The writing below is a segment from my research on the use of Pinterest for improved brand performance. Please feel free to contact me with any questions and references.
It is crucial to understand and clearly state segmentation methods used in a marketing strategy and campaigns. According to the Business Dictionary (2018), segmentation refers to the process of defining and subdividing a large homogenous market into clearly identifiable segments having similar needs, wants, or demand characteristics. Without clear idea of market segments and their behavior, marketers can not effectively target customers and provide them maximum value. Moreover, as presented by Wharton’s marketing guide, segmentation allows a company to grow profits, which is an important organizational goal, therefore a core theoretic rationale for its use (Wind, 2007). Accordingly, if segmentation methods are not clear, the research outcomes will be skewed, and maximum benefits from using a social media campaign cannot be derived. Consequently, the following part of the literature review examines the common methods for segmentation, modern methods, the methods that Pinterest supports and successful Pinterest profiles use. Finally, the information will be used in the research methods section of Chapter3, to identify optimal segmentation method for the Pinterest campaign and overall company’s marketing communications strategy.
Commonly Used Segmentation Methods (Overview)
According to Wind (2007), segmentation is at the core of the company that aims to be customer-driven. Wind identifies multiple methods of segmentation based on potential derived benefits/specific use:
· For positioning: product usage, product preference, benefits sought or a hybrid of the variables above.
· For new product concepts (and new product introduction): intention to buy, preference over current brand, benefits sought.
· For pricing decisions: price sensitivity, deal proneness, price sensitivity by purchase/usage patterns.
· For advertising decisions: benefits sought, media usage, psychographic/lifestyle. A hybrid of the variables above with or without purchase/usage patterns.
· For distribution decisions: store loyalty and patronage, benefits sought in store selection.
· For general understanding of a market: benefits sought, or in industrial markets, the criterion used is purchase decision, product purchase and usage patterns,
In Wind’s identified segmentation methods, there is no demographic segmentation; its absence will be reviewed more specifically in the discourse on modern segmentation methods below. For the purpose of the research and social media marketing campaigns, I will focus more specifically on the segmentation that is used for advertising, general understanding and positioning reasons. Therefore, the primary segmentation methods that are relevant for social media (Pinterest) usage include:
· Product usage
· Product preference
· Benefits sought
· Media usage
· Product purchase.
Above, the theoretically suggested segmentation methods were identified, however, from the perspective of marketing practitionership, some of the methods above are not viable. Therefore, below, the methods that are used by practitioners for segmenting the digital customer will be described.
This segmentation method identified by Brodo (2015), the co-founder of Advantexe, bypasses complicated methods of segmentation and divides customers simply by the stage and the level of involvement in a purchase decision.
He outlines three primary segments:
· High Actual Customers: People who are actively looking for a product but have not decided on the brand (suggested action-Customize offer).
· High Potential Customers: thought about buying the product, but have not decided yet (suggested action-provide additional value as a deciding factor).
· Everyone Else: Don’t want or are not looking for the product (suggested action-abandon).
Although this model seems intuitive, it is important to highlight its limitations, Brodo suggests that marketers should not bother with “Everyone Else” because the costs of reaching and converting them is too high; however, almost all companies that use “Blue Ocean” strategy, need to do some sort of market education. Innovative companies such as Apple in its early days, Uber and Airbnb, would not have a market segment if they did not target “Everyone Else.” Additionally, knowing these attributes does not guarantee customer conversion. Marketers need to know more attributes of a segment to clearly position a company to each segment and increase conversion rates.
The Most Frequently taught method
Most commonly taught marketing segmentation methods include:
· Psychographic: based on cultural, lifestyle, status, personality cluster
· Decision Maker: based on the individuals that make decisions
· Behavioral: based on product usage
· Geographic: based on the physical location of a user
· Distribution: segmenting distribution channels.
· Demographic: age, gender, interests etc.
Although these are the most common segmentation methods, which are familiar to marketers, most of the practitioners and experts, such as Yankelovich and Gary Vaynerchuk recognize their limitations. These methods of segmentation are quite myopic as they focus on the identification and grouping based on customer characteristics, but miss the fact that given characteristics in Glocal economy might not be sufficient to determine or influence a customer’s decision. This misguided practice was revealed by multiple psychologists, Kirkman (2016), writes that people often do not realize that occupation and socio-economic status are far better predictors of cultural similarities then either geographic or demographic indicators. Simply put, scientist in India might be more similar to a younger scientist in Egypt with similar socio-economic status, rather than a businessman of the same age in India. Although useful, marketers should avoid myopic reliance on common segmentation methods and pay attention to the hidden attributes of their customers. Another good example of limitations of traditional market segmentation comes from Psychographic segmentation.
Psychographic segmentation is heavily based on the research by Stanford scientists Arnold Mitchel, who launched Values and Lifestyle (VALS) program in 1978 (Yankelovich & Meer, 2006). VALS was adapted by advertisers and marketers who assumed that people’s resonance to the VALS was the primary explanation for their decision-making. It brought psychographic segmentation to the masses and popularized the widespread usage. Yankelovich suggests that actual purchasing decision is much more influenced by other attributes such as, purchasing history, loyalty, propensity to trade up etc.
