Why it matters
Practitioners can feel the work land. Nobody has been able to show it.
Watch why we built this, and what it means to move and measure the emotions that drive performance.
Why this matters: the case for measuring what moves when teams talk about emotion.
What it is
One survey. Two streams of insight.
Our community uses the ECD cards and tools every day, in workshops, team sessions, coaching and client work. They kept telling us the same thing. They could feel the work landing in the room, and they had nothing to show the people deciding what happens next. At the same time, researchers needed fresh field data without burdening teams with extra surveys. The Emotional Evidence Platform answers both. It is built for anyone using the cards, and it stands on its own, separate from any R&E delivery programme. One QR code and one short survey give each practitioner an Impact Report for their own room, and every result also feeds an anonymous, aggregated dataset for research at Wharton.
It does not measure how people feel on the day. It measures what moves because of the emotional culture conversations practitioners are creating. The conversations are the work. The data is the evidence of what they shift.
If you love data
Eighteen questions on a fixed improvement scale, reported as the share of people improving. Small groups suppressed, every method rule written down, and a monthly report that shows what dipped as well as what rose.
If you build things
A live platform in invited early-user testing: Vue 3, SurveyJS and Highcharts on Supabase and Vercel, with a schema keyed by measure. We are interested in how it could grow, and in who would like to help.
If you tell stories with data
Every number is the share of people who said something got better, never a score or a verdict. The monthly story, slides and report are written for people first, and free to use in your own talks.
Why it is different
Every practitioner can be their own emotion data scientist and storyteller.
Until now, showing that emotional culture work moved something meant stitching together a survey tool, a spreadsheet, some statistics know-how, a charting tool and a slide deck, and having the skills to use them all. Most facilitators are brilliant with people and have never had the time, or the wish, to become analysts.
The platform takes all of that away. One QR code, one short survey, and the analysis, the charts and the story structure are done for you. You keep doing the human work. The evidence builds itself around it.
One place, not five tools
Set up the project, collect the answers, read the Impact Report and share it, all in the same platform.
No analyst skills needed
The improvement scale, the top-3-box method and the small-group rules are built in, so you do not have to learn them to use them well.
Backed by a Wharton professor
The survey and method were developed alongside academic research with Professor Michael Parke at the Wharton School, so the evidence you put in front of a client stands on proper research.
You bring the conversations. The platform brings the data and the story. Behind every data point is a human being.
How it works
From a QR code to evidence.
A practitioner sets up a project
After a session using the ECD cards or any ECD tool, whether a workshop, a team conversation or a coaching session. The platform gives them a link and a QR code.
People answer, anonymously
Participants scan, answer 18 short questions in about two minutes, and never log in. No individual is identified.
The platform builds an Impact Report
The share of people reporting improvement on each measure, for that room, ready to put in front of a client or sponsor.
The evidence builds across every room
Every project adds to an anonymous, aggregated dataset. That dataset is what the monthly reports below describe.
What it measures
Fifteen improvement measures. Three satisfaction headlines.
Each person rates how much has changed since the work began. The report shows the share scoring 3 or higher, which means any improvement.
Individual
- Understanding my own emotions
- Managing my difficult emotions
- Sharing my difficult emotions
- Understanding other people's emotions
- Talking with my team about their emotions
Team
- Psychological safety
- Raising tough issues
- Trust
- Relationship quality
- Connection
- Empathy
- Loneliness
Outcome
- Enthusiasm for work
- Speaking up about problems
- Burnout
Three satisfaction headlines sit alongside: how effective, how satisfied, and interest in doing more.
How we keep it honest
Claims we make, and claims we do not.
People tell us what shifted.These are self-reported scores from people who chose to respond. We never say the work caused a change.
Not a benchmark.Every respondent starts from a different place, and a single workshop is not comparable to a long programme.
High is good, even on the hard ones.A high share on burnout or loneliness means more people got better.
Submitted responses only.We count responses that were saved when someone pressed submit. Every question is required, so each saved response is complete. We cannot yet see anyone who started and did not submit. Saving partial answers is a planned feature, so we can report started against finished.
Aggregates only.No group smaller than three is ever shown, and no individual is ever identified.
See it working
Watch it work, start to finish.
About nine minutes. Setting up a project, generating the feedback link, and reading the results.
Before we share the data
“Remember, data is just the beginning. Data must always tell a story, and that story must be about people.”Brian Solis
We cannot move the workplace with numbers alone. The numbers below are here to support a story about people, and to help influence what happens next.
