You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: src/engage/journeys/faq-best-practices.md
+9-9Lines changed: 9 additions & 9 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -65,37 +65,37 @@ Follow these best practices to test your journeys:
65
65
- For early version journeys, scaffold Send to Destination steps without connecting to your production advertising or messaging destinations.
66
66
- Verify individual users' progress through the Journey in the Profile explorer view.
67
67
68
-
## Frequently asked questions
68
+
## FAQs
69
69
70
-
### How often do Journeys run?
70
+
####How often do Journeys run?
71
71
72
72
Journeys run in real-time, like real-time Audiences in Engage. This means that users will progress through Journeys as Segment receives new events.
73
73
74
-
### Can a user re-enter a Journey?
74
+
####Can a user re-enter a Journey?
75
75
76
76
Yes. Users must first exit a Journey, however, before entering it again. To learn more about Journey re-entry, read the [Journey re-entry section](/docs/engage/journeys/build-journey/#journey-re-entry) of the [Build a Journey](/docs/engage/journeys/build-journey/) page.
77
77
78
-
### What destinations does Journeys support?
78
+
####What destinations does Journeys support?
79
79
80
80
Journeys supports all Engage destinations, including Destination Functions. Read more in Send data to destinations.
81
81
82
-
### What are the reporting capabilities of Journeys?
82
+
####What are the reporting capabilities of Journeys?
83
83
84
84
When building a Journey, if you check **Use historical data**, you can see the estimated number of users in the initial cohort.
85
85
86
86
Once published, Journeys displays the number of users are in each step of the Journey at any given time.
87
87
88
-
### How are users sent to downstream destinations?
88
+
####How are users sent to downstream destinations?
89
89
90
90
The data type you send to a destination depends on whether the destination is an Event Destination or a List Destination.
91
91
92
-
### Which roles can access Journeys?
92
+
####Which roles can access Journeys?
93
93
For Engage customers, users with either the Engage User or Engage Admin roles can create, edit, and delete journeys. Users with the Engage Read-only role are restricted to view-only access.
94
94
95
-
### Why am I seeing duplicate entry or exit events?
95
+
####Why am I seeing duplicate entry or exit events?
96
96
97
97
Journeys triggers audience or trait-related events for each email `external_id` on a profile. If a profile has two email addresses, you'll see two Audience Entered and two Audience Exited events for each Journey step. Journeys sends both email addresses to downstream destinations.
98
98
99
-
### How quickly do user profiles move through Journeys?
99
+
####How quickly do user profiles move through Journeys?
100
100
101
101
It may take up to five minutes for a user profile to enter each step of a Journey, including the entry condition. For Journey steps that reference a batch audience or SQL trait, Journeys processes user profiles at the same rate as the audience or trait computation. Visit the Engage docs to [learn more about compute times](/docs/engage/audiences/#understanding-compute-times).
Copy file name to clipboardExpand all lines: src/engage/user-subscriptions/subscription-groups.md
+13-19Lines changed: 13 additions & 19 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -108,37 +108,34 @@ To set susbcription groups with the Identify call, you'll need to include a key-
108
108
}
109
109
```
110
110
111
-
## Frequently asked questions
111
+
## FAQs
112
+
113
+
#### How many subscription groups can I have?
112
114
113
-
{% faq %}
114
-
{% faqitem How many subscription groups can I have? %}
115
115
Your Engage space includes up to 25 subscription groups.
116
-
{% endfaqitem %}
117
116
118
-
{% faqitem Can I use subscription groups with templates I've already built? %}
117
+
#### Can I use subscription groups with templates I've already built?
118
+
119
119
No. Templates you've previously created aren't compatible with subscription groups. To use subscription groups, you'll need to create new templates that include new unsubscribe and manage preference links.
120
-
{% endfaqitem %}
121
120
122
-
{% faqitem What happens if I delete a subscription group? %}
121
+
#### What happens if I delete a subscription group?
122
+
123
123
If you delete a subscription group, Engage will still maintain the preferences of the group's end users.