Modern Segmentation Methods (alternatives)
In Harvard Business Review article, Yankelovich & Meer (2006) state that physiographical profiling that is often called segmentation is a wasteful diversion that does not carry the original weight and importance of segmentation, which is discovering customers whose behavior can be changed and the needs can be satisfied. Yankelovich continues to argue that demographic traits such as age, sex, education and income level are no longer enough for useful segmentation and proposes that non-demographic traits such as values, preferences and tastes are more likely to influence consumers’ behavior. Moreover, he argues that even the psychographic characteristics in their narrow sense (lifestyle, attitudes, self-image and aspirations), should not be enough for marketers, because they lack a predictive precision. Although it might be useful in understanding what category of product a customer is likely to purchase, it does not provide enough information to assume which product within a category an individual might purchase.
Google has become a market leader in understanding targeting and market segmentation of an online customer. According to Google (2018), Individualized marketing campaigns that are focused on signals and intent targeting have 20% higher brand recall and 50% higher brand awareness (measured by brand Lift). These statistics suggest that approach used in the Simple Segmentation Method is actually quite useful. Intent of sales and search signals would clearly make up a “high actual customer” segment of the simple model.
When talking specifically about digital segmentation and Google Analytics, marketers have information regarding five distinct dimensions of a customer (Han, 2017):
· Demographics- age, gender, interests.
· Location- access point to the website, city and continent.
· Behavior- How much users consume, repeated visits, engagement and session length.
· Devices- What devices are used to access website.
· Channels- Source of the traffic, direct, referral, social media, organic.
Although, browser extensions are needed, the users can be segmented through these characteristics in Google Analytics into different clusters that define a persona. Afterwards, a marketer can analyze what type of user persona is more likely to make a purchase and is more valuable to be targeted. It is important to note, that unlike traditional marketing where similar elements for segmentation might have been used, digital marketing and Google Analytics allows the formulation of a targeted customer persona through the empirical research, whereas in a non-digital system, the creation and testing of the numerous clusters based on multiple characteristics would not be cost effective.
Joel Rubinson (2013), the president of the Rubinson Partners proposes four alternative approaches for segmentation for a digital customer:
· Moments: A segment with a specific intent or stage in life can be isolated and targeted. A simple example: data regarding a customer’s online behavior can suggest that they are pregnant, which means that they are at the stage that is more susceptible for crib, or baby-toy advertisement.
· Brand Loyalty: A segment that is “switchable” (more likely to switch between brands) is identified and personalized offer is catered, as they are also more responsive to the advertising.
· People as Shoppers: This segmentation method is more relevant to the retailers; however, it suggests that knowing how people shop for their products (search intensity, decision channel, relative goods), can help stores and companies structure their offering in a manner that is more persuasive. Ex: alignment of the items in the physical store to customers internet search patterns so that they are more likely to purchase a product in their cognitive search basket.
· Targetable interests and values: The behavioral segments that can be created through user information clustering.
Combined, above stated approaches to segmentation, provide a crucial targeting opportunity for digital marketers, to reach their customers and efficiently identify segments that carry the most value.
Segmentation and Targeting Methods on Pinterest.
Pinterest is a social media and search platform that utilizes intent segmentation by default. As Hutchinson (2018) noted, 55% of the users browse Pinterest in order to or with an intent to buy a product. This means that marketers can simplify their campaign targeting and focus on users that are searching for a product or a category. Pinterest has useful tools in order to target the audience. Pinterest as a platform (without using third party extensions), utilizes five primary characteristics for segmentation and targeting (Pinterest, 2018).
· Interests- Pinterest has around 400 preset customer interests, therefore, a marketer can choose to target groups with specific interests.
· Keywords- Pinterest is as much of a search engine as it is a social media platform, therefore it allows marketers to target people who search for specific keywords.
· Personas- Targeting people by 40 interests and 4 life-stages.
· Demographics- targeting by location, language and gender.
· Devices- Targets based on web, mobile web, iPhone, iPad, Android phone and Android tablet.
Additionally, Pinterest has the capacity for:
· Retargeting- allows Pinterest to target people who have already interacted with a company’s content through liking, repining, commenting etc.
· Customer list targeting- Allows a company to import a customers list, and then use the list to target these specific customers or avoid targeting them. This is usually done through importing mailing lists, however to successfully use this targeting, provided mail should match Pinterest’s registrar. The match rates on Pinterest are 30-50% (Pinterest, 2018).
· Act Alike – Allows a company to target people who are similar to their current best campaign respondents.
It is important to note that Pinterest utilizes three types of costs for targeted advertising/promotion: CPM (cost per-impression), CPE (cost per-engagement), CPC (cost per-click).
A clear illustration of the importance of correct targeting and of using Pinterest tools is Elle Germany. Through a simple integrated marketing campaign, with the primary drive on promoting Pinterest, the company reached 4000% more click throughs, 1000% more saves and 3 times more saved pins post-campaign (Pinterest, 2018). The steps that Elle Germany took were quite straightforward, they added Pinterest save button to their mobile and desktop websites, promoted their Pinterest profile on their website, blogs, Facebook and Twitter and maybe most importantly, adjusted and targeted their content to the customer segment with the highest engagement. Clear understanding of customer needs through the analytical tools allowed Elle Germany to properly segment and cater to their desired market. As the importance of segmentation of targeting, as well as multiple methods used in the industry have been presented in this part of the literature review, the next part will focus on another important factor for successful Pinterest performance- the content strategy.
 Each of the segmentation methods will be explained in more detail once the method of choosing segments for Pinterest campaign is clearly defined.