The Economics of Visibility: Algorithmic Power and Women-Led Small Businesses on Instagram in Pakistan

Abstract

This paper examines the intersection of algorithmic design and economic outcomes for women

entrepreneurs in Pakistan operating through Instagram contrary to the magnitude of studies

available for solely the contribution of small businesses to Pakistan's economy. Drawing on

behavioral economics and digital platform theory, it investigates how Instagram’s content

recommendation algorithm shapes visibility, consumer engagement, and revenue generation for

women-led small businesses. Utilizing a mixed-methods approach, combining survey data,

algorithmic trend analysis, and detailed case studies, the research reveals how platform biases

and socio-cultural dynamics interact to affect entrepreneurial trajectories. The study affirms that

80% of the adult population of Pakistan lives in areas served by mobile broadband (3G or 4G)

networks, this has thereby allowed many women to reap the benefits of e-commerce and start

selling goods through these platforms which may not be possible through traditional methods.

The study affirms that the number of entrepreneurs in each region is directly linked to the

region's development and digital literacy. These findings underscore the need for more

transparent algorithmic governance and localized platform support to promote equitable access

for underrepresented digital entrepreneurs in South Punjab. This research contributes to platform

economics by showing how visibility functions as an allocative mechanism in informal digital

markets.

5

Introduction

This paper begins with the assumption that Instagram’s algorithm produces asymmetrical

outcomes for women entrepreneurs, disproportionately affecting mom‐and‐pop enterprises,

informal, home‐based ventures that lack formal infrastructure, digital training, or access to

professional marketing networks. A large share of the women entrepreneurs profiled in this study

operate as mom‐and‐pop enterprises, small, owner‐run, home‐based ventures without formal staff

or storefronts. These informal enterprises rely entirely on Instagram for customer discovery.

Their limited resources, no dedicated marketing team, no professional analytics tools mean they

must juggle production, family duties, and constant trial‐and‐error with the algorithm to stay

visible.

These small owners‐run businesses ranging from custom cake shops in Lahore to handcrafted

crochet operations in Multan must not only craft compelling products but also master an opaque

set of platform rules in order to be seen. Without transparent analytics or targeted support, many

women find their hard work buried in the feed, turning what could be a gateway to economic

empowerment into yet another barrier. Social media platforms have revolutionized marketing

opportunities for entrepreneurs in developing economies. In Pakistan, a growing number of

women-led small businesses are using Instagram to showcase products and reach customers. Yet

women entrepreneurs remain underrepresented: they account for only about 14.4% of business

owners despite comprising nearly half of the population (USAID, 2022). At the same time,

Instagram’s user base in Pakistan is sizable, around 17.7 million users in early 2024, with over

one-third female (Napoleon Cat, 2024). In this setting, Instagram’s algorithm, which determines

which posts users see, can have significant economic impact. Recent reports suggest that

frequent algorithmic changes for example, favoring video content have reduced the organic reach

and sales of small firms (York, 2022) . Moreover, a survey of social media creators found that

88% expect future algorithm shifts to strongly affect their businesses (Business Wire, 2024).

These observations indicate that algorithmic visibility is a critical factor for online marketing

success.

This paper asks: How does Instagram’s algorithm affect the visibility and economic outcomes of

women-led small businesses in Pakistan? Our objective is to analyze the relationship between

algorithmic content ranking and measures of firm performance such as engagement, customer

reach, or sales for women-owned enterprises. We investigate whether the platform’s feed-sorting

systematically advantages or disadvantages these firms, and identifies the economic mechanisms

at work. By addressing this question, the study aims to shed light on how digital platform

6

dynamics influence market access for an economically marginalized group.

We frame the analysis within relevant economic theory. Instagram functions as a two-sided

platform, matching content suppliers with consumers. Platforms like this promise suppliers that

if they offer relevant or popular content, they will be rewarded with user attention (Cambridge

University Press, 2022). In practice, algorithmic recommendation systems act as an “invisible

hand,” efficiently allocating finite user attention based on many data-driven signals (Cambridge

University Press, 2024). However, scholars note that these markets often become

“winner-take-most” – once a platform establishes dominance, it can extract attention and value

from the ecosystem by controlling visibility (Cambridge University Press, 2024). In economic

terms, user attention is a scarce factor of production that can create rents and reshape value

allocation among competing firms. In addition, algorithmic feedback loops can reflect behavioral

patterns for example, favoring highly engaging or visually striking posts, linking our topic to

behavioral economics of attention. By applying these frameworks, the study connects the

specific case of Instagram to broader questions of platform power and market efficiency.

Studying Instagram’s algorithmic impact on women entrepreneurs is academically and socially

significant. It contributes to the literature on platform economics and media markets, especially

by focusing on gender and entrepreneurship in a developing-country context. It also has policy

relevance: understanding these dynamics can inform strategies to support women-owned SMEs

for example, through training on digital marketing or regulatory discussions about platform

transparency. In summary, this introduction highlights the Main research question and situates it

within economic theory. The following sections will review existing evidence, present our data

and methods, and analyze how algorithmic visibility on Instagram shapes outcomes for

Pakistan’s women-led small businesses.

The Rise of Digital Markets in Pakistan

Women count approximately 50% of the total population in Pakistan. However, The female labor

force participation rate, the percentage of women aged 15+ who are either working or actively

seeking work, was 24.46% in 2023, up slightly from 24.4% in 2022. According to a global

survey, women’s earnings have also been reported to be almost half those of men. Predominantly

Eastern traditions are still being followed in Pakistan where men are considered as the providers

for the family and are the sole wage earners and women as homemakers (Sadiq & Ali, 2014). In

Pakistan it is common to see women pressured to leave their jobs after marriage to serve their

families and look after the house. The Internet has opened up the doors of opportunity in many

ways. It has been a blessing for women in Pakistan as it allows them to take care of the

household whilst working and pursuing their passion.

In Pakistan, there has been exponential growth in the usage of social media since the past few

years. Pakistani users on Instagram, the most used social media, have crossed the mark of 17.7

7

million users (NapoleonCat, 2024). It has opened the doors for small scale young entrepreneurs

to open business through this platform, especially for women. They have made their profile

pages with details of their business and products. Uploading of items’ snaps, regular updates

about price cuts, new arrivals and regular answer of queries have made it a laudable business

medium. Economically, this reflects the growth of informal digital marketplaces, where low

entry barriers are coupled with high competition and algorithmic mediation. In platform

economics, such models shift market power from producers to intermediaries like Instagram,

who control access to visibility and attention.

Instagram as a Multi-Sided Market

Instagram is more than a visual content-sharing platform; it functions as a

multi-sided marketplace where users, businesses, and advertisers interact through

algorithmic mediation. Abidin and Leaver (2020) define Instagram as “a gigantic

database of images, videos, captions, comments, geolocation tags, likes, and

emojis”, but such a description only partially reflects its growing economic

significance. For small businesses–especially women-led microenterprises in

emerging markets–Instagram serves as a low-cost alternative to traditional retail,

offering built-in marketing tools such as reels, story features, and product tagging. It

enables entrepreneurs to reach customers, communicate offers, and build brand

identity without formal infrastructure. In platform economics terms, Instagram

operates as an intermediary that matches small businesses with consumers, while

simultaneously collecting data and monetizing attention through advertising.

