Your event tells you more than your guests do.
We listen.
Every registration, every check-in and every tap in the app says something about your guests. We turn that into answers you can actually use: What worked, where is there still room to grow, and what can you do even better next time?
Five questions your event data can answer
Every answer ends with a recommendation for your next event.
The idea
Data is the feedback your guests never put into words
After an event, you ask your guests how it went. Some reply, most don't. Yet every one of them has already answered, through what they did.
Our products record these moments anyway, because they're needed to run the event. Event analytics means we put that data side by side, spot patterns and turn them into recommendations. No jargon, and no spreadsheets nobody reads.
Who registered, and who actually showed up?
Which content was read, and which was never opened?
When did your guests respond to a message, and when did they ignore it?
What we measure
From the invitation to the day after
Registration
How many invitees open the link, how many complete registration, and where do the rest drop off? How many of those who booked actually show up?
Event app
Who uses the app, when, for how long and for what? Which content is in demand, and which push notifications get opened?
On site
Check-ins, test drives, workshop attendance and capacity by time of day.
Feedback
Star ratings, surveys and free-text comments, visible live in the app during the event if you like.
The bigger picture
Data becomes valuable once you connect it: for example, whether guests who opened the app early are more likely to show up or more satisfied.
Question 1: When do I reach my guests?
An event day in data
This is what an event day looks like when you make it visible: 107 people, 955 app sessions. Each row is one person, each line a glance at the app. You can see when guests arrive, when the breaks are and when everyone checks for information at the same time.
Each row is one anonymised person, each line one app session. Top: concurrent open sessions.
Training event with test drives, 2026 · 107 people, 955 sessions in one day · dashed: push notifications
Active participants 30 minutes before and after a push notification, totalled by time slot
Training event with test drives, 2026 · Waves 2–20 · 984 participants · 97 push notifications
When a push notification works
The result at this event: A push notification at 7:30 am brought almost three times as many guests into the app as in the half hour before. At midday, the number doubled. By 7:49 pm, guests had already wrapped up their day. The same message the next morning would have done more.
Not every push works the same
Confirmed across 19 waves: A training event with test drives ran in 19 waves, meaning 19 runs of the same programme, each with new guests. Across all waves, we analysed 97 push notifications. The pattern held in every single wave. The message the day before brought almost six times as many guests into the app as in the preceding half hour, and the message at the start of day one brought two and a half times as many. The effect was smallest in the evening.
Active participants 30 minutes before and after a push notification, totalled by time slot
Training event with test drives, 2026 · Waves 2–20 · 984 participants · 97 push notifications
All interactions from waves 2–20, mapped onto a single event day, in 5-minute intervals
Training event with test drives, 2026 · Waves 2–20 · 90,960 interactions · dashed: standard push times
19 waves in one day
19 waves in one day: Stack all 19 waves on top of each other and you see what the platform would have to handle in a single day: up to 278 guests in the app at once and 975 interactions in five minutes. The peak comes at 1:20 pm, when a driving block ends and everyone submits their rating at the same time. We factor moments like these into load testing and message timing.
What's in it for you:
You know when your most important message gets read. Put your call for ratings, the reminder about the evening event and the invitation to your next event into exactly these windows.
Question 2: When does an event start?
Long before the welcome, and it doesn't end with the goodbye
The app lives before and after the event
At a dealer conference with 16 waves and more than 2,600 invitees, we analysed each wave individually. The pattern was the same across all of them: 72 percent of guests who used the app opened it the day before, and on the first event day it was 93 percent. Two days after the event, one in four was still checking in. The key takeaways summary was viewed more than 2,100 times after the event.
Share of active users who opened the app on a given day, averaged across 16 waves
Dealer conference in 16 waves, 2024 · 2,184 active users · Day 1 and 2 = event days
Participants by pre-event login: usage at the event and return afterwards
Training event with test drives, 2026 · Waves 2–20 · 984 participants
Guests who look in early are more active at the event
The day before matters: At the training event with test drives, three out of four guests opened the app before day one. These guests used it much more at the event, with a median of 84 instead of 62 interactions, and came back more often afterwards.
The app stays in use after the event
And afterwards: Within a week, one in four guests came back to the app. After three months, it was 40 percent. Mostly they opened presentations and key takeaways. The app becomes a reference tool.
Share of participants who returned at least once after the event (cumulative)
Training event with test drives, 2026 · Waves 2–20 · 984 participants · Data up to the end of September
What's in it for you:
The days before the event are your best chance to set expectations. The days after are your best chance to make your message stick. Plan both based on data, not gut feeling.
Question 3: What can I learn from wave to wave?
Back-to-back events: every wave makes the next one better
16 waves, 16 reports
When a format runs in many consecutive waves, the analysis pays off twice. At the dealer conference, the project team received an event report after every wave: usage, satisfaction, popular content, languages and markets. That meant they could make adjustments between waves, not just after the final day.
At the training event with test drives, the ratings themselves provided the data: Across 19 waves, guests submitted more than 7,300 vehicle ratings, an average of 7.5 per guest for twelve vehicles. The number of ratings stayed steady from morning to late afternoon, a sign that guests stayed engaged right to the end.
