Sample workspaceWebsite ↗

Attribution

Which touches contributed, and how much

attr-v2
Not causal proof. Attribution assigns credit across observed touches. It is not causal proof that any touch caused the opportunity.

Acme reliability platform · opened Sep 28

18 eligible touches · 2 excluded · confidence high

TouchLast touchGraph assisted

Grace Wu liked a post

Jul 9 · 80.8d before open · ne_002e2e8b23bf8b

0%
0.1%

Sam Patel commented on a post

Jul 13 · 76.8d before open · ne_18fa8a43c074db

0%
0.1%

Sam Patel liked a post

Jul 15 · 74.8d before open · ne_1eab48516ef580

0%
0.1%

Ines Moreau liked a post

Aug 10 · 48.8d before open · ne_1500ba15185f7c

0%
0.3%

Grace Wu liked a post

Sep 13 · 14.8d before open · ne_17e1f6d24e3473

0%
2.5%

Tomas Lindqvist liked a post

Sep 13 · 14.8d before open · ne_1f84617824e335

0%
2.8%

Tomas Lindqvist commented on a post

Sep 19 · 8.8d before open · ne_1a2c715997173f

0%
9.3%

Daniel Reyes replied to Tomas Lindqvist

Sep 20 · 7.8d before open · ne_1a58b1f95272e1

0%
8.7%

Sam Patel liked a post

Sep 22 · 5.8d before open · ne_15f4813de53ff5

0%
1.7%

Noor Haddad liked a post

Sep 22 · 5.8d before open · ne_07e6cc4ebef813

0%
1.7%

Lena Fischer introduced Tomas Lindqvist

Sep 24 · 4.1d before open · ne_02baa8a6cda902

0%
23.5%

Grace Wu commented on a post

Sep 24 · 3.8d before open · ne_1cebc70e1508e3

0%
10.6%

Sam Patel commented on a post

Sep 25 · 2.8d before open · ne_1fc16c4e1d4d64

0%
4.9%

Priya Shah replied to Grace Wu

Sep 25 · 2.8d before open · ne_1e77f1dc6bbec7

0%
9.9%

Meeting: Acme reliability roadmap intro

Sep 26 · 2.0d before open · ne_13ca1d203daf8d

0%
10.1%

Grace Wu liked a post

Sep 27 · 0.8d before open · ne_0a85e55dd6c75a

0%
4.9%

Noor Haddad commented on a post

Sep 27 · 0.8d before open · ne_13725c9a6764f5

0%
5.4%

Ines Moreau liked a post

Sep 27 · 0.8d before open · ne_0388cb1aabf0ea

100%
3.4%
Sum of weights1.0000001.000000

How Graph assisted explains each weight

  • Lena Fischer introduced Tomas Lindqvist23.5%

    type 1.6 × recency 0.82 × relationship 0.63 × identity 1.00 × source 0.85

  • Grace Wu commented on a post10.6%

    type 1 × recency 0.83 × relationship 0.57 × identity 0.90 × source 0.75

  • Meeting: Acme reliability roadmap intro10.1%

    type 1.4 × recency 0.91 × relationship 0.25 × identity 1.00 × source 0.95

  • Priya Shah replied to Grace Wu9.9%

    type 0.8 × recency 0.88 × relationship 0.57 × identity 1.00 × source 0.75

  • Tomas Lindqvist commented on a post9.3%

    type 1 × recency 0.66 × relationship 0.63 × identity 0.90 × source 0.75

  • Daniel Reyes replied to Tomas Lindqvist8.7%

    type 0.8 × recency 0.69 × relationship 0.63 × identity 1.00 × source 0.75

  • Noor Haddad commented on a post5.4%

    type 1 × recency 0.96 × relationship 0.25 × identity 0.90 × source 0.75

  • Sam Patel commented on a post4.9%

    type 1 × recency 0.87 × relationship 0.25 × identity 0.90 × source 0.75

  • Grace Wu liked a post4.9%

    type 0.4 × recency 0.96 × relationship 0.57 × identity 0.90 × source 0.75

  • Ines Moreau liked a post3.4%

    type 0.4 × recency 0.96 × relationship 0.39 × identity 0.90 × source 0.75

  • Tomas Lindqvist liked a post2.8%

    type 0.4 × recency 0.49 × relationship 0.63 × identity 0.90 × source 0.75

  • Grace Wu liked a post2.5%

    type 0.4 × recency 0.49 × relationship 0.57 × identity 0.90 × source 0.75

  • Noor Haddad liked a post1.7%

    type 0.4 × recency 0.76 × relationship 0.25 × identity 0.90 × source 0.75

  • Sam Patel liked a post1.7%

    type 0.4 × recency 0.76 × relationship 0.25 × identity 0.90 × source 0.75

  • Ines Moreau liked a post0.3%

    type 0.4 × recency 0.10 × relationship 0.39 × identity 0.90 × source 0.75

  • Sam Patel commented on a post0.1%

    type 1 × recency 0.03 × relationship 0.25 × identity 0.90 × source 0.75

  • Grace Wu liked a post0.1%

    type 0.4 × recency 0.02 × relationship 0.57 × identity 0.90 × source 0.75

  • Sam Patel liked a post0.1%

    type 0.4 × recency 0.03 × relationship 0.25 × identity 0.90 × source 0.75

Excluded touches

Shown so nothing is silently dropped

  • Sam Patel liked a post — occurred after the opportunity opened
  • Sam Patel liked a post — occurred after the opportunity opened

Caveats

  • Attribution assigns credit across observed touches. It is not causal proof that any touch caused the opportunity.
  • Only touches captured by connected sources are visible; offline conversations are missing by construction.
  • Graph-assisted factors are hand-set heuristics (see docs/attribution.md), not a learned model.