Algorithmic Gatekeeping and Economic Power

A 2023 survey by Business Insider reported that 88% of content creators believe algorithm

changes significantly affect their income (Business Insider, 2023). Instagram’s recommendation

system functions as a powerful algorithmic gatekeeper, determining which users and businesses

receive visibility. Introduced in 2018, Instagram’s machine learning-based algorithm prioritizes

content based on engagement, relevance, timeliness, and user interaction patterns (Gillespie,

2014; Cotter, 2021). These algorithmic decisions are not neutral; they shape economic outcomes

by controlling which posts appear in users' feeds or on the Explore page. In platform economics,

this creates a bottleneck effect, where attention—a scarce and monetizable resource—is

disproportionately allocated to high-performing or trend-aligned content. For small businesses,

especially in low-resource settings like South Punjab, this means that platform fluency and

content strategy become necessary forms of digital capital. Visibility is not earned solely through

product quality but through alignment with algorithmic preferences, which remain opaque to

most users. As a result, the algorithm both empowers and excludes, making it a key mechanism

through which economic inequality can be amplified in digital markets. The visibility economy

8

on Instagram is shaped by both platform design and pre-existing inequalities, making it a crucial

area of study for understanding digital entrepreneurship. This paper contends that visibility on

Instagram is not just a product of aesthetic quality or creativity but is engineered through

algorithmic preference. As such, it acts as a determinant of market access for small-scale women

entrepreneurs, making it both a tool for empowerment and a mechanism of exclusion.

Research Objective and Economic Significance

This study begins with the assumption that algorithmic systems like Instagram’s

content-recommendation algorithm do not impact all users equally. In particular, it posits

that women-led businesses, due to structural and cultural barriers, face a unique set of

economic and digital challenges that may amplify the algorithm’s asymmetrical impact on

their visibility, reach, and revenue. The research explores this disparity by investigating

how algorithm-driven engagement affects business decisions and outcomes for women-led

enterprises.

In Pakistan, where female labor force participation remains under 25%, Instagram and

other digital platforms offer an accessible pathway to economic independence for women

excluded from formal employment (World Bank, 2023). However, the opportunity to

generate revenue through social media is increasingly shaped by opaque and data-driven

recommendation systems. This study highlights the risk of a new kind of inequality,

algorithmic exclusion, where businesses fail not due to poor product-market fit, but due to

lack of platform fluency.

The research contributes to the field of platform economics by situating visibility as a

scarce resource that must be understood, optimized, and often paid for. For policy makers,

the findings underscore the importance of digital literacy training and the need for fairer

platform practices. For development economists, this case illustrates how algorithmic

systems are reshaping microenterprise success in low and middle-income countries.

This paper therefore asks:

Main RQ. How does Instagram’s content‐recommendation algorithm asymmetrically

influence the visibility, consumer engagement, and revenue outcomes of mom‐and‐pop,

women‐led small businesses in Pakistan?

RQ 1.1. Which engagement metrics do women‐led mom‐and‐pop entrepreneurs monitor

most closely, and how do those metrics affect their decisions on product variety, pricing,

9

or posting schedules?Which engagement metrics do women entrepreneurs monitor most

closely, and how do those metrics affect decisions on product variety, pricing, or posting

schedules?

RQ 1.2. To what extent do self‐reported algorithmic understanding (platform fluency) and

the entrepreneurs’ geographic location moderate the relationship between

algorithm‐driven visibility and business performance?

The following literature review contextualises these questions within platform-economics and

behavioural-economics scholarship.

Literature Review

The Attention Economy and Digital Competition

According to the United Nations, the concept of attention economy was first coined in the late

1960s by Herbert A. Simon, characterizing the problem of information overload as an economic

one. The concept has become increasingly popular with the rise of the internet making content

increasingly abundant and immediately available, and attention becoming the limiting factor in

the consumption of information. Huge levels of data are owned by large digital platforms with

significant societal and economic impact. This has allowed these big tech companies—now the

most capitalized companies in the world—to acquire unprecedented power in the functioning of

the national and global economy. Equipped with big data and artificial intelligence, they can also

detect and eliminate nascent competitive threats, further consolidating their power and delaying

innovation and individual welfare. With their asymmetric bargaining power, digital platforms

tend to underpay content developers requiring intervention of the State as the

Australian-Facebook fight over how much content providers are compensated demonstrates. The

erosion of individuals' control of their own personal data has a profound effect on the human

psyche, influencing people’s beliefs, how they relate to the physical world and creating a sense

of information overload. The way data is currently valued in these markets creates a race to grab

individuals’ attention at the lowest possible cost. This degrades the experience leading to

maximizing the time users spend on a platform at the expense of an individual’s well-being and

even affecting individuals’ intentions.

Algorithms are programmed to increase engagement by maximizing virality of the content,

which often promotes highly “incendiary, controversial, or polarizing” content to drive

interactions. This increases exposure to unhealthy content often based on false information,

significantly impairing conscious decision-making and risks creating addiction, desensitization

and radicalization. The current attention economy model treats human attention as just another

10

factor of production; a resource to exploit. Attention can be perceived as labor: activities done on

digital platforms generate a surplus that is monetized by the platforms. From the perspective that

attention is captured and monetized through being converted into personal data, there is an

argument for treating attention as a new type of factor of production, one with economic

properties that distinguish itself from labor, capital and technology and therefore requiring

updated analytical tools. Direct contributions to the GDP have been calculated in the trillions, but

we have not yet estimated the harm it creates. The UN work-stream on measures of progress

beyond GDP must include the negative externalities of this extractive model. It is important to

understand that technologists at large, and programmers, are to be regarded as the next

generation of rights defenders. The current lack of technical guidance in the shape of, for

instance, taxonomies or readily available libraries, is a main impediment to their involvement.

We need to develop theory around the economic interests of “users” distinguished from the de

facto conditions of experiential dispossession, datafication, control, and commodification

introduced in the attention economy, and enforced by its unique and ever-widening power.

Behavioral Economics in Platform Environments

As touched on earlier, platformization defines “the penetration of infrastructures, economic

processes and governance frameworks of platforms in different societal sectors and spheres of

life (Poell, Nieborg & Van Dijck, 2019). The decision-making patterns of women-led small

businesses on Instagram, as observed in this study, strongly reflect the dynamics described by

behavioral economics, particularly in platform-mediated environments where information is

incomplete and feedback is both immediate and emotionally charged. Unlike classical economic

agents who act with perfect rationality and full information, the entrepreneurs surveyed operate

under significant cognitive and informational constraints. As such, their choices are often guided

by heuristics, mental shortcuts that allow them to simplify complex decisions, such as when to

post or which products to promote. For instance, In the Findings Chapter under Figure 4 we can

analyze that 100% of respondents reported adjusting their product offerings based on post

engagement, while 80% changed their posting times. These behaviors suggest a reliance on

visible engagement metrics like the reach as proxies for consumer demand, despite limited

understanding of how Instagram’s algorithm actually functions.

This pattern aligns with Herbert Simon’s theory of bounded rationality, which posits that when

individuals cannot process all available information or predict outcomes precisely, they

“satisfice”—choosing options that are good enough under the circumstances, rather than optimal.

