Benchmark presents an average market value for a given metric. This is calculated by analysing a particular metric across influencers in a given genre and arriving at an average score for that particular metric. An influencer’s score for that metric in the particular genre can then be said to be above or below benchmark.
Growth rate is the ability of an influencer to be consistently relevant.
Yearly growth rate is calculated by the number of followers an analysed influencer had 365 days ago, subtract the number of followers they had yesterday, and compare the difference to calculate the percentage. This is updated on a daily basis.
Location is detected basis the following:
The algorithm analyses multiple patterns and prioritises them and identifies most frequently repeated ones with highest priority being given to location geo tag in bio and least given to geotag in last analysed post.
In order to correctly identify topics we analyze bio and content of influencers' posts. We use methods of mathematical statistics and machine learning to understand their meaning and, based on this, indicate the appropriate topics. At the moment we have more than 130 topics in The database.
The results are checked daily by a team of trained QA and are used to further train the topic detector. In addition, influencers themselves can suggest edits to the topics on the reports, and this data is viewed by The QA and used for the additional training of the algorithm.
Similar Influencers feature is an effective way to reach new people who are likely to be interested in the business because they're similar to the audience of the best-performing creators.
We identify the common qualities of influencers and their audiences and provide you a list of similar accounts considering:
We update The Instagram report as per the following metrics on a daily basis:
The system detects new trends and behavioral patterns by recalculating the above metrics every 30 days.
AQS is a high level 1-100 metric which is used to measure an influencer’s overall quality in the influencer marketing industry.
Influencers with AQS '100' are most likely to have a high-quality audience and alternatively the ones with AQS '20' have a low quality audience who will have little impact Influencers with AQS '100' are most likely to have a high-quality audience and alternatively the ones with AQS '20' have a low quality audience who will have little impact
Get access to a Clan report to get a bigger picture and dive deeper into the metrics.
Yes, it can change but the % change may not be substantial in a short span of time. AQS is updated and calculated on a daily basis. It is a relative metric to audience quality to accounts with similar follower count and also reflects any changes happening in the market such as the audience quality of an influencer. All the account metrics are compared with similar accounts only.
It is incorrect to make assumptions about the true worth of an influencer just based on the Engagement Rate only. Engagement Rate may not always be the true indicator of its comments, number of likes and quality of the followers, which are very important elements for any influencer.
Ways to check the audience quality of an influencer:
If some anomalies such as an abnormal growth rate is identified in the Growth metric then it’s most likely that some inorganic methods were used to inflate the followers number artificially.
It used techniques such as image recognition to identify a user's Ethnicity. The algorithm is based on an extensive database of different human faces. The technology tracks multiple facial features and produces an estimate on the ethnicity of each individual.
It is ascertained through the content of a user’s posts and texts in their bio and processes and identifies their language basis that.
The Quality Audience metric reflects the number of followers whose activity is not identified as suspicious. On Clan reports, Quality Audience includes 'Real People' and Influencers and excludes ‘mass followers’ and ‘suspicious accounts’.
The Authentic Engagement metric shows the average number of likes and comments that come from ‘Quality Audience’ or real people and influencers per post.
We categorize Instagram users on the basis of the following Audience Types-
Instagram user who have more than 5,000 high quality followers and less than 1,500 following (so that it’s not identified as a mass follower) is an influencer.
Instagram users with more than 1500 followings are identified as Mass Followers.
Most of these accounts tend to take advantage of third party follow/unfollow tools to grow their followers' count in an inauthentic way by following loads of accounts in the hope that they’d follow them back, and when they do, the software used by these accounts unfollows the profiles that followed them. Mass Followers do not see posts from all users they follow. The Instagram algorithm shows posts only from those accounts with whom this user has interacted lately, hence bringing down the reachability of the influencer.
These are the accounts that come under The radar for using Instagram bots, fake or stolen accounts, third party services for fake likes, comments and followers. There is a machine learning process to identify such suspicious accounts. The algorithm's classification is based on a decision tree which takes the following metrics into consideration while identifying them-
A high follower count does not always signify good because it can cause major harm to your account and business potential if you have a high number of fake followers. A good quality of followers will always impact your engagement rate positively.
For e.g. A ghost follower is that suspicious looking Instagram account that follows you but never engages with anything you post (i.e. never likes or comments on any of Instagram content). Look for the following red flags while spotting ghost followers-
It does not serve a true purpose for serious social media marketers, influencers, and content marketing professionals to only have a huge following. They must be watchful of average engagement rate & authenticity rate. Audience quality takes more prominence while having ghost followers may become a block for positive growth of business.
if you have a business account then you should manually block them or remove them if you have a personal account. It may take time but it will be worth the effort for your social marketing and monetization in the long run.