The data, month by month
What the data is saying.
Each month we publish the story, what moved and what held, and three slides. Choose Overall for everything to date, or a month for what changed. If a number looks bad, it goes in.
October 2026
Platform evidenceThe numbers
325 responses to date (18 Jan to 24 Sep 2026), 321 with at least one answer, 250 fully complete F18s, 5,562 individual question answers. Growth on the July baseline in brackets: responses +56 (+21%), with at least one answer +56 (+21%), fully complete +56 (+29%), individual answers +1,008 (+22%). All 56 new responses arrived between 20 Jul and 24 Sep and every one is a complete F18. That is one in late July, 12 in August and 43 in September.
What we noticed
The gradient from July holds. Top-3-box (any improvement) is highest on understanding other people's emotions (92%) and your own (89%), sits in the low 80s across the team measures, and softens on sharing your own difficult emotions (77%), loneliness (66%) and burnout (57%, the lowest). Among the 56 new responses, psychological safety, raising tough issues and connection each sit about four to seven points below the earlier group (79% against 82%, 86% and 83%), and loneliness sits higher (77% against 64%). With n=56 we are not reading anything into it, and it goes in because a number that looks bad goes in. Satisfaction headlines (1 to 5 scale averages): effective 3.64, satisfied 4.00, want more 3.67.
The story beat
Three months on, the number has moved and the shape has not. People still tell us this work changes how they see emotion fastest, and that the heavier outcomes take longer. I feel the honest headline is that 56 more people spoke up, every one of them finished the survey, and the burnout and loneliness numbers are the ones to keep watching.
Things in the numbers you might not expect. Every figure is counted from the live data. They show what goes together, not what causes what.
When sharing gets easier, trust usually does too.
Nearly everyone who found it easier to share difficult emotions also felt trust improved (91%). Among people who did not find sharing easier, fewer than half did (45%).
This shows the two go together, not which causes which. The counts are 219 of 240 against 32 of 71.
Understanding others and sharing yourself go together.
People who understand other people's emotions better usually find it easier to share their own too (83%). Among the few who did not improve at understanding others, only 12% found sharing easier.
The second group is small, 3 of 25, so treat it as a hint, not a rule. The first group is 237 of 286.
Burnout does not move with everything else.
Four in five people whose burnout did not improve still understand their own emotions better (81%). Almost everyone was satisfied with the work either way, so satisfaction alone cannot tell you whether burnout is shifting.
Burnout is also our lowest measure, at 57%. The counts are 109 of 134, and satisfaction of 177 of 178 against 132 of 134.
Mostly, it held. Across 15 improvement measures, the overall picture has barely moved since July, and that is what I would expect when 56 new voices join 269 existing ones. The more useful question is whether the newer voices tell a different story to the earlier ones. For most measures they do not.
I feel the most important finding is a quiet one. The shape is steady. Understanding emotion improves fastest, and loneliness and burnout are the slowest to shift.
Moved up
Loneliness. 77% of the newest responses report improvement, against 64% in the earlier group. It is the clearest shift in the dataset, and still borderline at this sample size. Loneliness is also the second-lowest measure overall, so there is room to move.
Leaning, not proven
Up: managing my own emotions, confidence talking with the team about their emotions, understanding my own emotions. Down: raising tough issues (79% in the newest responses against 86% earlier). None of these clears the bar for certainty with only 56 new responses.
Held
Trust, empathy, relationship quality, enthusiasm, understanding other people's emotions, sharing difficult emotions, burnout, and all three satisfaction headlines. Differences are within what chance alone would produce.
Measure by measure
Share reporting any improvement (scoring 3, 4 or 5). Read left to right for the trajectory. Newest vs earlier compares the 56 responses since July with the 255 or so before. Moved means beyond what chance would usually produce, Leaning means it might be real, Held means no meaningful gap.