The women in this study demonstrated satisficing behavior by replicating content that previously

performed well or copying strategies from more visible competitors, instead of testing new

formats or making data-driven adjustments. This highlights how platform opacity and

algorithmic unpredictability create an environment of trial-and-error learning, where decisions

11

are shaped more by anecdotal success than by economic logic. In this sense, Instagram visibility

becomes not just a technical feature, but a behavioral signal that entrepreneurs interpret

sometimes incorrectly as a guide to demand.

In addition to bounded rationality, the emotional dimensions of Instagram’s feedback loops give

rise to loss aversion—another core principle of behavioral economics derived from Prospect

Theory (Kahneman & Tversky, 1979). Multiple respondents in this study reported reducing

posting frequency or abandoning certain content strategies after experiencing low engagement.

This reaction demonstrates that negative performance i.e., a post that receives fewer likes or

views than expected is perceived as a loss and is therefore weighted more heavily in

decision-making than equivalent gains. This aligns with Prospect Theory’s assertion that losses

have a greater psychological impact than gains of equal magnitude. In platform environments,

where metrics are instantly visible and socially amplified, this emotional response can lead to

irrational underinvestment in content that may have long-term value but lacks short-term

engagement.

Furthermore, Instagram’s algorithmic structure rewards content that gains traction quickly,

encouraging entrepreneurs to prioritize visibility over value. As a result, many users in this study

reported following trends, replicating viral formats, or posting more frequently in pursuit of

algorithmic favor—actions that align with present bias, where short-term rewards are prioritized

over long-term goals. These behaviors may appear rational in the short run but often divert time

and energy away from tasks that generate sustainable income, such as refining products or

strengthening customer relationships. In economic terms, this represents a misallocation of

resources, driven not by market signals but by algorithmic nudging.

The implication of these behavioral patterns is that Instagram’s design not only influences what

users see, it actively shapes economic behavior. By rewarding high-engagement content,

penalizing underperforming posts, and offering little transparency about what determines

visibility, the platform creates a decision environment where women entrepreneurs are nudged

toward reactive, short-term strategies. This highlights the need for policy interventions and

educational programs that address not only technical skills but also the cognitive and emotional

challenges of navigating algorithmic markets. Incorporating algorithmic literacy into digital

training for women entrepreneurs—such as those offered by SMEDA or GRASP—could help

reduce heuristic-driven errors and improve decision quality. In parallel, platforms like Instagram

could redesign their analytics tools to emphasize business outcomes e.g., conversions, inquiries,

or customer retention over vanity metrics like likes and follower count. Together, these changes

could create an environment where women-led businesses are empowered to make more

deliberate, economically sound decisions rather than reacting to emotionally charged signals in a

high-pressure digital marketplace.

Marshall McLuhan (1962) introduced the theory underpinning this literature, arguing that no

medium has an isolated meaning or existence, as it is in constant interplay with other media. He

proposed that media influences societal progression and that significant periods in history can be

categorized by the technological medium utilised during that time, and how it affected

12

civilisation (McLuhan & Staines, 2004). Neil Postman then coined the term ‘media ecology’ in

1968, asserting that “the medium is a technology within which a [human] culture grows” (2016).

Postman focuses on the way media communication “affects human perception, understanding,

feeling, and value”, and how interactions with such media facilitate, or impede, our chances of

overall survival (Postman, 2016). This literature broadly compares media to an infrastructure that

connects the nature and culture of a society and studies the movement between the two

(Postman, 2016). According to Strate, media ecology scholars employ broad categories of “oral,

scribal, print, and electronic cultures... based on the notion that ‘communication, not economics’,

influences social life most significantly” (2008).

José van Dijck (2013) conceptualises the intersection among social networking

platforms, mass media, users, and institutions as ‘social media logic’—the norms,

strategies and mechanisms that underpin contemporary online dynamics. Platforms

such as Instagram are described as ‘microsystems’ that form part of a collective

media ecosystem. Each microsystem is sensitive and functions according to changes

in other components. As a result, traditional ‘media species’ like radio, film, and

television must compete with newer platforms such as Instagram and adopt

interactive features to survive (van Dijck, 2013). The concept of evolution thereby

“creates a theoretical framework for studying the history of media and suggests new

concepts and questions about media extinction, survival, and coevolution” (Scolari,

2012).

Nieborg and Poell (2018) argue that platformization “entails the replacement of

two-sided market structures with complex multi sided platform configurations,

dominated by big platform corporations”. These configurations impact the

production, distribution, and circulation of cultural content. Producers are “enticed

by new platform services and infrastructural changes,” yet must also respond to

“seemingly serendipitous changes” in platform governance, ranging from content

curation to pricing strategies. This aligns with Instagram’s stakeholder dynamics, in

which users adapt to platform updates that directly affect their visibility and

economic outcomes.

Social Media and Women Entrepreneurs

Social media is a digital environment that allows all users to mix personal and professional

information, allowing businesses to gain a new communication channel with their customers

(Gustavsson, 2017). Dollarhide (2021), on the other hand, defined social media as a digital

platform for exchanging ideas, thoughts, and information via virtual networks and communities.

In Malaysia, the most popular and trending social media platforms are TikTok, Instagram,

Twitter, YouTube, and Facebook. According to Boyd & Ellison (2007), social media was created

initially to connect friends, but has since grown into a global phenomenon since the creation of

13

Facebook in 2008. Moreover, according to Mazza & Palermo (2018), Facebook is used as a

communication medium by billions of people worldwide, so businesses should leverage it for

marketing and consumer engagement.

Women’s involvement in economic activities has become a bench- mark for economic progress

in developing countries (IMF, 2013). For the last few years, the number of women who are

self-employed has increased exponentially. Women entrepreneurship significance has increased

not because of the economic strain on the only breadwinner at home but because of the act that

women have now more access to better information channels and have enhanced desire for

self-actualization as compared to the past, which has enabled them to earn for themselves and

their families with a sense of self-worth, while also catering their social and religious

compulsions (Melissa et al., 2015).

One of the most prominent features of this new business context is flexible working, allow- ing

women to increase their share in the workplace. Social media has offered enough benefits to

women entrepreneurs to pursue their businesses successfully, including fast and cheap

information disclosure, display of the photos and videos of products without any cost, wide

networking and instant messaging (Genç & Öksüz, 2015). It is claimed that flexible working

facilitate women to take care of their children too, thus plummeting their stress and making them

more prolific and contented with

Women in developing countries like Pakistan still have to face many obstacles in planning and

starting their own business and that is the reason that women entrepreneurs’ percentage is still

low; this low percentage can be attributed to strong gender stereotypes, cultural values, norms

and limited access to capital and developmental opportunities. In rural areas, a woman’s

principal responsibility is to look after her husband and children, with no permission to start her

own businesses or to start a job that involves dealing with men. Women around the globe are

generally involved in low paying jobs requiring lesser skill. There is a huge gender disparity

regarding communal and legal protection in the country causing marginalization of women.

President Women Entrepreneurs Association of Pakistan has claimed that 95% of the women in

Pakistan are unable to fully comprehend their potential.

