Data strategist seeks fairness in subscription analytics
Tolani Akinrinoye, a U.S.-based data strategist and MBA candidate at the University of Pittsburgh, has called for a more humane, inclusive, and intelligent approach to customer prediction. Her recent co-authored publication in the International Journal of Social Science Exceptional Research, titled Predictive Analytics for Customer Lifetime Value in Subscription-Based Digital Service Platforms, offers a profound […]
Tolani Akinrinoye, a U.S.-based data strategist and MBA candidate at the University of Pittsburgh, has called for a more humane, inclusive, and intelligent approach to customer prediction.
Her recent co-authored publication in the International Journal of Social Science Exceptional Research, titled Predictive Analytics for Customer Lifetime Value in Subscription-Based Digital Service Platforms, offers a profound rethinking of how businesses should calculate the value of each user in subscription ecosystems.
And while the paper is a collaborative effort involving professionals from Amazon, Kwara State University, and California-based research institutes, Tolani’s contribution is distinct for its visionary focus on ethical intelligence, fairness in algorithm design, and the contextual complexities of emerging markets.
At a time when companies are increasingly dependent on Customer Lifetime Value (CLV) to drive decisions around pricing, retention, and personalization, Tolani Akinrinoye is asking deeper, more consequential questions: Who gets counted? Who gets missed? And are we optimizing for profit at the expense of equity?
“We talk a lot about data-driven personalization,” Tolani explains, “but we don’t talk enough about who the data leaves out. Models are trained on the past, but what if that past is unjust?”
Tolani’s section of the study stands out for its emphasis on what she terms predictive fairness, a methodology that not only builds models based on transactional behavior but ensures that these models are transparent, explainable, and free from the biases that often plague automated decision systems. Her concern is especially urgent for emerging markets, where consumer behavior tends to be more complex, infrastructure less stable, and digital footprints inconsistent.
Rather than treating churn or irregular usage as a sign of poor customer quality, Tolani argues that such patterns often reflect systemic barriers like energy shortages, unreliable internet, or shared mobile devices in large families. “If someone goes dark for a few days, it doesn’t mean they’re disloyal,” she notes. “It may mean their prepaid data ran out, or they had to choose between school fees and entertainment. That nuance matters.”
Tolani’s call to recalibrate these models is not just a technical one; it is deeply moral. In the paper, she outlines several case scenarios where poorly designed CLV frameworks have led to exclusionary outcomes. In one striking example, she discusses a telecom provider whose algorithm labeled customers as high churn risk based solely on nighttime inactivity, only to later discover that many of those users were in rural zones with erratic power supply. Rather than tweaking the model, the company simply offered those users fewer promotions and higher prices. For Tolani, such decisions are symptomatic of a larger problem: a failure to humanize the data.
To remedy this, she proposes a context-aware CLV framework, a layered approach that blends transaction history with social context, behavioral alternatives, and geographic conditions. Her framework incorporates community-level data, shared device logic, and economic seasonality, recognizing that in much of the Global South, value creation does not always fit neat Western formulas.
“Too often, models are imported from the West and plugged into African or Asian markets with minimal adaptation,” she explains. “But a subscription user in Nairobi or Ibadan lives in a different reality. They may value a service but lack the infrastructure to use it consistently. Are we accounting for that?”
Tolani’s questions are now shaping broader conversations beyond academia. Several fintech startups and digital healthcare ventures have reached out for consultation, looking to integrate her principles into their retention strategies. She is currently advising one education platform on how to identify high-potential learners based on engagement proxies, rather than payment frequency alone, a move that has already boosted retention by 27 percent.
What makes Tolani’s contribution especially timely is that she refuses to separate prediction from justice. In a world where AI is fast becoming the invisible gatekeeper for access, affordability, and even opportunity, she sees predictive analytics not as a neutral tool but as a social force that can either widen or narrow digital divides.
“CLV is powerful,” she says. “But if we only use it to serve the already-connected and the always-on, we’re building systems that privilege the privileged. That’s not just bad business, it’s dangerous.”
Her stance on ethical explainability, making predictive systems understandable to both technical and non-technical users, adds another layer to her leadership. She insists that platforms using predictive modeling must offer end users clear, accessible explanations of how decisions are made. She proposes visual dashboards that show users why they were offered a discount or denied one based on interpretable metrics. In low-literacy environments, this transparency could build unprecedented levels of trust.
“I believe trust is the most undervalued currency in digital services,” she emphasizes. “If users understand the logic, they stay. If they feel profiled or tricked, they leave.”
Beyond her technical acumen, Tolani’s advocacy extends into education. She is currently designing a CLV Fairness Toolkit, a free, open-source resource for small and medium-sized businesses in the Global South. The toolkit will offer explainable templates, bias testing protocols, and ethical auditing workflows, enabling even non-coders to build predictive models that respect context and inclusion.
The paper also outlines a framework for hybrid segmentation, a concept Akinrinoye helped formalize. It merges digital analytics with ethnographic fieldwork, recognizing that some of the best signals of loyalty and engagement are still invisible to algorithms. By combining qualitative interviews, payment heuristics, and referral network analysis, companies can now model CLV in a way that honors lived experience rather than just clickstream data.
Tolani’s research also tackles the issue of model fragility. In volatile economies, she warns, predictive systems trained on last year’s patterns may be dangerously outdated. She champions adaptive retraining cycles and community-led feedback loops, where insights from local agents or call center staff are looped back into the system. This participatory approach does not just refine accuracy; it creates digital platforms that feel accountable and rooted.
One of her more radical ideas is to flip the CLV equation entirely, to think not just in terms of how much value a customer provides to the platform, but how much value the platform provides to the customer. She introduces the concept of Platform Lifetime Value, where metrics include learning gained, wellness improved, income stabilized, or time saved. This reframing has enormous implications for social enterprises and donor-backed digital programs.
Asked what motivates her in a field where profitability often overshadows fairness, Tolani is candid. “Because I’ve seen what happens when people are excluded. I’ve seen good platforms fail because they misunderstood their customers. And I’ve seen bad platforms succeed by exploiting gaps in data protection. I want better models, and I believe they’re possible.”
Tolani Akinrinoye is not content with being a thought leader. She is a builder of frameworks, of tools, and of futures. She is shaping a generation of ethical data strategists who understand that intelligence is not just computational, but cultural. And she is proving that even the most technical of metrics, Customer Lifetime Value, can become a site of empathy, dignity, and justice.
As the subscription economy deepens across the globe and algorithms take on more responsibility in shaping how services are distributed, it will be the voices like Tolani’s, rooted in ethics, fluent in culture, and bold in vision, that lead the way. She is not just calculating value. She is creating it.