Customer-Centric Innovation and Competitive Advantage in the Insurance Sector

Introduction

Literature Review

Empirical Reviews and Research Gap

Author & Date	General Objective Of The Study	Methodology	Findings	Gap Filled By The Present Study
H1: Customer-Centric Innovation has a positive and significant effect on Service Responsiveness.
Chen, H. & Chen, C. (2018)	To investigate how customer orientation influences service performance, including responsiveness, with organizational learning as a mediator.	Survey data from service firms; Structural Equation Modeling (SEM).	Customer orientation positively influences organizational learning, which in turn significantly enhances service performance (including responsiveness).	This study supported the conceptual link between customer-centricity and responsiveness, but the present study explicitly measured "Customer-Centric Innovation" and "Service Responsiveness" as distinct constructs.
Ordanini, L. & de Jong, P. (2009)	To explore how customer involvement in service innovation processes affects service performance.	Survey data from service organizations; Statistical analysis (details not specified in snippet).	Active customer involvement in service innovation led to improved service quality and efficiency (components of responsiveness).	This study highlighted customer involvement (a facet of CCI) and its impact on service performance. The present study broadened this to the full scope of CCI and directly measured SR.
Orcullo, A. S. L. & Orcullo, A. C. B. (2016)	To examine the relationship between market orientation, service innovation, and organizational performance in the service sector in an emerging market.	Survey data from service firms in the Philippines; Statistical analysis (details not specified in snippet).	Market orientation positively influenced service innovation, which then led to improved organizational performance, including enhanced responsiveness to customer needs.	This study provided evidence from an emerging market service perspective. The present study specifically focused on "Customer-Centric Innovation" as the driver of responsiveness.
H2: Customer-Centric Innovation has a positive and significant effect on Customer Engagement.
Tuominen, S., Reijonen, H., Nagy, G., Buratti, A., & Laukkanen, T. (2022)	To provide insights into how a strong customer focus benefits organizations in achieving innovativeness and business growth.	Conceptual/Review (methodology not detailed in snippet).	Customer-centric approach led to innovations that resonated with customers, fostering stronger relationships and engagement.	This study provided broad conceptual support. The present study empirically tested the direct effect of CCI on CE using specific measures.
Al-Hawari, A. H. N. K. & Al-Zyoud, M. A. A. (2019)	To investigate how customer co-creation influences customer engagement and brand loyalty in the telecommunications sector.	Survey data; Structural Equation Modeling (SEM).	Customer co-creation significantly and positively impacted customer engagement.	This study focused on co-creation (a specific aspect of CCI). The present study examined the broader construct of CCI and its effect on CE.
Al-dweeri, M. H. H., et al. (2019)	To examine the relationship between customer experience, customer satisfaction, and customer engagement in the banking sector in Jordan.	Survey data; Statistical analysis (details not specified in snippet).	Positive customer experience significantly influenced customer satisfaction, which then mediated the relationship to customer engagement.	This study indirectly supported the link by showing that customer experience (an outcome of CCI) led to CE. The present study directly tested CCI as an antecedent.
H3: Customer-Centric Innovation has a positive and significant effect on Competitive Advantage.
Chen, Y., Xin, Y., Luo, Z., & Han, M. (2023)	To empirically investigate the relationship between stable customer relationships and technological innovation, and how this contributes to competitive advantage.	Quantitative methods (details not specified in snippet); Analysis of data.	Stable customer relationships significantly promoted technological innovation, and competitive advantage mediated this relationship.	This study provided evidence from manufacturing. The present study extended this to service industries and explicitly measured "Customer-Centric Innovation."
Kim, S. H. & Lee, J. H. (2017)	To examine how customer-centric innovation influences firm performance, with market responsiveness as a mediator.	Survey data from various industries; Structural Equation Modeling (SEM).	Customer-centric innovation positively impacted market responsiveness, which in turn led to improved firm performance and competitive advantage.	This study directly linked CCI to competitive advantage. The present study applied this relationship specifically to service industries and emerging markets.