It's possible to have some suspicious accounts following you as some of them might have abandoned their account as they forgot their credentials or quit social media altogether. In addition, there are some commercial Instagram accounts by e-commerce or business companies looking to promote their page using the follow/unfollow method.
The goal here is to keep the rate of these suspicious accounts as low as possible. You can do that by simply checking the Audience Authenticity section in the Clan report on a regular basis. This metric raises a red flag if the number of suspicious accounts following you gets high.
Audience Reachability suggests the ratio of the followers who can see the posts in their feed. According to Instagram’s algorithm, audiences who follow less than 1,500 accounts are considered reachable, as they’re more likely to see the post in their newsfeed.
Scroll below to see the comparison of an 'Average' and 'Excellent' Audience Reachability score. In the first case, the user's audience is slightly less reachable, since about 15% of its followers are mass followers who follow more than 1,500 accounts.
Audience Authenticity suggests the percentage of real and authentic audience amongst the followers. It counts real people and influencers following you as real profiles and excludes mass followers and suspicious accounts.
These ranks indicate whether the quality of audience is very good, average or can be improved. Again, each metric has a set benchmark amongst similar size influencers.
Reachability is determined by calculating the number of followers that have less than 1500 followings. As per Instagram’s algorithm, these are the accounts that are more likely to see the post on their scroll feed, hence making the account reachable.
To improve the audience reachability score you must get rid of the suspicious accounts and mass followers.
Estimated Reach means the estimated number of unique people a post might reach to or the number of people who may see it.
The system calculates the Estimated Reach by taking the following metrics into account:
Have a closer look at the examples down below and try to spot the difference between Audience Interests of 2 Instagram users. Can you take a guess as to which one is a fashion blogger and which one is a youtuber?
The system processes and analyses the content of the posts of the active and engaged audience of the influencer to determine their Audience Interests. Clan’s sophisticated technology uses topic modeling algorithms to detect topics for each account to do the same.
Followers growth analysis is used to identify any suspicious or unusual patterns in the growth of an influencer’s audience. Sudden spikes in the followers of an influencer indicates that they might have resorted to an artificial way of boosting followers.
Having said that, there are some exceptional cases wherein a sudden follower growth happens when a user participates in a loop giveaway or becomes part of a TV show.
At Clan, we’ve implemented an automatic graph analysis that is performed by The machine learning algorithm regularly that tells you about any suspicious patterns via the graph.
Check The blog post "How to read Followers & Following graphs on Clan" to get more information about Followers’ & Followings graphs.
We first collect enough historical data to analyse and show the dynamics via a growth graph and The system takes about a week to update this data. If you are not able to see the graph in the report, come back and check for it later.
Engagement Rate of a post defines the level of engagement that a piece of content receives from an audience. I.e how many people liked, commented and interacted with that piece of content as per the total number of followers the influencer has. The overall ER, however, defines the average engagement all the pieces of content get on the influencer profile using the same metrics.
We calculate the ER by combining the number of likes and comments and dividing it by the follower count and the number of posts for the given period.
At Clan we use the latest 12 posts to calculate influencer's ER.
Likes Spread denotes the spread of the number of likes amongst different posts on an influencer’s feed. It shows the difference in the number of likes between their posts. If an influencer has a low level of Likes Spread, it might mean that they have resorted to inorganic ways of boosting the number of likes per post. This is an important metric in ascertaining fake engagement because on an average, every piece of content on a feed receives different engagement depending on the quality of that post. That's why it's suspicious if all posts have the same amount of likes.
As the name suggests, Comments Authenticity is a metric that checks for the authenticity of comments which means checking whether the comments come from real or suspicious accounts or accounts that are a part of Instagram Pods (Engagement Pods/Boost groups) that provide extra exposure to these accounts to boost their post engagement in a non-authentic way.
While checking for Comments Authenticity, multiple factors like the content of the comment and the quality of the account that left that comment are taken into account. Clan identifies a comment as inauthentic if it matches several negative factors at once:
If a post has only about 1- 3 suspicious comments out of 20, we don’t mark it as inauthentic. However, If 50% or more of the comments per post are marked as inauthentic, it means there’s something wrong. During the first round of check, even the suspicious looking comments are not marked into an inauthentic group. It’s only when the account who left those comments is identified as performing some suspicious activity is when we mark the comment as inauthentic.
