Knowing what to say and where to play

As with the interpretation of any dataset, context is an absolute must-have and it is even more important with social media. We have previously highlighted this as an essential best practice in strategic social media listening.

In this post we share an example of how strategic social listening can power informed decisions and inspire actions for consumer insights, R&D and marketing teams.

We investigated a hypothesis from Unilever’s former R&D Director of Hair Care Europe, Inge Terpstra,  that strategic social listening could be used to:

  • Segment consumers
  • Identify habits, trends and which of the key segments were driving these
  • Understand consumer language
  • Know what to say and where to play

Knowing what to measure

Setting clear research objectives, knowing what to measure and in what context is critical to deliver actionable insights. This was clearly echoed by Forrester in its report from August 2013 on the Roles and Responsibilities for Social Intelligence which stated that “finding and acting on insights gleaned from online and social discussions requires the right combination of a clear objective, business planning, technology and, most importantly, people to do the work.”

Unilever had been implementing natural formulations only to find out that consumers didn’t always recognise that its products were more natural than others’ that make “naturally inspired claims.” This prompted a need to understand whether the presence of natural ingredients in formulations was a consideration for consumers but also to understand consumers’ needs better to ensure that future formulations would meet those needs.

With the prior understanding that different segments may have different needs, we decided to focus on women with frizzy hair.

Our key objective was to identify trends, habits and the language used by this group, with a view to highlighting new ways of messaging consumers.

  • We aimed to answer the following key questions:
  • Are natural oils a hot trend amongst the fizzy hair segment?
  • What other trends / habits are emerging from the frizzy hair segment?
  • Which brands are mentioned in connection with natural/organic or natural oils?
  • Are “trend setters” and “influential individuals” advocating natural oils and natural and organic products?

Knowing how to measure

Knowing how to measure requires a combination of tools and people with the right expertise. The research objectives should help determine the methodology and tools used so that the approach is robust, repeatable and understandable, and, the insights actionable. This was again echoed by Forrester in its Evaluate and Evolve Social Intelligence report published in May 2015: “Listening to data gleaned from social media channels, ratings and reviews, and forums has existed for a decade, yet most CI (customer insights) pros still cannot demonstrate the value of monitoring social data.”

Using social media as a source of insight, we applied qualitative research techniques at scale on conversations among women with frizzy hair.

We gathered content across all social media using intelligent search queries combining known hair care consumer language focusing on natural or organic solutions. We then refined our search to focus on the frizzy hair segment.

Within the frizzy hair data set we qualitatively evaluated a sample of mentions for themes, and when present, views on products available.

We also identified influential consumers based on our proprietary approach which looks at Reach, Relevance, Resonance, and Activity.

Contextual insight

Our research highlighted some key insights:

  • Natural oils are a hot trend among consumers who identified themselves as frizzy-haired. This appears to be driven by a belief that ‘chemicals’ cause frizz and natural products are more effective. The focus was invariably on whether the product/solution actually works to eliminate frizz.
  • Women with frizzy hair are prepared to pay a premium for a solution that works suggesting that effectiveness trumps price but while efficacy is the most important consideration, price does come into the decision-making process as a price premium is synonymous with quality.
  • Women with frizzy hair describe or recommend an entire hair care routine, incorporating the use of products but also washing, styling, and drying techniques that are believed to reduce and manage frizz. The CG method or Curly Girl method was prominently discussed as a potential solution as were glycerin-free gels and wet gels.
  • Other routines were also discussed such as ‘No-Poo’ (infrequent or no shampoo use) and ‘Max Hydration’ (a 5 step regimen that systematically increases moisture levels).
  • Women with frizzy hair appear very knowledgeable on potential solutions and this is evidenced by the use of acronyms in their discussions (e.g. FSG: Flaxseed gel or DT’s: Deep Treatments). Humidity not surprisingly was mentioned frequently as a problem. Terms such as Hair Canopy (the mass of hair in the centre) or “taming medusa hair” were used to describe their problems. John Frieda’s Frizz-ease range and salon keratin treatments were both seen as the most effective products / solutions for controlling frizz.

Actions: Know what to say and where to play

There are multiple ways these insights could be applied.

Brands could advocate popular hair care routines (CG, No-Poo or Max Hydration) into content marketing aligning brands around consumer routines. Brands could also consider developing a range of products to support these routines.

Brands could mimic the way some dedicated social networking sites drive engagement by getting site visitors to take a quiz to understand their hair type on the landing page, directing users to the right product and routine for their hair.

Brands could use the insights to inform Adwords and PPC campaigns. Insight can also be used to shape straplines to appeal to consumers with frizzy hair using recognised consumer language e.g. canopy hair, reinforce the fact that treatments are proven.

The time and place or potential activations is also key:

  • Time marketing to run prior to when humidity levels rise in markets (late spring and during summer).
  • Focus campaigns on popular YouTube vlogger videos & forums. Mumsnet and Naturallycurly display banner adverts would be an ideal vehicle.   

Validate and quantify strategic social listening insights with other research

Strategic social listening is very effective at an early stage of the planning and research funnel. Insights found should therefore be further validated through both qualitative and quantitative message-testing techniques.

We chose to validate our recommendations with natural language processing from Relative Insight who used a linguistic analysis tool (originally designed for crime-fighting) to validate our findings. The tool analysed 1.5 million mentions about fizzy and non-frizzy hair going back 10 years form the key sources identified through social listening ( and More text than in the bible!

Our findings corroborated in that willingness to pay a premium for something that works was clearly of importance.

The consumer language identified in our qualitative social research was also found in the wider linguistic analysis. The needs and thoughts were the same amongst women with frizzy hair on the main hair care communities, however there were differences in the language used on (UK forum) and (USA forum).

Amongst people that do not have frizzy hair, the hair style was mentioned most frequently. The conversations were less about finding solutions or managing one’s hair, instead focusing on one’s cut and hair length. Once the hair reaches a certain length it can become unmanageable, and consumers with non-frizzy hair discussed ways to manage hair growth.

Prior to conducting strategic social listening it is vital to have a clear set of research objectives and an idea of how the research will be used. It is also very important to have the right research expertise to derive insight and provide recommendations to enable you to reach your objectives. As Anand Rao, Innovation Lead for PwC’s US Data and Analytics practice said, “the benefits of social listening come when you feed the insights gained back into your organisation, when and where they’re needed. In this way, social listening contributes to your broader big data mission to gain intelligence in the moment.”

Search article

Find out more

Gaelle Bertrand
Head of Social Media Intelligence

Call me +44 (0)20 7264 6307