.png)
Metadata and Participant attributes
To be able to make a distinction between different customer groups, personalize your survey, etc., we advise sending metadata with your participants. How you can do this depends on the type of touchpoint you're working with. To learn how to do this for the different touchpoint types, you can consult the following sources:
IN THIS ARTICLE
1. What is metadata?
Metadata is data that describes other data. In the context of Hello Customer, it's data that describes your participants or the interaction they had with your company.
The two main reasons why you would add metadata are:
- Personalize your survey to increase response rates.
- Filter your results for a certain customer group and compare two groups against each other.
There are two main types of metadata:
- Personal data
- Personal information about the participant: email address, first name, last name, telephone number, and customer ID. These specific metadata are treated a bit differently than other metadata since they are not available as filters and are mainly used to personalize your survey.
- Information that is not sensitive but is also linked to the person, for example, the gender of a person, the birth date, etc.
- Transactional data
- Data that describes the context of the interaction between a customer and a company Examples are:
- In which city or country did the customer buy an article?
- Was there a discount on the articles or not?
- Data that describes the context of the interaction between a customer and a company Examples are:
We can also differentiate a third type:
- Data about the actual interaction with your company. Examples are:
- Total amount spent
- Number of items purchased
- Purchase date
It is important to decide which metadata you want to send to the Hello Customer platform. Only send along data that you really want to use to personalize your survey or filter on.
2. Adding and managing metadata keys per touchpoint
2a. How to get there
You can manage your metadata keys in the settings of the relevant touchpoint.
Option 1: Click on the touchpoint settings icon of the touchpoint of your choice.
Option 2:
Step 1: Click on touchpoints in the left navigation menu
Step 2: Choose the touchpoint you want to add or manage the metadata for. If you do not find your touchpoint immediately, try clicking "show more".
Once you are in the settings of the touchpoint of your choice, you can manage the metadata keys:
2b. Add a new metadata key
To add a new metadata key, click on "Add new metadata" in the manage metadata screen:
- Choose a name for your new metadata key. In order to work properly, this name needs to be the same as the column name of your Excel or CSV file you use to upload your participant (Email touchpoints) or needs to match the name you use in your integration code (Website and In-app touchpoints) or URL (Ask Anywhere touchpoints)
- Indicate if the metadata key is mandatory. In case a key is mandatory, participants without a value on that key will not receive an invitation email (Email touchpoints) or answers will not be added to the platform (Ask Anywhere, Website and In-app touchpoints)
- Choose if the metadata key is visible as a filter in the platform.
2c. Managing metadata: mandatory and visible
For existing keys, you can choose whether you want to make them mandatory and if they need to be visible in the platform. You can edit these options for an existing touchpoint by clicking on the checkbox
In Email touchpoints, there are two mandatory fields you need to add for all your participants:
- The email address of the participant; otherwise, we are not able to send out an invitation mail to this person.
- The language in which you want to survey your participant. This is the language in which the participant will receive the invitation email. The person can still change the language of the survey once they fill it out.
- Participants in e-mail touchpoints without a value for this metadata key won't receive an e-mail invitation to fill out the survey.
- Answers without values for the mandatory metadata keys integrated in the URL (Ask Anywhere touchpoints) or in the integration code (Website and In-app touchpoints) will not be added to the platform.
Visible metadata
When you make a certain metadata key invisible, it will not be available in the platform to filter on and it will not be displayed in the conversation manager. This is mainly useful if you add metadata to personalize your survey, but that is not handy to filter on. Example:
- Key = Title
- Values = "Sir", "Madam"
- Personalized invitation mail: "Dear Sir", "Dear Madam"
Note that this metadata won't be available in the conversations export either or in your Snowflake environment.
IMPORTANT
Changes on visibility take a day to be processed, so once changed, the attributes will be visible (or not shown) the day after the changes have been made,
2d. Remove metadata keys
To remove metadata keys, you can click the trashcan icon which appears next to the Visible checkboxes. This icon is only enabled (in red), if the touchpoint has not received any answers with values for that metadata key. If values have already arrived, the trashcan is disabled (grey) and the metadata keys cannot be removed from the touchpoint. An alternative to deleting a metadata key is unchecking the Visible checkbox. This way, it won't show on the conversations and won't appear on the exports of conversations either.
3. Turning dumb metadata into smart participant attributes
In order to make filtering on the platform easier than ever, ensure data quality and facilitate analysis across several touchpoints in the future, the concept of participant attributes has been introduced. In the general settings you can find an overview of all the metadata used throughout your organization. Here you can tell the platform of which type of values the metadata exists, so an appropriate filter can be chosen, and you can say whether it's personal or transactional data. This enriched form of metadata is what we call Participant Attributes.
Attributes that are not yet defined, will appear as 'undefined' attributes in the overview below.
3a. Attribute type
As Hello Customer, we noticed that a lot of our customers send similar data to us. Therefore, we created some predefined attributes to allow for market benchmarking in the future and to have predefined data types and context, so you need to choose less settings yourself. As a first step to define your attributes, you can indicate if your participant attribute is one of these predefined attribute types or not. In the dropdown, you will find all possible options.
IMPORTANT
Choose a pre-defined attribute only once per environment. Otherwise, your filters will not work properly.
