January 2024

Objective:  

• Boost TikTok's MIV to accurately reflect the evolving landscape of social media.

Date: Change effective from January 1st, 2024 in Discover, Insights, and Data Services.

Changes

To align with the evolving dynamics of social media, particularly the increased prominence of TikTok in luxury brand strategies, we have updated our Media Impact Value (MIV) calculation algorithm.

We have adjusted the MIV algorithm to give greater weight to engagement metrics such as the number of shares and comments. This decision comes from our analysis that these metrics are significant indicators of content virality and audience interaction.

Viral posts on TikTok will now see an increase in MIV by 60-70%. This adjustment ensures that individual high-engagement posts receive the recognition they deserve.

Additionally, the median MIV for TikTok's channel will remain relatively stable, maintaining a balanced view of overall channel performance.

Another notable outcome of this update is the improved comparability of top-performing TikTok posts with their Instagram counterparts. This means posts with similar performance on both platforms will now be more equally valued in terms of MIV, facilitating a more consistent cross-platform strategy.

October 2023

Objective: Improve MIV for Chinese social media platforms.

Date: Change effective from October 1st in Discover, Insights, and Data Services.

Re-computation of all data from January 1st 2023.

Changes

Click here to see the dedicated article

January 2022

Objective: Integrate content types in Social MIV formula.

Date: Change effective from January 1st  in Discover, January 10th in Insights and Data Services. No re-computation of historical data.

Changes

Social

Content types
Content types are a new qualitative attribute in the MIV formula. The objective is to differentiate the types of content published on social channels and their valuation. Indeed, depending on the channel, a video, an image or a carousel generate distinct engagement. Some content types such as videos are more incentivized by the platforms to content creators. To account for this behavior and to add more volatility in the formula, a boost/decrease is applied to the MIV of a placement depending on its content type.
ie: Two placements on the same channel with the same qualitative attributes and quantitative metrics but one is an image and the other one is a video. The video placement receives a boost in MIV compared to the image which gets a decrease.

Content types and impact in MIV formula by channel

Note: content types are not taken into account in chinese socials MIV formula

Expected impact: no impact globally or at channel level, small impact between -1% and 2% evolution at brand level.

April 2021

Objective: Make the formula more clear and simple as well as update it according to new use cases.
Boost FLB media Online and adapt to new use cases on Social: Instagram giveaways, YouTube music videos, Facebook Owned media.

Date: Change effective from April 12th in Discover, April 13th in Insights and Data Services. Re-computation of all data with updated MIV from January 1st 2021.

Changes

Online

To calculate MIV online, one of the attributes is a media score based on the relevancy of this media for your brand. Media that are FLB-focused (Vogue, Cosmopolitan, Elle) and more valuable to the brand have a high score compared to less relevant media for the brand (generalist media such as the Daily Mail).

While a weight was always applied to FLB media, we have reviewed and boosted a number of FLB medias higher than other generalists. This will allow for more differentiation between FLB and non-FLB MIV.

Expected impact: Growth of very FLB-oriented media MIV such as Vogue (+40%), and decrease in MIV for generalists media such as the Daily Mail (-10%). This will not impact overall Online MIV and global MIV for brands.

Social

Instagram

Our algorithm detects giveaways on social media which artificially raise the engagement of a post (number of likes, comments, views and shares) and applies a discount to the engagement account to calculate MIV.

To better detect these giveaway posts and provide accurate MIV we’ve updated the list of words the algorithm uses to flag them.

Expected impact: More giveaway posts detected. No significant change in Social MIV or global MIV for brands.

YouTube

Brands' MIV coming from YouTube music videos is very high when the brand is mentioned in the lyrics. The reason behind is that music videos have a higher amount of views due to the audience listening to the song and therefore watching the video repeatedly.

To adjust the MIV of these videos for brands we add a component in the algorithm to detect this content. If a brand is mentioned in a music video, the number of views taken into account in the MIV calculation is discounted.

Expected impact: Lower MIV for Youtube music videos with a high number of views. These placements should appear less often in top placements for brands. Small impact on Social and on global MIV for brands.

Facebook

Multinational brands have one global page on facebook from which they can publish content targeted at different regions. In order to avoid counting the full global audience of the page for each local post, we implemented a change in the formula to reflect the true impact of the posts.

MIV calculation changes; the balance between engagement and audience is more tilted towards engagement so the audience doesn’t weigh too much in the formula.

Expected impact: This change will heavily lower Facebook Owned Media MIV for brands that have a global FB page with different regions (up to -60%). This also has an impact on global MIV for brands (up to -10%). However, if you reference the above, it's limited to global brands that have a Facebook page with regionally targeted posts, so while the individual post impact can be big, it's limited to a small amount of posts/clients ie. global clients with their own Facebook pages that then have regional posts.

February 2021

Objective: Fix the audience issue coming from third party Similarweb.

Date: Change effective from January 1st 2021 onwards. This fix may impact MIV values significantly compared to before.

Changes

Online

We were only getting from Similarweb the audience of root urls (ex: www.elle.com) and not specific sections of a website (ex: www.elle.com/it). This explains for example why the audience was the same for Elle - ES, - IT and - US. They all got the full audience of the website which artificially inflated their MIV.

The problem is fixed, the media now receive a relevant audience calculated based on the visits on the specific section of the website.

Expected impact: Lower audience and MIV for articles in Online Media that used to have a global audience: Cosmopolitan, Elle, Esquire, Glamour, Grazia, Harper’s Bazaar, HypeBeast, i-D, Lifestyle, Men’s Health, Prestige, Refinery29, Women’s Health.