JavaScript SDK for "Do They Use" service
"Do They Use" service is made to help teams to make data-informed decisions for their products. The way it helps to make data-informed decisions - is by providing ready-to-go quantitative data reports about their product usage by users for the teams. Teams get visual answers on various questions about their product's elements and features usage and can make decisions based on that data.
Service works similarly to Google Analytics, but aims not Marketing and Sales teams, but Product teams. This service helps teams to find answers on the following quantitative questions:
- Do users use specified (tagged) web elements (buttons, links, check-boxes, etc.)?
- If they do use them, then how many, how often they use each of them?
- Which data they choose (in check-boxes, drop-downs, etc.) and how many?
- And by the way, who of users and users groups exactly - maybe team simply don't care of them in a product (trial users, testers, managers, etc.)
This information helps team to:
- decide to remove not used element(s) at all;
- decide to take actions to increase or decrease element(s) adoption (number of users who uses them);
- decide to take actions to increase or decrease element(s) usage frequency (number of element(s) calls by some group of users);
- compose questions for gathering qualitative data from users - to ask users more specific questions, like: "we noticed, that you use these (element(s), feature(s)) a lot, can you explain why?" and then use answers on these questions when talk to users who uses less or don't use at all.
Addressed answers on these questions lead to better UX both for Product teams and for Product users. Product gets better and this better is driven not by assumptions but by real usage data by real users.
- SDK is made for "Do They Use" web service to simplify:
- adding listeners for specified events ("click", "change", etc.) for specified elements (link, checkbox, dropdown, etc.) of web page(s) it is installed for;
- translation these events into web log reports;
- sending these reports to "Do They Use" service API.
- This repository starts teamwork and collaboration for users of "Do They Use" service and of this SDK.
- Open SDK demo HTML page
- Open browser's dev console
- Change drop-down value on demo page
What to expect: on each drop-down value change in console will appear a new report.
When installed in web page (see for example SDK demo HTML page), SDK listens to events from specified (tagged by data-dtu attribute by default) elements on web page, builds JSON reports and sends them to specified callback (which is console.log by default) function. callback could be a function that sends these reports to API (and this is how this SDK is used in "Do They Use" service).
List of currently supported elements types is available here. You can play all currently supported elements in a Story Book.
All currently supported web elements are available in a Story Book published with GitHub Pages.
- At the bottom of your web page (right before
</body>closing tag) import and init SDK with anyctag(for example,TEST CTAG):
<script src="https://alexeyhimself.github.io/dtu_sdk_js/dtu_sdk.js"></script>
<script type="text/javascript">
const dtu = dotheyuse({
'ctag': 'DTU CTAG', // ctag - is an analogue of gtag in Google Analytics - a unique key for different orgs for DTU deployed as a SaaS
});
</script>(check out how it is done in code of SDK demo HTML page and how it works on SDK demo HTML page itself).
- Check that SDK was installed correctly. In browser's dev console execute:
dtu.statusCorrectly installed SDK will reply:
'Ready'(you can try this in SDK demo HTML page as well).
Incorrectly installed SDK will reply with an error message related to that specific problem.
- To any (of supported types) element of your web page add
data-dtuattribute. For example:
<select data-dtu="some dropdown">
<option>value 1</option>
<option>value 2</option>
</select>- Fire supported event for this element. For example, change value in this drop-down.
- In browser's dev console you will see:
{ctag: 'DTU CTAG', topic: 'default', feature: 'some dropdown', value: 'value 2', …}You can try it in in SDK demo HTML page.
- Set your own
callback(my_custom_function, for example) function:
<script type="text/javascript">
const dtu = dotheyuse({
'ctag': 'DTU CTAG',
'callback': my_custom_function
});
</script>- Change default (
data-dtu) bind attribute (todata-testidfor example):
<script type="text/javascript">
const dtu = dotheyuse({
'ctag': 'DTU CTAG',
'dtu_attribute': 'testid'
});
</script>Please note, that data- preffix is omited.
- Specify
topic,uid,ugidsexplicitly:
<script type="text/javascript">
const dtu = dotheyuse({
'ctag': 'DTU CTAG', // analogue of gtag for Google Analytics
'topic': 'custom', // subdomain for ctag
'uid': '123',
'ugids': ['Free Trial', 'Professional', 'Admin']
});
</script>| Parameter | Mandatory | Default | Examples | Description |
|---|---|---|---|---|
ctag |
No | 'DEMO MVP' | 'x1298yveve778', 'BI Team', 'Some company department' | Originally stands for "company tag". It's an analogue of gtag for Google Analytics. ctag is made to support DTU to be deployed as a SaaS and support many orgs. ctag must have any (non-empty) string value for self-hosted setups (for example, name of the department or a product inside your org. |
topic |
No | 'default' | 'Some product v1', 'Some team, Some product, v1.5' | topic is made to allow sending analytics to different isolated spaces inside 1 ctag. Imagine ctag as a company key, as a domain, and topic as a product / project key, as a subdomain. If you need to slice analytics, and send new data to an empty space, then you may change topic - and data in a new topic will be separated from the data with other topics |
uid |
No | 'you@example.com' | '123456', 'sdf876080870ewv', '98765:12345678', 'name@email.com' | Stands for "User ID". uid may be real UID (a number, email, etc.) or any substitute for UID (salted hash, random string or number). uid is made to let you distinguish data for different User IDs. If not set, then SDK will automatically generate and save synthetic uid into user's browser and all the data will be from 1 default uid |
ugids |
Yes | ['Visitor'] | ['Paid', 'Admin'], ['Free Trial', 'User', 'Owner'] | Stands for "User Groups IDs". ugids is made to introduce different dimentions for the roles and permissions of the uid |
dtu_attribute |
No | 'dtu' | 'dotheyuse', 'analytics' | Was introduced to avoid situations when you already have data-dtu attribute in your product, and it is not related to DTU analytics. In this case, you can set another data attribute name to bind to for a DTU analytics: data_attribute: someting - and now DTU will bind elements with data-something instead of elements with data-dtu |
callback |
No | console.log | my_custom_function | Made to let you define custom callback functions instead of sending data somewhere |
api_url |
No | 'http://localhost' | 'https://yourcompany.com/dtu/analytics', 'http://10.10.1.1' | To set any API endpoint to send analytics |
- Install Jest framework:
npm install --save-dev jest- In project directory run:
npm test- In project directory run:
npx jest --coverage- In project directory will appear
coveragedirectory withindex.htmlanddtu_sdk.js.htmlfiles