Businesses have more social media data than ever, but are they using it strategically?
Businesses increasingly track social media conversations, audience behaviour and campaign performance. Yet collecting information is not the same as allowing that information to shape important marketing decisions.
A new peer-reviewed study proposes a structured way to measure this gap. Published in Management & Marketing on 5 October 2026, the research develops a 42-item instrument for assessing how deeply social media data analytics is embedded in a firm’s marketing strategy.
The framework examines five areas: targeting and positioning; products, services and brands; communication and influence; channels and logistics; and pricing.
The authors, Teissir Benslama and Rim Jallouli, argue that evaluating analytics in only one of these areas provides an incomplete picture of a business’s strategic maturity.
The study does not prove that analytics increases revenue or profitability. Instead, it provides a measurement tool that researchers and managers can use to investigate how analytical insights are translated into decisions.
Why collecting data is not enough
A company might have sophisticated social listening software, campaign dashboards and a team of analysts, but still make major product or pricing decisions without consulting those insights.
Another company might have fewer tools yet routinely use social media evidence when choosing customer segments, designing products and refining distribution channels.
These businesses differ in a way that software expenditure alone cannot capture.
The researchers call this difference social media data analytics-enabled marketing strategy, or SEMS.
The concept describes the extent to which social media analytics informs and supports marketing strategy across its principal domains.
It focuses on what an organisation actually does with analytical information rather than merely whether it owns the technology.
This distinction matters because businesses can invest in data infrastructure without developing the processes and decision-making habits needed to use it.
The framework therefore shifts attention from analytics adoption to strategic integration.
How the researchers built the framework
The researchers began with a systematic review of 120 academic articles examining social media analytics and marketing strategy.
They also drew on expert interviews and established marketing research to identify activities that could represent the construct.
An initial pool of 62 potential survey items was refined through expert review, producing a 47-item questionnaire.
A pretest involving 30 marketing and analytics professionals assessed clarity and questionnaire flow.
Additional exploratory analysis reduced the instrument to 44 items before the main study.
The researchers then distributed the survey to 600 professionals, receiving 317 responses.
Of these respondents, 132 reported that social media analytics was applied in their firm’s marketing strategy.
After removing responses with missing information and outliers, the final analytical sample consisted of 123 participants.
Further factor analysis eliminated two items, leaving a 42-item instrument across five dimensions.
This multi-stage approach was intended to establish whether the questionnaire measured a coherent concept rather than an arbitrary collection of marketing activities.
Five areas where analytics can shape marketing
The first dimension concerns targeting and positioning, including how businesses identify customer groups and decide what distinguishes their offerings.
Social media data can reveal emerging interests, customer concerns and changes in how audiences discuss competing products.
The second dimension covers products, services and brands. Here, analytics may inform decisions about product development, brand associations and customer expectations.
The third dimension examines communication and influence, such as campaign planning, message development and engagement with relevant audiences.
The fourth concerns channels and logistics, reflecting decisions about how offerings reach customers and how distribution choices relate to market information.
The fifth dimension addresses pricing, where social signals may provide supplementary evidence about consumer preferences, competitor activity and perceived value.
Importantly, the study does not say that social media should determine every decision in these domains.
Instead, it assesses the extent to which analytics contributes useful information to each part of a firm’s existing strategy.
Taken together, the five dimensions offer a broader view than a dashboard focused only on campaign clicks, reach or engagement.
What the statistical testing found
The researchers evaluated the instrument using exploratory factor analysis and partial least squares structural equation modelling.
The factor analysis reproduced the proposed five-dimension structure, with 14 items for targeting and positioning, nine for communication and influence, eight for products, services and brands, five for channels and logistics, and six for pricing.
The five factors accounted for 74.523% of the variance in the analysis.
The reported Kaiser-Meyer-Olkin statistic was 0.902, indicating that the correlation structure was suitable for factor analysis.
Subsequent tests supported internal consistency, convergent validity and discriminant validity under the study’s selected criteria.
In simpler terms, the survey items generally behaved as expected for a measurement instrument with five related but distinguishable components.
These findings support the proposed scale within the analysed sample, although they do not guarantee that it will perform identically in every country or industry.
That is why independent replication remains important.
The difference between analytics activity and strategic use
The researchers also tested whether their new measure was related to the broader practice of social media data analytics.
That relationship was positive and statistically significant, with a reported t-statistic of 2.57 and p-value below 0.01.
The model reported an R-squared value of 0.415, indicating that the analytics construct explained 41.5% of the observed variance in the new strategic-integration measure.
This is a statistical result about relationships between survey constructs, not a claim that analytics improves business performance by 41.5%.
The study also found that analytics activity and analytics-enabled strategy could be distinguished empirically.
That distinction supports the central argument that possessing analytical capabilities and integrating their outputs into marketing strategy are not identical.
A business may collect large volumes of data while using only a small portion of that information in important decisions.
Managers therefore need to examine the organisational processes that connect analysis to strategy, rather than evaluating technology adoption alone.