Active app users per wave (bars) and share of all invitees
Dealer conference in 16 waves, 2024 · 2,184 of 2,624 invitees active (83%) · Avg. likelihood to recommend 9.6 out of 10
Vehicle QR code scans per hour on both event days, all waves combined
Training event with test drives, 2026 · Waves 2–20 · 9,171 scans · Vehicles anonymised, sorted by total
The rotation spreads guests out without bottlenecks
The vehicle QR code scans show when each vehicle was in demand. Across both days, all twelve vehicles were in demand at every hour, without any bottlenecks. That confirms the rotation plan, and any bottleneck would have shown up right away.
More engagement from wave to wave
The result: 2,184 of 2,624 invitees used the app actively. That's 83 percent. Guests rated their likelihood to recommend at 9.6 out of 10 on average and gave the app 4.8 out of 5 stars. Their own posts in the event feed rose from an average of 0.2 per active guest in the first eight waves to almost one per active guest in the last three.
Participant posts in the event feed per active user
Dealer conference in 16 waves, 2024 · 929 participant posts in total
Question 4: Where am I losing registrations?
Your registration flow as a measuring tool
For a test drive tour by a premium car manufacturer across six locations, we analysed more than 13,000 registration journeys, from the click on the invitation to the completed test drive.
27 out of 100
invitees who opened the link completed registration. The biggest drop-off happened at a single step: model selection.
8 out of 100
booked guests didn't show up, ranging from 4 to 14 percent depending on the location.
Around 1,200
additional test drives came from a single product change: guests could book several vehicles in one step.
1
location recommendation for the following year, based on catchment areas, purchasing power and registration behaviour.
The deadline predicts no-shows
The deadline as an early warning system: At the training event with test drives, 11 percent of guests who had confirmed by the registration deadline didn't show up. Among those who hadn't responded by the deadline, it was 45 percent, four times as many. Follow up on deadline day and you'll know who's worth the effort.
What's in it for you:
You see which step is costing you guests, how much you can overbook and where your next event will have the biggest impact.
Share of invitees who end up not attending, by status on the day of the registration deadline
Training event with test drives, 2026 · Waves 1–20 · 1,047 invitees · Data as of 30 September 2026
Question 5: What does the app give me beyond a single event?
Every event builds on the last
A single event has its own curve before, during and after. When the same app accompanies your guests across several events, part of the usage carries over after each event and keeps growing with the next one: agenda, presentations, contacts and news all in one place, even between events.
We're already measuring that guests come back: At the training event with test drives, 40 percent of guests returned to the app within three months, without any particular reason to do so.
What's in it for you:
Plan event series, annual conferences and training programmes as one platform instead of many separate apps. That way, each event doesn't start from zero, but with the guests who are already there.
When the same app accompanies your guests across several events, usage keeps growing between events
Schematic illustration, not measured data
Live and after the event
Results while the event is still running
Live in the event app
Ratings and surveys are visible in the event app right away, either as a distribution per question or as a comparison of several products. At test drives, guests can see how others rated the same vehicle right after their own drive.
After the event
Then comes the analysis a live view can't deliver: trends over time, the impact of each message, comparisons across roles, markets and waves, a summary of free-text comments and concrete recommendations for your next event.
How we work:
1. Define goals
What should your event achieve? That tells us what to measure.
2. Capture automatically
Registration, app and check-in deliver the data with no extra work for your team.
3. Monitor live
The dashboard shows usage, registrations and ratings in real time.
4. Analyse and recommend
After the event, you get an event report with charts and clear recommendations.
5. Plan better
At your next event or in the next wave, you put the recommendations into practice and see what they achieve.
Three levels:
Live dashboard
Included in every project with registration or an event app: usage, registrations, opt-ins, content and ratings over time.
Event report
The post-event analysis: what worked, what didn't, and five specific levers for next time. For back-to-back formats, after every wave, with fixed modules: guests and satisfaction, daily usage, opt-ins, most-used content, feed and messages, languages and markets, and key takeaway views.
Strategic analysis
Year-on-year comparison, location and format evaluation, registration and no-show patterns. For event series you develop over several years.
Data protection
We analyse behaviour, not people
Our analyses show aggregated values. Where a chart shows individual journeys, they're anonymised and can't be traced back to anyone.
We use pseudonymised data for our analyses.
We don't report on groups of fewer than five people.
We only use data from your own channels, never social media profiles.
All processing is GDPR-compliant, on servers in Germany.
Since 2018
Data analysis has been part of dgtl.ai from day one. Back with the IBM Think event series, we were already analysing how guests used the app and which content they cared about. Some of those insights still hold true today: During an event, app sessions are short and focused. Before and after, people read.
In 2026, we measured it again across 19 waves and 19,000 app sessions: Before the event, a session includes a median of six interactions, at the event only three. The recommendation is the same as back then: an “everything you need to know” message for the evening before and a recap for the morning after.
Interactions per app session (median). A session ends after 10 minutes of inactivity.
Training event with test drives, 2026 · Waves 2–20 · 18,981 sessions · First measured at an event series in 2018
Why it works
Data that's already there + an analysis you can understand + recommendations you can act on = an event that gets better every time
Your next step
What is your last event telling you?
Send us the exports from your last event, or tell us about your next one. We'll show you which questions your data can answer.
You might also be interested in:
Last updated: November 6, 2026