The Economics of Visibility

Digital platforms like Instagram function as markets for attention, where information

asymmetries and signaling distort traditional economic relations. In this context, large firms or

the platform itself hold more information about how content is ranked e.g. the proprietary

recommendation algorithm than individual entrepreneurs, creating an information asymmetry

that can lead to inefficiencies. Economist Joseph Stiglitz famously argued that information

asymmetry, situations where one party has more or better information than the other can lead to

power imbalances and even market failure when transactions become inefficient (Stiglitz, 2002).

This principle is highly relevant to platforms like Instagram, where small businesses cannot

directly observe user preferences or the algorithm’s weighting system. Instead, they rely on

14

observable cues. By posting professional photos, using branded hashtags, or maintaining high

engagement metrics, women entrepreneurs attempt to signal product quality or relevance. This

behavior reflects Spence’s signaling theory, in which a sender (the entrepreneur) uses a costly

signal (e.g, high-quality content, verified profiles) to credibly convey value to the receiver (the

consumer). In this context, likes and comments serve as demand signals, interpreted by

Instagram’s algorithm as indicators of interest and thus rewarded with visibility.

The technology also amplifies the role of bounded rationality in decision-making. As Herbert

Simon noted, real agents have limited information and computational capacity, so they cannot

perfectly optimize. Instagram’s algorithm and market dynamics are extremely complex, so

entrepreneurs use simple heuristics rather than global optimization. For example, business

owners might “satisfy” by posting at times they believe are optimal or by emulating popular

trends, rather than calculating an optimal schedule. In cognitive terms, entrepreneurs follow rules

of thumb e.g. posting daily, using trending filters because there is no transparent formula to

guarantee success. Simon’s notion of bounded rationality implies that these women make

“rational” choices only within the limits of the information they can process. In practice, a

shopkeeper in Lahore or Islamabad will iterate through trial-and-error to find what works,

trading depth of analysis for manageable decision-making.

Finally, classic economic concepts like opportunity cost and market failure help explain the

trade-offs imposed by platform dynamics. Every hour spent optimizing Instagram content has an

opportunity cost: time not spent on production, offline sales, or other tasks. In other words, the

entrepreneur must give up the next-best alternative use of that. When the platform biases

visibility for example, favoring celebrity posts or highly engaging content, the resulting

allocation of exposure may not align with true product value, resembling a market failure. As

one definition notes, a market failure occurs when “each individual makes the correct decision

for himself but those prove to be the wrong decisions for the group,” leading to an inefficient

distribution of goods and services. If algorithmic noise or asymmetry prevents efficient matches

between buyers and sellers, the Instagram marketplace departs from the ideal free market

outcome. Thus visibility on Instagram can be understood as an economic good governed by

information gaps, signaling strategies, and trade-offs in resource allocation.

Monopolistic Competition in Digital Marketplaces

Albarran (2016) argues that social media platforms exist in the structure of monopolistic

competition, where “many sellers or suppliers of products are similar, but not ideal

substitutes for one another,” leading to firms engaging in product differentiation to stand

out. In such an environment, “price is set by a combination of market forces and the firms

themselves”—a process comparable to how social media platforms tweak their algorithms

to control the way users receive visibility and maintain competitive advantage. Gomery

(2009) similarly describes mass media as economic institutions that must compete in a

15

dynamic landscape, where performance is shaped by structure and strategic conduct.

Theoretical Framework

Utility, Choice, and Algorithmic Influence

Visibility functions as a proxy for utility in this digital market, entrepreneurs treat increased

engagement as a signal of value creation, which guides future resource allocation. The

economics of visibility on Instagram is grounded in the concept of utility maximization, where

small business owners make calculated choices to optimize reach, engagement, and ultimately

revenue. Empirical evidence from a 2020 Malaysian study by Altaf Akbar confirms this

economic behavior: in a survey of 352 SMEs, the study found that Instagram competence,

cost-effectiveness, innovative behavior, and interactivity significantly influenced the use of the

platform for business purposes. More importantly, the findings indicate that Instagram use had a

strong positive effect on both financial and non-financial performance of these businesses. This

demonstrates that algorithm-driven visibility on Instagram serves as a utility-enhancing tool,

especially when entrepreneurs adapt their strategies to suit algorithmic patterns e.g., maximizing

interactivity or visual appeal. Entrepreneurs allocate time and resources to content creation and

engagement techniques because they perceive measurable returns in visibility and consumer

demand, reinforcing the platform’s role as an economic decision-making environment.

Instagram's algorithm indirectly dictates producer choice, nudging business owners toward

specific types of content or engagement strategies that align with the platform’s visibility logic.

As the algorithm rewards frequent posting, Reels, and high interaction rates, women-led

businesses often constrained by limited time and labor are forced to weigh these algorithmic

demands against their opportunity costs. This creates a decision framework not unlike classical

utility theory: producers allocate scarce resources to maximize expected benefits. However,

when the algorithm changes without transparency, it disrupts this choice mechanism. Business

owners may be unable to determine whether drops in engagement are due to poor content,

external variables, or a shift in the algorithm itself. This ambiguity limits rational economic

planning and can result in suboptimal allocation of marketing efforts, especially for

microentrepreneurs unfamiliar with the platform’s backend logic.

Furthermore, the algorithm’s influence on choice introduces inequality in utility gain,

particularly for women-led businesses in developing countries. Studies from similar contexts,

such as India and Bangladesh, suggest that women entrepreneurs are less likely to access digital

marketing training or analytics tools, meaning their ability to adapt to algorithmic shifts is

diminished. This technological barrier leads to a lower marginal return on their content compared

to more resource-rich businesses. Thus, while Instagram may offer high potential utility through

exposure and sales, the benefits are not evenly distributed. Instead, algorithmic systems act as

16

invisible gatekeepers, rewarding businesses that understand and exploit their logic while

penalizing those who do not. In this sense, algorithmic visibility not only shapes economic

choices but also reinforces structural inequalities in digital entrepreneurship.

Information Asymmetry in Platform Economics

Instagram’s algorithm operates as a black box, creating substantial information asymmetry

between the platform and its users, especially small businesses. Entrepreneurs do not know

precisely how visibility is calculated, how engagement is weighed, or when algorithm updates

occur. This prevents rational expectations and introduces inefficiencies in marketing strategies.

Women-led businesses in Pakistan are particularly vulnerable, as they are less likely to have

access to paid analytics tools, digital marketing education, or agency-managed insights. Unlike

larger firms or influencers who test and adapt through data feedback, these entrepreneurs operate

in an environment of systematic uncertainty, where identical actions may yield vastly different

results due to unseen algorithmic shifts.

This asymmetry introduces a digital version of economist George Akerlof’s classic

concept of the “market for lemons”, a theory explaining how quality uncertainty in a

market leads to adverse selection and inefficiency (Akerlof, 1970). In traditional

markets, when buyers can’t distinguish between high- and low-quality goods, they

assume average quality and offer lower prices, driving out good products. On

Instagram, however, the uncertainty lies not in product quality but in content

performance. Entrepreneurs don’t know whether a post’s poor reach is due to

content flaws, suboptimal timing, or invisible algorithmic shifts. Lacking this clarity,

they often waste time and effort adjusting strategies without meaningful results.