124
-
{% endfaqitem %}
125
124
126
-
{% faqitem What subscription group events does the Engage Channels Source send? %}
127
-
The [Engage Events Source](/docs/connections/sources/catalog/cloud-apps/engage-events/) tracks four subscription group events: `Email Unsubscribed`, `Email Group Unsubscribed`, `Channel Subscription Updated`, and `Group Subscription Updated`.
125
+
#### What subscription group events does the Engage Channels Source send?
128
126
129
-
{% endfaqitem %}
127
+
The [Engage Events Source](/docs/connections/sources/catalog/cloud-apps/engage-events/) tracks four subscription group events: `Email Unsubscribed`, `Email Group Unsubscribed`, `Channel Subscription Updated`, and `Group Subscription Updated`.
130
128
131
-
{% faqitem How can users opt back in if they've unsubscribed from all groups? %}
129
+
#### How can users opt back in if they've unsubscribed from all groups?
132
130
If a user unsubscribes from all of your subscription groups, they'll need to re-subscribe by explicitly opting back in to each group.
133
-
{% endfaqitem %}
134
131
135
-
{% faqitem Do subscription preference links work in test emails? %}
132
+
#### Do subscription preference links work in test emails?
133
+
136
134
Yes. Test emails include fully functional unsubscribe and subscription preference links. If a test email recipient unsubscribes using a test email, Segment updates that user's subscription state. <br><br>
137
135
138
136
Test emails temporarily override an email subscription state. This means that an unsubscribed email address can receive a test email but won't receive regular email campaigns from which they've unsubscribed.
139
-
{% endfaqitem %}
140
137
141
-
{% faqitem Should I follow any conventions when naming a subscription group? %}
138
+
#### Should I follow any conventions when naming a subscription group?
142
139
143
140
Yes. Keep the following table in mind when you name a subscription group:
144
141
@@ -151,6 +148,3 @@ Yes. Keep the following table in mind when you name a subscription group:
151
148
| Spaces in Group Names | Spaces aren't allowed at the beginning and/or end of the Group name |
152
149
| Unsupported characters for Group Names | `!@#$%^&*()_+\-=\[\]{};':"\\|,.<>\/?` |
153
150
| Unsupported accent characters for Group Names |`á, é, í, ó, ú, à, è, ì, ò, ù, ë, ï, ã`|
Copy file name to clipboardExpand all lines: src/unify/Traits/predictions/using-predictions.md
+21-23Lines changed: 21 additions & 23 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -70,49 +70,47 @@ Predictions may not be as beneficial in the following situations:
70
70
-**When you sell limited but highly-priced items**, like enterprise software, complex medical machines, and so on; this also applies if you're in the B2B sector.
71
71
-**When you don't yet have enough data**; your model could produce errors if, for example, your target is too new and lacks sufficient data. Waiting a month could allow Segment to gather more predictive data.
72
72
73
-
## Frequently asked questions
73
+
## FAQs
74
+
75
+
#### What type of machine learning model do you use?
74
76
75
-
{% faq %}
76
-
{% faqitem What type of machine learning model do you use? %}
77
77
Segment uses a binary classification model that uses decision trees.
78
-
{% endfaqitem %}
79
78
80
-
{% faqitem What level of confidence can I have in my predictions? %}
79
+
#### What level of confidence can I have in my predictions?
80
+
81
81
Once Segment creates your prediction, you can check the model statistics page, where Segments shows you how the model was created. Segment also maintains automated systems that monitor model performance and will alert you if your model is not predictive.
82
-
{% endfaqitem %}
83
82
84
-
{% faqitem How long do predictions take to create? %}
83
+
#### How long do predictions take to create?
84
+
85
85
Trait creation depends on the amount of data, but Segment expects predictions to be completed in around 24 hours. For larger customers, however, this could take 48 hours. Predictions shows a status of `In Progress` while computing; Segment updates this status when customers are scored.
