About Ylytic Data and Insights

What is 1st party data, and how does Ylytic help fetch that?

1st party data is obtained with the creator's consent and is directly provided by the respective platform (Instagram, YouTube, etc.) through their official APIs. Ylytic has integrated with these platforms and provides tools (Custom Microsite, OAuth Flows, Data, and Insight APIs) that enable businesses to get creators’ consent and access this data without the need for them to do any development work of their own. Our OAuth flows provide a 2-click permission experience for creators in seconds.

First Party API

What is 3rd party data, and how does Ylytic help fetch that?

3rd party data is public data that a creator has made open to all on their social accounts. Ylytic uses a combination of official platform APIs, services for reading public data (both in-house and 3rd party), and ML and AI estimation techniques to provide a good estimate of this data. While the publicly available data points are available as they are seen, it is important to note that insights from 3rd party data are based on sampling and estimation techniques and may not always be comparable to 1st party data. E.g. Audience demographic may be of overall followers given by a platform but using 3rd party Ylytic processes audience demographics of only those followers that it deems authentic or who have recently engaged with the creator. In this case, therefore, 3rd party data could give a better representation of a creator’s overall metric.

Third Party API

What sample size does Ylytic use for computing the audience demographics?

Ylytic uses dynamic sample sizes varying between 500 and 5000 followers, likers, and commenters for computing audience demographics of any creator. In this method, sampling continues till a stable value is found for a metric and the value stabilizes or doesn’t change much on increasing the sample size. This makes the sampling statistically significant and reliable.

How do Audience Demographics for 1st and 3rd party data differ for various social platforms?

Social platforms have direct access (either provided by user themselves or permitted collection) to their users’ information like their location, gender, age, interests (though all of these may not be authentic accounts). However, to computing audience demographics via 3rd party methods, Ylytic uses only publicly verifiable information to construct AI/ML estimation models (Image recognition, Text recognition, Location Detection and so on). These models are checked for statistical significance. One advantage of this process is that it considers data of only the likely authentic audience of the creator and also takes into account recent engagers (likers, commenters) in addition to followers, many of whom are passive. Hence, this may be a better representation of a creator in many cases.

How does Ylytic estimate Audience Authenticity for a creator?

Audience Authenticity is estimated through proprietary algorithms as below. Fake/Inactive Followers: Categorized basis an account’s name, display picture, followers, following, content posted, and frequency of content posted. Mass Followers: If an account follow more than 1500 accounts on Instagram. Such an account's probability of interacting with one particular creator's content is extremely low.

How does Ylytic estimate Audience Age and Gender for a creator?

Ylytic's proprietary image recognition algorithm can detect gender and age from public pictures an account posts/uses. We also validate results with AWS Image recognition which is a global standard. The image processing accuracy for valid images is 99%+ on an aggregated basis.

How does Ylytic compute Audience Location and Interests for a creator?

Location and Interests are computed by processing a creator's likers, commenters, and followers' profiles and posts. If a user has tagged multiple locations, the location with highest frequency is selected keeping in mind a few other probabilistic and time factors. Similarly, the content of numerous followers' posts is accessed and categorized to assign various interests to that account.

How does Ylytic calculate the Engagement Rate of a creator?

Engagement Rate is computed as an average percentage of interactions by the audience for the latest* posts/reels.

How does Ylytic calculate the Effective Follower Rate of a creator?

Effective Follower Rate is the ratio of active viewers to total followers. It is calculated by taking median views of the last 90 days' posts as a percentage of followers.

How many posts does Ylytic compute for calculating metrics like a creator's Engagement Rate, Average Views, and Effective Follower Rate?

For YouTube, Ylytic processes the latest 10 videos and shorts each for computing various metrics; for Instagram, Ylytic processes the latest 12 reels and posts each.

At what frequency does the data refresh on the platform?

Third party creators’ analytics except audience is refreshed every 7 days for 30 days. First party creators’ analytics except audience is refreshed every 3 days for 30 days. Campaign insights except audience is refreshed every 24 hours for 30 days or till the end date selected.

How does GDPR affect third-party data?

Ylytic only analyses explicitly consented or publicly available data and doesn't deal with personal or sensitive data, which is a crucial focus of the GDPR.

How long does it take to set up?

Our self-serve products can be set up entirely by you within minutes if you use manual uploads (CSVs) or exposed API endpoints. If you have to integrate with databases (recommended), a technical team member can do this through our easy-to-embed code within a couple of hours. We also provide live support for this integration free of charge.

Is my data protected and secure with Ylytic?

We understand that your data is private and ensure it stays protected. We do not share private creator data permitted for a client with anyone else including Ylytic. If you want us to share your data with your partners, we do that only after your explicit consent. A creator’s private data is not shared with anyone except those they have provided explicit consent for. Whenever we use any data for mining insights, we use it in a wholly anonymized form with no personal footprints. All data is stored in cloud servers with world class encryption standards. To know more about how we protect your data write to us at [e-mail]