From an economic standpoint, this represents a form of market failure: an

unregulated digital monopoly, Instagram, withholds the rules of visibility, impairing

rational decision-making and distorting resource allocation. In response, some

entrepreneurs over-invest in superficial signals e.g., buying likes, further weakening

the link between content quality and visibility just as low-quality goods can

dominate in markets plagued by asymmetry.

Rational and Irrational Behaviours in Response to Algorithmic Changes

In theory, economic agents are rational and responsive to incentives, they make decisions in

order to maximise their utiliy. Yet in practice, algorithmic opacity and psychological pressure

trigger a mix of rational and irrational behaviors among women-led small businesses. Rational

responses include increased posting frequency, use of trending audio, and investing in ads,

actions rooted in maximizing engagement returns. However, many women entrepreneurs also

17

engage in irrational or speculative behavior: abandoning the platform altogether after a dip in

visibility, copying viral content without brand fit, or emotionally reacting to temporary drops in

followers. These actions represent a departure from rational economic behavior and instead

mirror the uncertainty-driven behaviors seen in volatile financial markets.

Behavioral economics helps explain this. Instagram’s unpredictable algorithmic changes function

like information shocks, producing overreactions and inconsistent investment in content. For

entrepreneurs without training or support, even minor dips in engagement are internalized as

personal failure or business decline. This leads to loss aversion, where the pain of a drop in

visibility outweighs the gains of experimenting with new content strategies. In Pakistan, where

women entrepreneurs often operate without mentorship or support systems, these irrational

reactions are compounded by emotional labor and social pressure. What should be a rational

economic decision becomes a psychologically taxing and inconsistent process thereby further

limiting their ability to grow through digital means.

Together, these theories suggest that algorithmic visibility is not only an output of user activity,

but also a market-shaping force. Entrepreneurs respond to feedback not just rationally, but

emotionally and heuristically creating a complex, nonlinear decision-making environment that

blurs classical market models.

Research Methodology

Qualitative Approach

According to Denzin and Lincoln, online qualitative research is a “transdisciplinary and

sometimes interdisciplinary field that crosscuts the humanities, the social sciences, and the

physical sciences” (2000).

A mixed-method approach was used for collecting both primary and secondary data to help

us understand the market decisions in greater detail. To analyse the degree of effect that the

algorithm plays on deciding the business activity of small enterprises primary research data was

collected through questionnaires. Primary data is the data that is collected through direct

interaction with the women owners of these enterprises. The methodology of how the data were

to be collected was established :

1. Sampling

2. Surveys

3. Data Collection

4. Analysing the data

18

Throughout the project, The researcher applied a reflective positionality lens, acknowledging

her own experience as a home bakery owner with 3k followers since the past 5 years. By actively

monetizing content on Instagram, insider access to platform dynamics and audience response

was provided.

Sampling

This research employed purposive sampling, initially distributing outreach to over 100 women

entrepreneurs via digital networks, online forums, and business groups. From these, a subset of

five valid and complete responses was selected for in-depth profiling based on diversity in

business type, digital engagement level, and regional context. These profiles were used to

examine varying experiences and strategies under the same platform dynamics.

This research employs a purposive sampling technique, supplemented by snowball sampling, to

identify and engage women entrepreneurs across various socioeconomic and geographic

backgrounds. Purposive Sampling was used in a way in which Initial participants were to be

selected based on a clear criteria:

(i) Women who independently run or co-own a small business or home-based enterprise, (ii)

Entrepreneurs with at least 6 months of operational activity, (iii) Individuals representing diverse

industries (e.g., food, crafts, retail, services), (iv) Use Instagram to sell their products. This

method ensures that participants have relevant experience with entrepreneurship, allowing for

richer and more insightful data regarding their challenges, resource needs, and views on potential

incubation support. Snowball Sampling was also used to ensure a larger data sample is collected.

Following the initial interviews, participants were asked to refer to other women entrepreneurs

within their networks who meet the same inclusion criteria. This technique enables access to

hidden segments, such as women in conservative communities who may not engage with formal

institutions.

Respondents Profiles

The five respondents operated a range of microenterprises, commonly characterized as

owner-operated or “mom-and-pop” businesses. These included custom cake shops, home

bakeries, crochet-based handicrafts, and boutique clothing operations. Most of these were

solo-run by women without formal staff or retail infrastructure. While two entrepreneurs relied

on the business as a main source of income, the rest operated them as side projects or

passion-based home ventures. Their varied motivations and degrees of digital fluency provided

rich insight into the asymmetric outcomes of algorithmic visibility.

19

A total of five valid responses were recorded. Although the sample is small, it represents a range

of product types including: (i) Custom cake businesses, (ii) Home bakeries, (iii) Crochet-based

handicrafts, (iv) Ice cream and dessert shops. Of these five respondents:

(a) Two stated that their Instagram business is their main source of income

(b) Three classified it as a hobby or side project

(c) All respondents had at least 6 months of experience using Instagram for commercial

purposes

(d) The duration of platform use ranged from 6 months to over 2 years

Variables studied

Eight questions were drawn up, each targeted towards helping understand how business

decisions are made. The first question inquired about the type of business they operate,

following that was a question asking them about how long they have been using the instagram

platform to sell their products. This helped us analyze whether they have familiarised

themselves with how the platform operates. Following that were questions related to how the

business owners decide the output and whether they operate the business as a main source of

their income or a side project. Questions following these were aimed at understanding whether

they have grasped the way that the algorithm works or not, mainly questioning whether they use

the insights, and whether business outcomes are based on audience engangement on their posts

or story. The areas that were a part of this study have been shown below and include mainly the

urban areas of Pakistan, and the demographic model lies between the ages of 18-30.

Limitations

This study faces several limitations that affect the generalizability and scope of its findings.

Firstly, the research is based on a small sample of five valid responses, which limits the statistical

strength of conclusions. However, this constraint is itself indicative of the broader issue under

investigation: the underrepresentation of women in Pakistan’s digital entrepreneurial landscape,

particularly in formal data-collection spaces. Secondly, infrastructural challenges most notably,

power outages and unstable internet were reported by respondents in regions such as Multan.

These disruptions led to delayed communication and highlighted the broader issue of digital

inequality that can hinder the consistency and efficiency of online business operations. Lastly,

the study is geographically concentrated in South Punjab. While this focus allows for a deeper

exploration of a specific context, it excludes perspectives from northern and remote regions of

20

Pakistan, where women entrepreneurs may face different cultural or infrastructural barriers.

Future research should aim for broader geographic coverage and larger sample sizes to capture

the diversity of experiences across Pakistan’s evolving digital economy.

Discussion

Algorithm Visibility and Strategic Adaptation

The results of our study can be interpreted through the lens of information

asymmetry and signaling theory. As Joseph Stiglitz demonstrated, uneven access to

information can lead to market inefficiencies. Similarly, Instagram’s opaque

algorithm creates a gap between what the platform “knows” about user preferences

and what entrepreneurs can observe. To cope, small businesses rely on visible

signals such as the style of their posts, image quality, or follower count to convey

credibility and appeal to consumers. This mirrors economist Michael Spence’s

(1973) theory of market signaling, where one party (the sender) uses costly,

observable actions to credibly convey unobservable qualities to another party (the

receiver). In this context, a like or follow becomes a proxy for product quality;

entrepreneurs adjust their strategies based on these cues. Crucially, each engagement

metric: likes, comments, saves acts as a form of demand signal, guiding both

consumer behavior and algorithmic exposure thereby functioning as an economic

signal within a platform-mediated market.

