaboration types tell you culture. A firm where “mentoring” dominates feels different from one where “client teamwork” dominates. Neither is wrong; both inform recruiting and development.
Evidence table is where you justify claims. When someone challenges a network finding, the evidence backs it up.
Tips
Network graphs require enough data to be meaningful. Small cohorts or sparse review content produce sparse networks. The page will hide some charts if there’s not enough connection data.
Pair with Collab Points. Collaboration evidence (this page) is what reviewers say; Collab Points is what reviewers allocate. Differences between the two are interesting — sometimes someone gets lots of words but few points, or vice versa.
Filter the evidence table by type. “Show me all the mentoring evidence” is often a better starting point for a talent conversation than the aggregate view.