| Measure | Jan to Mar 2026 | Apr to Jun 2026 | Jul to Sep 2026 | Newest vs earlier | Whole dataset | Verdict |
|---|---|---|---|---|---|---|
| Individual measures | ||||||
| Understanding my own emotions at workn=311 | 88%n=89 | 88%n=165 | 93%n=57 | +5 pts | 88% to 89% | Leaning up |
| Confident to manage my difficult emotionsn=311 | 72%n=89 | 86%n=165 | 88%n=57 | +6 pts | 81% to 82% | Leaning up |
| Confident to share my difficult emotionsn=311 | 65%n=89 | 82%n=165 | 81%n=57 | +4 pts | 76% to 77% | Held |
| Understanding team members' emotionsn=311 | 89%n=89 | 94%n=165 | 91%n=57 | -1 pts | 92% to 92% | Held |
| Confident talking with team about their emotionsn=311 | 80%n=89 | 87%n=165 | 91%n=57 | +6 pts | 85% to 86% | Leaning up |
| Team measures | ||||||
| Expressing true feelings is welcomedn=311 | 75%n=89 | 86%n=165 | 79%n=57 | -3 pts | 82% to 81% | Held |
| Members can bring up problems and tough issuesn=250 | 75%n=28 | 87%n=165 | 79%n=57 | -7 pts | 86% to 84% | Leaning down |
| We trust each othern=311 | 75%n=89 | 83%n=165 | 82%n=57 | +2 pts | 80% to 81% | Held |
| Close or effective relationships with teamn=311 | 81%n=89 | 79%n=165 | 84%n=57 | +4 pts | 80% to 81% | Held |
| Strong sense of connection with teamn=311 | 79%n=89 | 86%n=165 | 79%n=57 | -4 pts | 83% to 82% | Held |
| Team treats each other with care and respectn=311 | 76%n=89 | 84%n=165 | 84%n=57 | +2 pts | 82% to 82% | Held |
| Feelings of loneliness at workn=311 | 58%n=89 | 67%n=165 | 77%n=57 | +13 pts | 64% to 66% | Moved up |
| Outcome measures | ||||||
| Hopeful or enthusiastic about workn=321 | 87%n=99 | 78%n=165 | 82%n=57 | +1 pts | 82% to 82% | Held |
| Speak up with recommendations about problemsn=313 | 86%n=91 | 79%n=165 | 77%n=57 | -5 pts | 82% to 81% | Held |
| Feel burned out from workn=321 | 66%n=99 | 51%n=165 | 61%n=57 | +4 pts | 57% to 57% | Held |
What the trajectory tells us
The earliest responses (January to March) were the lowest on most of the individual and team measures. Managing my own difficult emotions sat at 72% and sharing them at 65%, rising to 86% and 82% in the April to June group, then holding through the summer. I would be careful about calling that progress in the work itself. Those early responses came from a smaller, earlier set of rooms, and different engagement lengths are not comparable.
Burnout is the one measure that wobbles without a clear direction: 66%, then 51%, then 61%. I feel that is worth watching each month rather than explaining away. It is the heaviest outcome and the one practitioners tell us takes longest to shift.
Satisfaction stays very high. 99% of respondents score the work 3 or higher on how satisfied they are, but with a ceiling that close there is very little room to see movement.
Three slides for this month, free to use in any talk or deck.
The whole report for October 2026: every measure, quarter by quarter, with the full method. Open it as its own page, or read it here. Also: the three slides full screen and the story on its own.
July 2026
Platform evidenceThe numbers
269 responses to date (18 Jan to 30 Jun 2026), 265 with at least one answer, 194 fully complete F18s, 4,554 individual question answers. (First entry, so no growth figure.)
What we noticed
Completion is strong: roughly seven in ten people who start the F18 finish all 18 questions. The Wharton studies (172 participants) remain a separate, Qualtrics-collected evidence base and are never mixed into platform counts.
The story beat
For six months the platform collected quietly while we built. This month we counted properly for the first time, and the number is real: 269 people have told us how emotional culture work landed for them. The next six months are about invited early users growing that number and teaching us what it proves.
Things the numbers show that most people would not expect. Counted by us from the platform data as it stood at the July baseline.
People stay with the survey to the end.
194 of the 265 people who answered at least one question finished all 18 (73%). Looking closer, 61 of the 71 who did not were not walking away. They took an earlier 17-question version, and the other 10 stopped in the first few questions.
We cannot point to a fair outside benchmark, so we will not call 73% high. The point is that most of the gap is the survey changing, not people losing interest.
Overall
Every response to dateAcross every response from 18 Jan 2026 to 24 Sep 2026, the highest share reporting improvement is Understanding other people's emotions at 92%, and the lowest is Burnout at 57%. Each bar is the share of people who scored 3, 4 or 5 on the improvement scale. A share, not a score. High is good, even on burnout and loneliness.
Every measure, whole dataset
Bars show everyone to date. The small figures on the right show the first to the latest quarter, so you can see the direction.
Individual
Team
Outcome
Want the detail? Open Full data report for the quarter-by-quarter table and what moved against earlier responses.
The whole report to date. Open it as its own page, or read it here.