At the same time, the findings highlight significant opportunity costs faced by these

entrepreneurs. Time and effort devoted to crafting Instagram content is time not spent producing

goods or pursuing other sales channels – a classic economic trade-off. Women in Pakistan’s

major cities must balance these costs, often under uncertainty. This situation exemplifies

bounded rationality: without perfect foresight, they rely on experience and heuristics for

example, focusing on features that have historically worked for them or peers. The data suggest

that those who invest in learning the platform tend to see higher returns, while novices face

slower growth due to their learning curve. In aggregate, these patterns imply a potential

inefficiency. When information is uneven and entrepreneurs cannot calculate optimal strategies,

the distribution of visibility may skew toward already-established accounts or those lucky

enough to guess the right strategy. In other words, the Instagram market may not allocate

attention to the “best” products in the absence of perfect information – a form of market failure

induced by algorithmic complexity.

This behavioral pattern aligns with the theory of bounded rationality: when entrepreneurs operate

with incomplete or uncertain information, they develop adaptive heuristics—such as posting

21

more during high-reach periods or copying past successful content. The fact that no respondent

felt confident in understanding the algorithm, yet all engaged in behavior aimed at “pleasing” it,

indicates a perceived but unverified causal link between visibility tactics and business outcomes.

These entrepreneurs are making rational choices within irrational systems—adapting their

business models to an opaque and changing set of rules.

In addressing Research Question 1.2, the study found no variation in self-reported algorithm

understanding across regions, but some indirect signs of a digital divide were evident.

Respondents with longer exposure to Instagram, or with businesses based in more urbanized

areas like Lahore, appeared to rely more consistently on insights and strategic posting. While the

small sample limits firm conclusions, this hints at a possible correlation between geographic

digital literacy and the ability to navigate visibility economics more effectively. The absence of

formal training or guidance on how Instagram works leaves many entrepreneurs vulnerable to

algorithmic shifts that can dramatically reduce engagement thus, revenue without warning or

explanation.

Beyond individual behavior, these findings carry broader economic implications. As Instagram

becomes a substitute for physical marketplaces, especially for women excluded from traditional

employment, visibility becomes a gatekeeper for financial independence. However, when that

visibility is algorithmically controlled and commercially opaque, it introduces structural

inequality into a platform that claims to democratize opportunity. This algorithmic exclusion

where lack of platform fluency leads to diminished economic viability deserves policy attention,

particularly in regions where women’s economic participation already lags behind.

Ultimately, this research underscores a critical paradox: Instagram offers entrepreneurial

freedom, but only for those who can navigate its invisible systems. As digital platforms

increasingly become economic infrastructures in emerging markets, ensuring algorithmic

transparency, localized education, and inclusive platform governance will be essential to make

the promise of digital entrepreneurship a reality for all.

Economic Barriers Created by Algorithmic Systems

There are many economic barriers created by algorithmic systems; Firstly, Change in Media

Content It is no trade secret that algorithms favor certain types of content on their platforms. An

example of this is YouTube only monetizing and promoting family-friendly videos rather than

videos of a violent or sexual nature. However, not only do algorithms favor certain topics and

themes, but also the media it is presented in. When Instagram announced that algorithm change

will favor video content over images, there was an outcry from many creators. Small businesses

that relied on captivating photos were now in shambles trying to learn video production to

maintain their online audience. If every online platform periodically makes these adjustments to

their algorithm, marketers bear the burden of changing their marketing styles to cater to it

22

Secondally, Outsourcing Marketing Costs, For some businesses, tracking and creating new

content everyday may become overwhelming to monitor. Because of how challenging content

creation can be, some business owners may outsource their social media and advertising to a

third-party. While a third party may create the content you desire, there are also increased costs

to consider. What was once an easy and resourceful method to advertise to consumers has

quickly grown into a complicated and costly burden. There is also the risk of losing a business’

personal touch to their marketing when outsourced to other companies.

Empowerment vs Exploitation

The attention economy refers to the incentives of advertising-driven companies, in particular, to

maximize the time and attention their users give to their product. Attention economics is an

approach to the management of information that treats human attention as a scarce commodity

and applies economic theory to solve various information management problems

“Humans are natural social learners. We are constantly scanning the environment to figure out

what other people are doing and what we can learn from that,” says William Brady, an assistant

professor of management and organizations at Kellogg. “Social learning happens whenever we

observe people, get feedback from them, mimic them, and incorporate this information into our

understanding of norms.” Social media represents a new frontier for this type of learning. What

happens when this all-important observing and mimicking of others is mediated by algorithms

controlled by tech companies whose goals are to keep people’s attention on platforms?

Invisible Labour

Invisible labor is a philosophical, sociological, and economic concept applying to work that

is unseen, unvalued or undervalued, and often discounted as not important, despite its essential

role in supporting the functioning of workplaces, families, teams, and organizations.The term

was coined by Arlene Kaplan Daniels in the 1980s.

Opportunity Costs in Algorithm Chasing

In microeconomic theory, the opportunity cost of a choice is the value of the best alternative

forgone which, given limited resources, a choice needs to be made between several mutually

exclusive alternatives. Assuming the best choice is made, it is the "cost" incurred by not enjoying

the benefit that would have been had if the second best available choice had been taken instead.

The New Oxford American Dictionary defines it as "the loss of potential gain from other

alternatives when one alternative is chosen". As a representation of the relationship between

23

scarcity and choice, the objective of opportunity cost is to ensure efficient use of scarce

resources. It incorporates all associated costs of a decision, both explicit and implicit. Thus,

opportunity costs are not restricted to monetary or financial costs: the real cost of output forgone,

lost time, pleasure, or any other benefit that provides utility should also be considered an

opportunity cost. Every single time a social media platform makes updates to its platform or its

algorithm, business owners who use those platforms have to adjust. That can mean loads of time

and mental energy relearning a platform, making updates to templates, and sometimes, learning

something completely new. The first thing that comes to mind for me here is reels on Instagram.

Policy Recommendations

Improving Digital Literacy for Women Entrepreneurs

Training and Digital Literacy

Firstly, Training of Male Community Members, Men must also be trained to progress in women's

entrepreneurship development truly. This can mean training community leaders, such as religious

or local political leaders. The objective is holistic social development that will gradually but

steadily pave the way forward for women micro entrepreneurs. Moreover, there needs to be a

greater emphasis on Family Planning. Family planning must be central to all efforts involving

women in Pakistan. Given the sociocultural norms of Pakistan, women's entrepreneurship cannot

progress in an isolated environment, which makes it necessary to address other factors that

heavily influence women's entrepreneurship simultaneously.

