When it comes to social media analytics, things often are more complicated than they seem. Instagram Insights are no exception. In this article, we share some interesting insights about Instagram and our data partner sources. Comparing them helps to understand their analytics better and make the right decisions.
First, it is important to underline that almost all data in Instagram Insights and our partner are estimated data. They are never straight counts as each proprietary metric is derived through statistical sampling or modeling. Both look at a sample of the data as a representative reflection of the larger population included in the entire set of data.
Sampling allows calculating metrics quickly with a high level of accuracy.
This is true for both Instagram Insights and our social data partner. However, there are a number of areas in which their approach differs.
Sources of data
Our partner and Instagram have different sources of data. On the one hand, Instagram analyzes a sample of an influencer's followers who were active on Instagram over the last seven days.
It means that if an influencer bought a lot of bot followers, who are not active at the moment, they'd disappear from the Insights 7 days after they became inactive. Usually, a bot becomes inactive as soon as it reaches 7500 followers.
On the other hand, to get audience insights, our partner analyzes a statistically significant sample of an influencer's audience who engages with the creator's content.
If an influencer does not use any inauthentic methods to grow their number of followers and engagement, our partner information is fairly similar to Instagram Insights. On the contrary, if an influencer uses some questionable strategies, the number of differences will grow and the influencer's ACS decrease (read more about the Audience Credibility Score).
To detect the audience's location (country and city), Instagram uses a sample of an influencer’s followers who have been using Instagram during the last 7 days. The location data comes from Location Access, a setting on mobile devices. It includes IP address or Wi-Fi connection and specific location information like the device's GPS signal.
Beware that bot accounts are registered automatically using scripts. Thus, these accounts do not have any location information, or the information is hidden using a web proxy. An influencer could easily buy bot followers with Europe or US IPs to look more natural in Instagram Insights. Although more expensive than bots with Russian, Asian, or Latin American IPs, they are not difficult to find.
Our data partner uses a different method. The location of an influencer's followers (country and city) is determined by analyzing the respective location tags, the languages of their bio descriptions and captions under their recent posts. The location tags and captions are gathered from the latest 150 posts (or more than 150 posts provided that they have been published within the last two months).
Based on location tags, a time map is created for the audience: it shows what location (country or city) covers the largest time frame. That way our partner makes sure that even if a person travels a lot (like celebrities or travel bloggers do), but keeps coming back to the same place - this particular country or city will be determined as their location.
Audience age for Instagram Insights is also an estimated metric that is calculated based on audience activity from the last 7 days. Unfortunately, Instagram does not provide its sources of data, but we suppose that it relies on information that users themselves provide.
For our partner, age is determined by analyzing profile pictures and the latest selfies with the help of their photo analysis technology and checking bio descriptions (where relevant). Audience age is available only as an aggregated percentage.
According to the information in the Facebook Creator Studio, gender for Instagram Insights is an also estimated metric calculated based on audience activity from the last 7 days.
Source: Facebook Creator Studio
Instagram doesn't provide its sources of data, but we similarly assume that it relies on information that users provide. If users do not disclose their gender on Instagram, they might use data from Facebook where users self-report their gender and select their preferred pronoun.
Our partner determines gender by analyzing profile pictures and the latest selfies with the help of our photo analysis technology; checking names and bio descriptions (where relevant). Similarly to age, audience gender is available only as an aggregated percentage.
Whenever you make a decision regarding which data to use for your business, pay attention to the sources and their reliability. Can these data help you achieve your goals and make the right decision when selecting influencers?
By partnering with our data provider, we seek to assist advertisers by providing them with data that will make their marketing campaigns more relevant and effective. If your influencers do not use any inauthentic methods for follower growth, our data will be very similar to the equivalent Instagram Insights. This prevents you from investing time and resources in collaboration that cannot help you reach your KPIs.