Clan marks comments to tag-to-win giveaways and contests as non-authentic because asking friends and followers to tag non relevant people in comments to win a giveaway or contest is an inauthentic way of growing engagement and is considered non-authentic. Instagram promotion guidelines states/recommends, "You must not inaccurately tag content or encTheage users to inaccurately tag content."
Engagement pods, also known as Instagram pods, Comment Pods, or Boost groups are private groups created by bloggers who engage with each other’s content in the view of gaining extra exposure on their social media posts, mainly, to boost the engagement activity on their profiles. These private group chats are created on Facebook, Telegram, Whatsapp or other such platforms
Users of such pods typically drop links of their post in the group and ask others to ‘like, comment (3 words and more), & save’ their post. They then return the favThe by engaging with other posts left by other bloggers. Influencers are usually not part of one, but multiple such groups. This type of activity is definitely an inauthentic way of gaining engagement and is hard to spot without an implementation of AI.
At Clan we’ve developed The technology in a way which detects when a comment comes from an Engagement Pod. We’ve done this by feeding The system with thousands of examples of comments that come from Engagement Pods. The comments are then analyzed and sorted into different groups according to various patterns. Not only does The system check the comments for any suspicious signs, but also checks the corresponding accounts, thus the error rate is extremely low.
Therefore, the higher the percentage of suspicious comments among the posts, the lower the Comments Authenticity and vice versa.
Keeping a check on the Like- Comments ratio can help ascertain whether an influencer
bought any of those.
Via the above graph you can see that if an influencer purchases the same number of likes and comments all the time, all the dots will be concentrated in a bubble indicating that they’ve adopted an inauthentic way of getting likes and comments, whereas a well distributed set of dots can hint towards organic engagement.
The price of a post is an indicative estimate and not definitive. There are multiple factors like geo/country, genre/category, followers, ER and audience quality which are factored in to arrive at a baseline for a specific type of influencer in a particular category. By approximating market values from industry learning added with machine learning, the price is arrived at.
Please note that this will just be an estimate and the actual price might differ from it. However, in multiple cases, the influencer may choose to disclose the price willingly in which case, the price is definitive.
Applicable mostly for an influencer or celebrity who has more than 8 million followers. This is because there are multiple factors which determine which are subjective and in most cases, the deal/cost is not known and it could also be a part of a bigger brand deal.
Hence, it is not possible to ascertain this value. However, there are other standard metrics which are available which may help you determine the effectiveness of the influencer/celebrity.
The price of a story is an indicative estimate and not definitive. There are multiple factors like geo/country, genre/category, followers, ER and audience quality which are factored in to arrive at a baseline for a specific type of influencer in a particular category. By approximating market values from industry learning added with machine learning, the price is arrived at.
Please note that this will just be an estimate and the actual price might differ from it. However in multiple cases, the influencer may choose to disclose the price willingly in which case, the price is definitive.
Advertising performance is a very important metric for brands trying to ascertain how well the sponsored post that the influencer did for them (in terms of likes and comments) worked in comparison to their usual posts. The algorithms detect, analyse, and process influencer's posts with AI (Artificial Intelligence) to optimise the advertising post. You can track the quality and effectiveness of the sponsored post by checking the Brand Mentions section in the Clan report.
Example- If an influencer were to have an average ER of 10.62% and that of the advertisement posts as 2.06% on average, it would mean that their audience didn’t engage with the specific sponsored post as much as they usually engage with their other posts. Perhaps the ad didn’t look natural enough.
The Brand Mentions section in the Clan report gives a list of brands that were mentioned by influencers in their post captions. It has a recency window of the last 180 days.
Example of a post with a brand mention:
To do this, the system uses an AI that analyzes the following attributes:
While working with influencers, one thing brands should be concerned about is their Advertising post frequency. i.e How often does an influencer perform advertisement activity and How many brands have they mentioned in the recent past. Advertising Post Frequency is the number of posts per week with brand mentions on the influencer’s feed.
Now the catch here is that the ideal influencer should have a balanced number of sponsored and organic posts. It’s not advisable to work with influencers who ONLY publish paid content. A decent percentage of advertising posts in the influencer’s feed should not be more than 10-20% of the total number.
Post Frequency displays the number of posts an influencer publishes per week. While it’s important to be consistent with posting the content, It is equally important to maintain a balance and make sure you’re not spamming the audience. While 1 post a day might keep the audience consistently engaged, posting 5 pieces of content in a day will only spam them.
Brands should keep this criteria in mind while selecting influencers to work with. In case a blogger posts several times a day, it is advisable to come to an agreement with them beforehand to only do that one post that day, as it will concentrate the engagement to the brand’s sponsored post.