HypeAuditor is constantly improving its algorithm, and now we are excited to announce that the Audience Quality Score formula was enhanced significantly. Now it includes a lot of additional information.
Previously, we recommend you to study the whole report in detail. AQS was the result of 8 criteria, which are split into four key categories:
- Engagement Rate
- Active Audience type
- Followers and Followings growth patterns
- Comments Authenticity.
From now, AQS includes 15 criteria, so the score becomes more comprehensive. Let’s look at some of these changes in more detail.
The summary of an account’s quality
Next to the AQS, you’ll find detailed information with the list of problems an account has or vise-versa their strengths.
More Tiers — a fairer assessment
Previously, we’ve divided all influencers into five tiers and compared all metrics within the Tier to assess whether the result is good or not. Here are these tiers:
Tier 1 – Mega Influencers and Celebrities over 1 million followers
Tier 2 – Macro Influencer — 100K—1M followers
Tier 3 – Mid-tier Influencer 20K—100K followers
Tier 4 – Micro-Influencer 5K—20K followers
Tier 5 – Nano-Influencer 1K—5K followers.
And now we have over 150 tiers. We compare all metrics with influencers who have almost the same number of followers. The assessment becomes much fairer.
Brands vs Brands
One of the key differences in the new AQS formula is that we started to separate personal and brand accounts. It is well known that the Engagement Rate of Brands’ accounts is much lower than the influencers’. The audience is more engaged in communications with other humans.
That’s why we decided to compare brands only with brand accounts to make the result fairer.
In most cases, AQS on brands’ accounts has increased. Below, you can see how these changes have affected Fashion Nova Instagram account.
Deeper engagement analysis
Our engagement analysis becomes more in-depth, and now we show the authenticity of the Engagement Rate.
Next to the AQS, you’ll find detailed information on likes and comments authenticity.
Furthermore, we didn’t check whether an influencer has restricted the comments on their posts, before. And now, we check it and do not vastly lower the AQS if the comments are off.
More data for analysis
To get audience insights, we analyze a sample of an influencer’s audience who engage with the creator’s content. The size of this sample is chosen by the machine learning algorithm and depends on a variety of factors, but this amount is always statistically significant.
Anyway, we decided to enhance the sample size, so the Audience Type analysis becomes even more accurate. The difference is only a few percents, but even small improvements are valuable for us.
To each his own
AQS considers Followers and Followings growth patterns. Previously, we had an equal punishment for any anomalies on the graph. Now we analyze how many spikes are there and lower AQS accordingly.
We also lower AQS if an influencer uses Comments Pods, but now we analyze how many comments are coming from Pods and decrease AQS, respectively.
There are a lot of influencers who do not post content for months. In the new AQS formula, we lower AQS on accounts that left abandoned for 90 days and more. Moreover, we’ll decrease the score every month until an influencer becomes active again.
Our goal is to make influencer marketing transparent, fair, and effective. And we are making every effort to evaluate influencers as efficiently and equitably as possible.
The changes have affected all brands’ accounts and those influencers who had problems with their authenticity before. The AQS on the reports of quality influencers has changed slightly.