Why this matters for marketing budgets
Marketing departments often justify analytical investments through metrics such as impressions, conversions and campaign return on investment.
These indicators can be useful, but they usually measure specific activities rather than the full strategic role of analytics.
A firm might improve the efficiency of its advertising while leaving its product strategy or customer positioning largely unchanged.
The new framework offers a way to identify such differences across marketing functions.
For example, a business could use the five dimensions to investigate whether social media insights influence customer segmentation but rarely reach pricing or distribution discussions.
That finding could prompt an internal review of reporting processes, decision rights and cross-functional collaboration.
However, the scale should not be treated as a proven formula for allocating marketing budgets.
The research does not calculate how much additional profit a company would earn by improving its score.
It is better understood as a diagnostic starting point for examining how information flows into decisions.
What managers could do with the findings
A practical first step is to identify the major marketing decisions made across the five domains.
Managers can then ask what evidence is considered when those decisions are made and whether relevant social media analysis reaches the responsible teams.
Businesses could compare departments to identify areas where analytics is used extensively and areas where its contribution is limited.
Such an assessment may reveal that a marketing analytics team produces reports that rarely influence senior management decisions.
It may also show that insights are used in campaign planning but not in longer-term brand or product discussions.
The framework could support more structured conversations about training, reporting and the integration of analytical findings into business planning.
Any organisational changes should be evaluated against outcomes that matter to the business, including customer experience, efficiency and financial performance.
The study provides a measurement instrument, not direct evidence that a particular managerial intervention will work.
Important limitations in the sample
The final analysis included 123 professionals, which is a modest sample for making broad claims about businesses worldwide.
Most participating firms were small and medium-sized enterprises, representing 63.5% of the sample.
The geographical distribution was uneven: 61.5% of respondents were based in Tunisia, 18.0% in France and 7.6% in Mexico.
The technology sector was particularly prominent, accounting for 37.8% of respondents.
These characteristics matter because organisational structures, digital maturity and marketing practices differ across industries and countries.
The study also relied on self-reported assessments from professionals rather than independently observed marketing decisions.
Respondents may overestimate or underestimate the extent to which analytics supports their organisation’s strategy.
The researchers tested for common method bias, but survey-based evidence cannot eliminate every source of measurement error.
Consequently, the findings should be considered preliminary validation rather than definitive proof of universal applicability.
The missing question: does it improve performance?
The most important unanswered question is whether higher strategic integration of social media analytics actually improves business results.
The authors explicitly identify the relationship between their new construct and marketing performance as an area for future research.
The current study tested the measurement model and its relationship with analytics practices, but it did not demonstrate a causal effect on sales, profitability or competitive advantage.
That limitation is essential when interpreting the findings.
A statistically reliable measurement tool can help researchers study a business practice without establishing that the practice produces financial benefits.
Future studies could examine whether firms with stronger integration achieve better customer retention, faster strategic responses or more effective marketing expenditure.
Longitudinal research would be particularly useful for testing whether improvements in strategic integration precede changes in performance.
Comparisons across countries and sectors could also establish whether the instrument operates consistently in different business environments.
Why the study is relevant to South African businesses
South African firms increasingly operate across digital platforms where customers share opinions, compare offerings and interact with brands.
The study offers a framework that local businesses could consider when evaluating how these signals inform their marketing activities.
For banks, retailers and service providers, the relevant question is not simply whether social media is monitored, but whether findings influence broader strategic decisions.
Nevertheless, the study did not validate its scale specifically among South African firms.
Local application would require testing its reliability and relevance across different industries, business sizes and languages.
Privacy and responsible data use are also important considerations when analysing customer-generated information.
Social media signals may not represent the views of an entire customer base, particularly where digital access and platform use differ between groups.
Businesses should therefore combine such information with customer research, transactional evidence and other appropriate data sources.
The framework is most useful as one component of a wider evidence-based approach to strategy.
A broader lesson for data-driven business
The research challenges a common assumption that greater access to data automatically makes an organisation more data-driven.
According to the authors’ framework, meaningful integration requires analytical evidence to influence several connected areas of marketing strategy.
The final 42-item scale provides a structured way to assess that integration.
Its five dimensions also highlight the organisational breadth of the challenge, extending beyond advertising to products, channels, pricing and customer positioning.
The study’s validation results are encouraging, but its modest and geographically concentrated sample calls for further testing.
Most importantly, the research does not establish that a high score guarantees stronger profits.
It offers businesses and researchers a more precise question to investigate: how much of the information generated by social media analytics actually reaches the decisions that shape a firm’s competitive position?
That question may prove more useful than simply counting dashboards, software subscriptions or the volume of customer data collected.
Source Information
Study: Social media data analytics-enabled marketing strategy: conceptualization, development, and validation of a multidimensional scale.
Authors: Teissir Benslama and Rim Jallouli.
Journal: Management & Marketing, volume 21, article 26.
Published: 5 October 2026.
Methods: Systematic review of 120 studies, expert input, survey development, exploratory factor analysis and partial least squares structural equation modelling using 123 retained responses.