Additionally, there is a need to collect more data on the determinants of digital enablement and

strategies for long-term improvement to empower women in Pakistan. Furthermore, investment

in research will help understand the specific technological needs and challenges and help design

customized training programs accordingly. Launching comprehensive awareness campaigns is

crucial for highlighting the benefits of technology adoption among women. To reach a broad

audience, these campaigns should leverage various media channels, including social media,

radio, television, and community outreach programs. Such initiatives will demystify technology

and build a supportive community where women feel encouraged to explore and integrate

technological solutions into their entrepreneurial ventures. Lastly, Online platforms for market

access are crucial. Just as training should move away from traditional methods, so should market

access platforms for women entrepreneurs. User-friendly e-commerce platforms tailored for

women should be developed to allow them to showcase and sell their products effectively

(Misbah Tanveer, 2023). DigiSkills is an online training program in Pakistan. This is an initiative

of the Government of Pakistan spearheaded by the Ministry of Information Technology and

Telecommunication through Ignite - National Technology Fund and executed by Virtual

University of Pakistan. DigiSkills provides online education in Virtual Assistant, Freelancing,

E-Commerce Management, Digital Marketing, Digital Literacy, QuickBooks, AutoCAD,

24

WordPress, Graphic Design, Creative Writing and Search Engine Optimization.

Government–Platform Partnerships and Subsidies

In Pakistan, where formal credit access for women remains low and digital skills are

unevenly distributed, digital subsidies represent a practical and targeted development tool. These

subsidies could include financial support for paid Instagram promotions, free access to business

account analytics, or government-sponsored vouchers for online marketing training. For many

women entrepreneurs operating home-based businesses, the cost of maintaining digital visibility

especially without guaranteed returns acts as a barrier to entry and growth. By subsidizing these

costs, the government could encourage formalization of microenterprises and reduce regional

disparities in platform engagement. Such subsidies would act much like traditional agricultural

or energy subsidies correcting market inefficiencies where private returns are high but initial

access is low.

The Pakistani government has long partnered with international organizations to expand

digital inclusion (e.g. with GSMA, UNDP, and Telenor). However, little effort has been made to

formally engage platform providers like Meta, despite the fact that Instagram and Facebook are

critical infrastructures for informal commerce. A policy-level public–private partnership (PPP)

could facilitate more transparent algorithm practices, platform training in regional languages, and

direct support for women-led businesses. For example, a national agreement with Meta could

support city-level "digital business incubators" for female entrepreneurs. Without this

intervention, the platform remains a private, unregulated gatekeeper of economic opportunity,

one whose biases and opacity disproportionately harm low-literacy and low-resource users.

Government involvement is therefore essential not only to protect users, but to ensure digital

platforms serve inclusive economic development.

Findings

Urban rural divide in Algorithm literacy

25

Figure 1

The areas from where the respondents operate their e-commerce businesses were mainly:

Lahore, Islamabad, Karachi, and Multan. These findings display a stark contrast between the

number of entrepreneurs using e-commerce in developed vs underdeveloped regions of

Pakistan. Areas such as Lahore, Karachi, and Islamabad where digital literacy is higher the

women indicate a higher usage of instagram as a tool for selling products. Whereas, areas such

as Multan are competitively underdeveloped, the internet usage is lower. The high number of

entrepreneurs in the developed regions can be attributed towards the algorithm, which curates

personalised content for each individual based on their preferences and exposes them to

enterprises that closely align with what they want. The algorithm constantly evolves with

customer behaviour and therefore not only allows the entrepreneurs to benefit by reaching a

wider audience but also encourages entrepreneurship mainly due to the high profits it generates;

thereby leading to a higher number of entrepreneurs in developed regions.

Respondents Profiles

The respondents were mainly aged between 18-30, hence were fairly acquainted with the use

of instagram. All of them were using the Instagram Marketplace for displaying their products;

26

pictures of their products were put up as posts and stories while some even had a website in the

bio for extra information. Notably, those who identified Instagram as a primary income source

appeared more likely to use features such as Reels, story highlights, and external shop links

suggesting a higher degree of platform fluency linked to financial intent. According to the survey

data, there was an equal number of women pursuing this as either a hobby or their main source

of income. Furthermore, most of the respondents were engaged in selling goods that require a

high level of skill and expertise, 80% of them were involved in professional baking such as cake

decorating, while the other 20% was involved in making handicrafts such as crochet products

and clothes designing.

Figure 2

Digital Skills as Human Capital

Like education in labor markets, digital fluency on Instagram functions as a

productivity-enhancing input. It amplifies exposure, boosts engagement, and ultimately increases

revenue, all while requiring upfront time investment, akin to traditional forms of human capital

accumulation. On Instagram, user engagement functions like a revealed preference or demand

signal in the marketplace of products. In classical economics, consumer choice reveals utility;

here, high engagement reveals consumer interest in a product or style. Entrepreneurs treat each

like or comment as if it were a vote of demand, signaling to the algorithm which posts to show to

more people. This process is analogous to price signaling in conventional markets: just as a surge

in sales signals strong demand for a product, a flurry of likes signals strong interest. The platform

responds by boosting that post’s visibility, creating a positive feedback loop. In effect, by

generating engagement entrepreneurs signal credibility and demand, overcoming some

informational gaps in the market.

Platform literacy: the skill and knowledge needed to use Instagram effectively yields economic

returns for entrepreneurs much like human capital in traditional economics. An owner who learns

about hashtags, content design, and features like Reels or Instagram Shopping will likely see

27

higher engagement and revenue than one who does not. This is analogous to returns to education:

investment in digital skills increases expected payoff. For instance, a shop owner who invests

time in mastering the platform gains an efficiency advantage, similar to a worker gaining more

productivity from education. These returns are especially steep because Instagram’s algorithm

disproportionately rewards savvy content strategies. Conversely, those who ignore new features

or best practices incur a form of hidden cost: their content is less visible and yields lower returns.

In this way, knowledge of the platform itself becomes a scarce resource with its own opportunity

cost, and it generates higher marginal revenue for the business.

Instagram visibility and Revenue Correlation

Instagram is more than just a social media platform it is a marketplace, an area where buyers

and sellers meet. Instagram algorithm chooses which businesses to boost and which ones to fail

through the audience engagement with the reels, posts, stories etc. It can be said that it decides

which businesses get seen or not. In instagram terminology the more you get seen, the more

profit you generate. Hence, visibility = revenue, making it essential for entrepreneurs to

understand the algorithm in order for the success of the business. Therefore, visibility and

revenue generation on Instagram can be said to have a positive correlation.

Figure 3

28

The Algorithm

Figure 4

The results of this question clearly illustrate the significant influence Instagram’s algorithmic

engagement metrics have on strategic business decisions. Notably, 100% of the respondents

reported that they had altered their product offerings whether in terms of variety, pricing, or

packaging based on how well their posts performed in terms of likes, shares, and reach.

This finding highlights a deeper behavioral adaptation among entrepreneurs: visibility metrics

are not simply vanity indicators, but active signals used to shape production and marketing

strategies. In the absence of direct consumer feedback, engagement becomes a proxy for

demand, and post performance becomes a form of real-time market research.

However, this reliance on algorithmically driven cues presents a risk. As post visibility is

governed by a non-transparent and fluctuating algorithm, product changes made in response to

such signals may be based on inconsistent or misleading data. This reinforces the need for

platform transparency, especially when users depend on these cues for economic

decision-making. However even though 100% of the respondents react to the algorithm almost

none of them feel they confidently understand how to use it. This further indicates that algorithm

literacy is low and does not vary significantly by region in the sample, but is still critical to

performance.