86
-
{% endfaqitem %}
87
86
88
-
{% faqitem What are AUC, log loss, and lift quality? %}
87
+
#### What are AUC, log loss, and lift quality?
88
+
89
89
These data science statistics measure the effectiveness of Segment's predictions when tested against historical data. For more information, refer to [ROC Curve and AUC](https://developers.google.com/machine-learning/crash-course/classification/roc-and-auc){:target="_blank"}, [The Lift Curve in Machine Learning](https://howtolearnmachinelearning.com/articles/the-lift-curve-in-machine-learning/){:target="_blank"}, and [Intuition behind log-loss score](https://towardsdatascience.com/intuition-behind-log-loss-score-4e0c9979680a){:target="_blank"}.
90
-
{% endfaqitem %}
91
90
92
-
{% faqitem What is the Prediction Quality Score? %}
91
+
#### What is the Prediction Quality Score?
92
+
93
93
The Prediction Quality Score factors AUC, log loss, and lift quality to determine whether Segment recommends using the prediction. A model can have a score of Poor, Fair, Good, or Excellent.
94
-
{% endfaqitem %}
95
94
96
-
{% faqitem How do you store trait values? %}
95
+
#### How do you store trait values?
96
+
97
97
The created trait value represents the user's percentile cohort. This value refreshes every seven days. If you see `0.85` on a user's profile, this means the user is in the 85th percentile, or the top 15% for the prediction.
98
-
{% endfaqitem %}
99
98
100
-
{% faqitem How frequently do you re-train the model? %}
99
+
#### How frequently do you re-train the model?
100
+
101
101
Segment rebuilds the machine learning model every 30 days.
102
-
{% endfaqitem %}
103
102
104
-
{% faqitem How frequently do you update trait values? %}
103
+
#### How frequently do you update trait values?
104
+
105
105
Every seven days.
106
-
{% endfaqitem %}
107
106
108
-
{% faqitem How many predictions can I have? %}
107
+
#### How many predictions can I have?
108
+
109
109
You get five predictions as part of Engage Foundations or Unify Plus. To purchase more predictions, reach out to your CSM.
110
-
{% endfaqitem %}
111
110
112
-
{% faqitem Are there any known Predictions limitations? %}
111
+
#### Are there any known Predictions limitations?
112
+
113
113
Yes. Keep the following in mind when you work with Predictions:
114
114
115
115
-**Predictions made for more than 100 million users will fail.** Segment recommends making predictions only for non-anonymous users, or, as an alternative, use the Starting Cohort to narrow down the audience for which you want to make a prediction.
116
116
-**Predictions may not work as intended if you track more than a thousand unique events in your workspace.** If this applies to your use case, [contact Segment Support](https://segment.com/help/contact/){:target="_blank"} for help with removing unused events, which will allow you to create predictions.
Copy file name to clipboardExpand all lines: src/unify/identity-resolution/index.md
+2-2Lines changed: 2 additions & 2 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -26,9 +26,9 @@ Identity Resolution sits at the core of Segment. The Identity Graph merges the c
26
26
4.**Maintains persistent ID** - multiple external IDs get matched to one persistent ID
27
27
28
28
29
-
## Frequently asked questions
29
+
## FAQs
30
30
31
-
### Can I use the Profile API on the client-side?
31
+
####Can I use the Profile API on the client-side?
32
32
For security reasons, Segment requires that the [Profile API](/docs/unify/profile-api/) only be used server-side. The Profile API allows you to look up data about any user given an identifier (for example, email, `anonymousId`, or `userId`) and an authorized access secret. While this enables powerful personalization workflows, it could also let your customers' data fall into the wrong hands if the access secret were exposed on the client.
33
33
34
34
Instead, by creating an authenticated personalization endpoint server-side backed by the Profile API, you can serve up personalized data to your users without the risk of their information falling into the wrong hands.
0 commit comments