Kaplan and Haenlein (2010, p. 61) define social media as "a group of Internet-based applications that build on the ideological and technological foundations of Web 2.0, and that allow the creation and exchange of User Generated Content". The definition is old, but its emphasis still matters: the defining feature of social media is that users, not brands, make most of the content and most of the meaning. A brand is a guest in other people's conversations, and a strategy that treats social platforms as free broadcast channels misunderstands the medium from the start.

Kietzmann et al. (2011) give managers a more practical lens. Their honeycomb describes seven functional building blocks: identity (how much users reveal about themselves), conversations, sharing, presence (knowing whether others are available), relationships, reputation and groups. Platforms emphasise different blocks, and so do audiences. The authors' point is not that a brand should try to excel at all seven, but that it should work out which functions matter to its audience on a given platform and build its activity around those, while monitoring the others.

For Moorfield Botanics, botanical skincare for sensitive and reactive skin, the strongest blocks are conversations, sharing and reputation: people with reactive skin ask each other what worked, share routines and judge products by other users' experience. Presence and identity matter far less. That analysis already narrows the job social has to do: support peer conversation and earn reputation, rather than simply display products.

A social media strategy, then, is a set of choices: which audiences, on which platforms, in what role, towards which measurable objectives, in service of business goals. It is equally a statement of limits. Social can build awareness, community and consideration, and it can drive sales, but a brand controls neither the algorithm nor the conversation; it cannot make people care about a category they don't care about, and it cannot fix a weak product. Felix, Rauschnabel and Hinsch (2017) treat social media marketing as a cross-functional activity that uses social media to achieve organisational goals by creating value for stakeholders, which makes it a matter of strategy rather than a fragmented set of tactics. The rest of this module builds that consistency, one decision at a time.

Platforms are not interchangeable pipes of different sizes. Each has affordances, the actions its design makes easy or hard (resharing, stitching, saving, commenting, searching), and each decides distribution by its own logic. Those differences shape what content spreads, how audiences behave and how much a brand's follower count is worth.

The most important shift of the last decade is from the social graph, in which you mainly saw posts from accounts you followed, to recommendation-led feeds that rank content by predicted interest. TikTok made this explicit: its For You feed ranks videos on user interactions, video information and device and account settings, and TikTok states that "neither follower count nor whether the account has had previous high-performing videos are direct factors" in recommendation (TikTok, 2020). Instagram followed. Mosseri (2023) explains that Instagram has no single algorithm: Feed, Stories, Explore and Reels each have their own ranking, with Stories weighted towards close connections and Reels and Explore built largely on recommendations from accounts the user does not follow. YouTube and Pinterest behave partly as search engines, so content can be found long after it is posted; LinkedIn's audience is there in a professional role, which suits corporate, recruitment and B2B communication more than consumer skincare.

The implication is double-edged. Reach depends less on audience size and more on how each piece of content performs, so a small brand can reach strangers. But followers are a less reliable asset, and organic reach becomes something earned post by post rather than owned. Much sharing also happens out of sight in messaging apps and email, what Madrigal (2012) called "dark social", so public engagement understates how much a brand is passed on.

For Moorfield, this explains a pattern Amara expects to find: the 84,000 Instagram followers do not guarantee that a post is seen, while a TikTok from the 41,000-follower account can reach people who have never heard of the brand.

Platform audiences and features change quickly. Any usage figure you use should be cited and dated; Ofcom's annual Online Nation report (Ofcom, 2025) is the authoritative UK source, and its December 2025 edition is the latest at the time of writing. Where you can't cite a figure, describe direction and relative differences instead.

A platform portfolio is the set of platforms a brand chooses to be active on, with a distinct role for each. The decision is a trade-off, because every platform has to be resourced properly to work: content made natively for its formats, someone to reply to comments, budget to amplify what performs. In a recommendation-led feed, a thinly run account is not a cheap presence; it is a stream of under-performing posts. Fewer platforms done well usually beats many done thinly.

A weighted scorecard makes the trade-off explicit. Five criteria cover most decisions. Audience presence asks whether the people the brand needs to reach use the platform, in that role, often enough. Objective fit asks whether the platform's logic serves the objective: recommendation feeds suit discovery, search-led platforms suit education and consideration, close-connection formats suit community. Content capability asks whether the team can produce what the platform rewards. Cost to resource, scored so that a higher score means more affordable, asks what it would take to do the platform properly. Competitive opportunity asks whether there is room to stand out. Weights reflect the brand's priorities; changing them and seeing whether the ranking holds is part of the analysis.

The scores rank platforms; they do not make the decision. The manager then assigns roles. A primary platform gets sustained investment and is expected to deliver against a main objective. A secondary platform has a narrower job, such as search and education, with lighter, lower-frequency investment. A limited-role platform serves a specific non-marketing purpose, such as corporate communication or recruitment, and is resourced accordingly. Dropping a platform is a legitimate outcome, provided the account is handled responsibly rather than left to decay.

Resource is where most portfolio decisions are really made. Moorfield Botanics has 38 employees, and its 2027 digital marketing budget of £720,000 allocates £110,000 to content production and £260,000 to paid social. Those sums have to cover every platform the brand keeps. Spread across four platforms equally, they would buy four mediocre presences; concentrated, they can buy two strong ones and a modest third. Felix, Rauschnabel and Hinsch (2017) treat scope as one of four strategic choices in social media marketing, ranging from defenders, who use social mainly to broadcast to customers, to explorers, who collaborate with many stakeholders. How widely to spread across platforms is a related choice, and it should follow from strategy and resources, not from what competitors happen to be doing.

Objectives connect social activity to the reasons the business invests in it. A useful structure is a cascade: business goals (revenue, customers, margin), marketing objectives that serve them (awareness among a target segment, acquisition, retention), and social objectives that state what each platform contributes. Every social objective should be traceable up the cascade. If it can't be, it is an activity measure, not an objective.

Doran (1981) gave managers the SMART test: objectives should be specific, measurable, assignable, realistic and time-related. The letters are now often read as achievable and relevant, but the discipline is the same. "Grow engagement on TikTok" fails; "increase the number of new customers who first discover Moorfield through TikTok, measured by post-purchase survey, by December 2027" passes most of the test, once a baseline is measured and a target is set against it.

The second distinction is between brand-building and activation. Binet and Field (2013), analysing campaigns in the IPA Effectiveness Databank, found that long-term brand-building and short-term sales activation work differently and that the most effective campaigns balanced the two, with roughly 60% of investment on brand-building. Treat this as a guideline from their analysis, not a law: their later work showed the right balance varies by category, brand size and context (Binet and Field, 2018), and their evidence is about advertising budgets in general rather than organic social specifically. The lesson for social is that reach-based, memorable brand activity and conversion-focused activity need different objectives and different measures, and a plan made only of activation is likely to under-invest in future demand.

That links to reach versus engagement. Engagement rates are easy to report but can rise while reach falls, and a small, highly engaged audience may be people who would have bought anyway. Follower counts, likes and impressions are vanity metrics when reported alone, because they do not show whether anything changed for the business.

For Moorfield, the business goals for 2027 are explicit: revenue from £6.2M to £7.5M (+21%) and new customers from 48,000 to 60,000 (+25%). Any social objective Amara sets has to show how Instagram and TikTok contribute to those numbers. The social objectives set here are the ones Module 7 later tests against evidence.