29

Case Study: Shaheen Ejaz and the Role of Digital Training

The successful case study of Shaheen Ejaz, that uses secondary data allows us to understand the

research question in greater detail. The experience of Shaheen Ejaz, a woman entrepreneur from

Quetta, offers a compelling illustration of how digital training and online visibility can transform

the financial outcomes of women-led microenterprises in Pakistan. Initially operating MNM

Products from her home with minimal sales and limited marketing knowledge, Shaheen

struggled to grow her herbal skincare business; an experience echoed by many of the survey

respondents in this study. Like them, she faced information barriers and lacked the skills needed

to navigate online commerce. This reflects Herbert Simon’s theory of bounded rationality,

wherein decision-makers operate under cognitive and informational constraints and rely on

satisficing strategies rather than optimization. Her case also highlights the economic returns to

human capital investment: after receiving e-commerce training under the EU-funded GRASP

initiative, Shaheen’s weekly revenue grew from approximately $20 to $175, a more than

eightfold increase. Her ability to rebrand, package, and promote her products online following

the training demonstrates how digital literacy can dramatically increase the efficiency and reach

of microenterprises.

While Instagram is not explicitly mentioned in her case, her shift to digital sales suggests entry

into the same platform-driven environment where visibility is algorithmically mediated and

directly tied to revenue. Her success reflects the argument presented in this paper that

algorithmic visibility acts as a form of economic capital, especially for women operating outside

of formal markets. Furthermore, her goal of expanding to a factory employing other women

points to the multiplier effect of women’s digital entrepreneurship: empowering one entrepreneur

has the potential to generate broader community-level economic benefits. From a policy

perspective, Shaheen’s journey validates the role of targeted digital training and support

programs in overcoming market failures related to information asymmetry and access to sales

channels. Her story underscores that with the right tools, even the most resource-constrained

women can participate in and benefit from the digital economy.

Conclusion

This study has explored how Instagram’s content-recommendation algorithm functions as both

an opportunity and a constraint for women-led small businesses in Pakistan. By treating digital

visibility as a form of economic capital, the paper has shown that algorithmic systems determine

not just which businesses are seen but which survive. Through survey responses, case study

analysis, and theoretical grounding in behavioral economics, information asymmetry, and

platform economics, it is evident that entrepreneurs adapt strategically to algorithmic feedback

despite low platform literacy. However, these adaptations often reflect satisficing behavior rather

30

than optimal strategy, highlighting the cognitive and informational constraints faced by

micro-entrepreneurs especially women in underserved regions.

The findings suggest that visibility on Instagram is positively correlated with revenue, yet this

visibility is neither neutral nor evenly distributed. It favors those who understand the mechanics

of engagement, hashtags, and content timing factors often inaccessible to entrepreneurs without

digital training. The case of Shaheen Ejaz underscores the transformative potential of digital

literacy: with minimal resources but targeted training, a woman entrepreneur was able to grow

her income significantly and contribute to community development. Her trajectory mirrors the

broader conclusion of this study that digital inclusion efforts must move beyond infrastructure

provision to include education, transparency, and algorithmic accountability.

While Instagram democratizes access to markets in theory, its opaque algorithms, regional

disparities, and unequal access to platform literacy create a system where opportunity is filtered

and fragile. Women entrepreneurs, particularly in underdeveloped areas, face a double bind: they

are excluded from formal markets and challenged by digital systems that reward familiarity,

frequency, and fluency. Without intervention, these dynamics risk deepening existing

socioeconomic inequalities.

Policy efforts must therefore prioritize inclusive digital training, public–private partnerships with

platform providers, and regulatory frameworks that ensure transparency in algorithmic

governance. Programs like DigiSkills, GRASP, and Women Ventures are a promising start, but

their success hinges on long-term support, context-sensitive curricula, and integration with

economic policy. If Instagram and similar platforms are to fulfill their potential as engines of

inclusive growth, they must not only enable access but also empower users with the tools to

succeed. Only then can digital entrepreneurship become a viable, equitable path to financial

independence for women across Pakistan.

31

References

Abidin, C., & Leaver, T. (2021). Instagram: Visual social media cultures. Polity Press.

Akbar, A. (2020). Factors influencing the usage of Instagram for business and its effect on

financial performance. Journal of Management & Research, 7(2), 45–58.

Gillespie, T. (2014). The relevance of algorithms. In T. Gillespie, P. J. Boczkowski, & K. A. Foot

(Eds.), Media technologies: Essays on communication, materiality, and society (pp. 167–194).

MIT Press.

GSMA. (2023). The mobile gender gap report 2023. https://www.gsma.com/r/gender-gap/

Meta for Business. (2024). Instagram advertising audience overview: Pakistan.

https://business.instagram.com/

Napoli, P. M. (2014). Automated media: Algorithmic culture and the search for meaning. The

International Journal on Media Management, 16(2), 63–74.

https://doi.org/10.1080/14241277.2014.918523

Pakistan Bureau of Statistics. (2023). Labour force survey 2022–23. https://www.pbs.gov.pk/

Simon, H. A. (1955). A behavioral model of rational choice. The Quarterly Journal of

Economics, 69(1), 99–118. https://doi.org/10.2307/1884852

Spence, M. (1973). Job market signaling. The Quarterly Journal of Economics, 87(3), 355–374.

https://doi.org/10.2307/1882010

Stiglitz, J. E. (2000). The contributions of the economics of information to twentieth century

economics. The Quarterly Journal of Economics, 115(4), 1441–1478.

https://doi.org/10.1162/003355300555015

UN Women. (2022). Digital inclusion of women entrepreneurs in South Asia: Challenges and

opportunities. https://www.unwomen.org/

van Dijck, J. (2013). The culture of connectivity: A critical history of social media. Oxford

University Press.

World Bank. (2024). Female labor force participation (Pakistan).

https://data.worldbank.org/indicator/SL.TLF.CACT.FE.ZS?locations=PK

Researcher-conducted survey. (June 2025). Form responses: Instagram’s algorithm and small

business decisions. Unpublished primary data.

NapoleonCat. (2024, March). Instagram users in Pakistan – March 2024. NapoleonCat.

32

https://napoleoncat.com/stats/instagram-users-in-pakistan/2024/03/

World Bank. (2023). Female labor force participation (% of female population ages 15+) –

Pakistan. TheGlobalEconomy.com. Retrieved June 12, 2025, from

https://www.theglobaleconomy.com/Pakistan/Female_labor_force_participation/

Appendix A

Research Questions:

1. What kind of business do you run on Instagram?

2. How long have you been using Instagram to promote/sell your products?

3. Is your business your main source of income or a side project?

4. How do you decide how much to produce or post in a week?

5. Have you ever changed your product offerings (e.g., variety, price, or packaging) based

on how well a post performs (likes, shares, reach)?

6. Do you feel confident that you understand how the Instagram algorithm works?

7. Have you ever changed the time/day you post in hopes of gaining more visibility?

8. Do you find yourself checking insights often to guide your business planning?

Optimal To view here ->

(All elements are viewable here)