If no existing attribute type matches your data, you can select 'custom' to define your own type.
3b. Data type
Next, indicate the right data type, to have optimal filtering options in the platform - for example sliders for numerical data - and to have data validation when data starts flowing in.
There are standard and custom attributes. If applicable, it is in your best interest to select standard attributes as they can later on be used to compare your results against the market/competition.
These are the standard attributes with their corresponding definitions:
| Name | Description | Data type | Context |
|---|---|---|---|
| Age | The age the surveyed person had on the moment they had an interaction with the company. | Number | Transactional |
| Birthdate | The date on which the surveyed person was born. | Date | Personal |
| Categories | The type of product / service the surveyed person has / makes use of. Examples:
Please note that this is only valuable when a customer can only have / make use of one product / service. When they can have / make use of several, it is better to add all separate products / services as custom - boolean. |
Text | Transactional |
| Country of transaction | The country where the transaction between the company and the surveyed person took place. Example:
|
Text | Transactional |
| Customer ID | An internal id the company gives to the surveyed person. This is automatically seen as PII in our platform. | Text | Personal |
| Date interaction | The date on which the interaction between the company and the surveyed person took place. Examples:
|
Date | Transactional |
| E-mail address | The email address of the surveyed person. This is automatically seen as PII in our platform. | Text | Personal |
| Employee | The employee who got in touch / helped / etc the surveyed person. Examples:
|
Text | Transactional |
| First name | The first name of the surveyed person. This is automatically seen as PII in our platform. | Text | Personal |
| First time right | When helping customers via email or phone, a first time right is true when you can help the person with 1 answer on the email or with 1 phone call. In all other cases, it was not a first time right, so this attribute is false. | Boolean | Transactional |
| Gender | The gender of the surveyed person. There are three categories:
|
Text | Personal |
| Language | The language in which the surveyed person was surveyed. | Text | Personal |
| Last name | The last name of the surveyed person. This is automatically seen as PII in our platform. | Text | Personal |
| Participant zipcode | The zipcode of the surveyed person. This represents the place where this person lives. | Text | Personal |
| Phone number | The phone number of the surveyed person. This is automatically seen as PII in our platform. | Text | Personal |
| Pieces | The amount of pieces a person bought during a certain transaction. Examples:
|
Number | Transactional |
| Reason of contact | The reason why someone had contact with the company. Useful in for example a customer service touchpoint:
|
Text | Transactional |
| Title | How you should address a person. This can be used to personalize survey emails, but will be less valuable to filter on. | Text | Personal |
| Transaction amount | The value of the transaction the customer had with the company. Examples:
|
Currency | Transactional |
| Type of customer | The different types of customer a company has based on a criterium. Examples are:
|
Text | Personal |
| Years as customer | The amount of years the customer is already a customer of the company (= a measure of loyalty) | Number | Personal |
For the custom attributes, these are the data types you can choose:
- String: the values can not be ordered. Examples are store names, regions etc.
- Currency: the values reflect an amount of money. Examples are fees paid, purchase amount etc.
- Boolean: the values are either true or false. Examples are owner of a loyalty card, VIP-member etc.
- Date: the values reflect a date in time. Examples are purchase date, beginning of contract etc.
- Number: the values are numbers and can be ordered. Examples are age, amount of pieces bought etc.
3c. Context
Indicate if your data is personal (linked to the surveyed person) or transactional (linked to the transaction itself). This context will help you ensure that GDPR ruling will be applied correctly and that generic, personal data will be available for all future transactions.
Please not that:
- Participant attributes mapped as personal will be overwritten every time the same participant (same email address) is added to the platform
- Those mapped as transactional are linked to the specific time the participant was uploaded.
4d. PII
PII, or Personally Identifiable Information, refers to data that can be used to identify an individual. In the platform, certain attributes are automatically flagged as PII, while others can be manually designated as such during configuration. Attributes marked as PII are identified with a green shield icon next to their name.
Marking attributes as PII ensures:
- Restricted filtering options: To protect sensitive information, these attributes cannot be used directly in filters.
- Compliance with GDPR and other regulations: Helps maintain adherence to data privacy standards.
By default, attributes such as Email Address, Phone Number, First Name, Last Name, and Customer ID are classified as PII. For custom attributes, you can enable the PII option if the data includes sensitive information.
It is important to correctly flag PII attributes to maintain data integrity and protect user privacy while ensuring compliance with applicable regulations.
4. Display names for metadata
Sometimes, it's not very clear what information is exactly stored in a metadata key. Also, this key can be in a certain language, while not all users of the platform speak this language. Therefore, we made it possible to add a display name in all user languages for metadata keys.
IMPORTANT
Make sure to still use the original name of the metadata key when uploading new participants in your e-mail touchpoints, when adding metadata in the URL of ask anywhere touchpoints or the integration code of website touchpoints.
There are two ways to add a display name for a metadata key:
- Simply click the display name
- Go to edit mode and click the pencil next to the display name
A modal will open where you can configure a display name in all user languages. It's not obligatory to change this for all user languages. By default, the display name is the same as the original metadata key name in all languages.
Once a display name is added, this will be used in throughout the platform:
in all filters available in dashboards, conversations, analysis and e-mail stats
- in the conversation details
in the touchpoint settings both the original name and the display name are shown
