Beauty Brains – Predicting Beauty Reviews

Project Description

Every brand I have talked to has said their most important marketing activity is reaching out to beauty reviewers (including bloggers, mainstream publications, vloggers and social media channels). It’s very important to them that they reach out to the right reviewers, and send them relevant products to review.

Brands currently struggle to understand who are the most relevant reviewers to engage with for each product, and they get bombarded with new blogs and websites who say they can review their products and promote it to their audiences. The brands do not have the resources or tools to accurately assess how influential a review from a particular website will be and who is the best person to review a product.

The beauty industry has a vast amount of new content being published each day that could be crawled and analysed to predict which reviewers are likely to write about specific products types, specific beauty concerns and specific brands.

By analysing content can we work out the following?

• Who is likely to write about a specific product type?

• Who is likely to write about specific beauty concerns?

• Who is likely to write about a specific brand?
• What beauty products types and brands will be popular in the next X months/years?
• How much influence does each review have over how successful a product will sell?

If we can work out these key questions we can then do the following:

• We can start to build an algorithm to work out the best content to serve ads on for specific brand campaigns. This is effectively the beginning of the Beauty Brains ad network.
• We can provide brands with insights into who the best reviewers are to send products to (even more brains!)….we can even proactively send products to reviewers and track how much ROI they generate as part of the Beauty Brains platform.
• We can launch this as a data product very quickly which we can use to attract brand’s interest into Beauty Brains before the launch of the main platform.
• We can provide review websites with new value propositions:
  • we can help them receive more relevant products to review
  • we can provide them with our information about their website’s content based on our analysis
  • by implementing our tag we can also provide them with information about the beauty concerns, products and brand preferences of their audiences.
  • we can help them increase the revenue they generate from selling more targeted advertising space (to us!)

Questions?

• How do you suggest we go about this project?
• Do we need to define some sort of methodology for selecting reviewers (as well as identifying new reviewers on an on-going basis)?
• Do we need to define beauty taxonomy, or will this emerge as we crawl sites and identify trends in terminology?
• How do you go about crawling the content of these websites?
• How do we assess how influential or authentic each site is (e.g. social media likes/followers? inbound links?)
• What do you think can be delivered for the initial $5,000?
• How long would you anticipate a project like this to take?