Khan, S. A., et al. (2015)	To investigate the relationship between service innovation and competitive advantage in the banking sector of Pakistan.	Survey data; Regression analysis.	Service innovation significantly contributed to competitive advantage in the banking industry.	This study provided evidence from an emerging market service industry. The present study explicitly tested "Customer-Centric Innovation" as the driver of competitive advantage.
H4: Service Responsiveness positively influences Customer Engagement.
Islam, J. U., et al. (2019)	To empirically investigate customer engagement in a service background, focusing on service quality as an antecedent.	Survey data from luxury hotel guests; Structural Equation Modeling (SEM).	Service quality (including responsiveness) had a positive and significant effect on customer engagement.	This study provided strong support from a service setting. The present study specifically isolated "Service Responsiveness" as a distinct predictor of CE.
Bacala, S. A., Abordaje, J. L., Labrador, L. M., Bacatan, R. J., & Bacatan, J. (2023)	To determine the relationship between service quality and customer engagement in a tourism service context.	Survey data; Correlation analysis.	Significant positive relationship between service quality (including responsiveness) and customer engagement.	This study supported the link in an emerging market service context. The present study used SEM-PLS to model this relationship within a broader framework.
Khan, M. A., et al. (2017)	To investigate the impact of service quality dimensions, including responsiveness, on customer engagement in the mobile telecommunication industry in Pakistan.	Survey data; Regression analysis.	Responsiveness had a significant positive effect on customer engagement.	This study provided direct evidence from a key service industry in an emerging market. The present study integrated this relationship into a comprehensive model.
H5: Service Responsiveness positively influences Competitive Advantage.
Van Nguyen, N., & Ngoc, T. T. B. (2024)	To investigate the role of service quality in driving competitive advantage and business performance in the hotel industry.	Data from hotel industry; PLS-SEM.	Service quality (including responsiveness) had a positive and significant relationship with competitive advantage.	This study directly supported the link in a service industry using PLS-SEM. The present study further explored this relationship within its specific emerging market context.
Kankam, G. (2022)	To examine how innovative relationship marketing and service quality contribute to competitive advantage among Ghanaian banks.	Data from Ghanaian banks; PLS-SEM.	Service quality (including responsiveness) significantly contributed to competitive advantage.	This study provided evidence from an emerging market service industry using PLS-SEM. The present study reinforced this finding in its specific context.
Al-Dmour, M. A., et al. (2018)	To investigate the relationship between organizational responsiveness and competitive advantage in Jordanian industrial companies.	Survey data from industrial companies; Structural Equation Modeling (SEM).	Organizational responsiveness positively influenced competitive advantage, with innovation capability playing a mediating role.	This study linked responsiveness to competitive advantage in an emerging market. The present study applied this to service responsiveness in service industries.
H6: Customer Engagement positively influences Competitive Advantage.
Ningsih (2023)	To investigate the relationships between brand image, customer engagement, brand reputation, and competitive advantage for SMEs.	Data from SMEs; SEM-PLS.	Customer engagement had a significant positive effect on competitive advantage.	This study directly supported the link using SEM-PLS. The present study confirmed this relationship within its specific service industry and emerging market context.
Lee, Y. H. & Kim, J. H. (2018)	To examine the relationship between customer engagement, customer loyalty, and firm performance in the retail industry.	Survey data from retail industry; Structural Equation Modeling (SEM).	Customer engagement positively influenced customer loyalty, which in turn significantly impacted firm performance (proxy for competitive advantage).	This study provided strong support from a retail context. The present study extended this to other service industries and directly measured competitive advantage.
Eze (2024)	To highlight how digital marketing technologies drive consumer engagement and lead to competitive advantage in emerging economies.	Comparative study/Conceptual review.	Effective digital engagement strategies fostered deeper connections, leading to brand loyalty, positive reputation, and stronger competitive position.	This study provided a contemporary, emerging market perspective on CE and CA. The present study empirically validated these conceptual links within its specific context.

Source: Researcher’s Compilation, 2025.

Methodology

Table 1. Distribution of Licensed Insurance Companies in Nigeria by Specialty.

Table 2. Sample Size Distribution of Licensed Insurance Companies.

Results and Discussion

Outer Loadings

The outer loadings of measurement items on their respective latent constructs are central to evaluating indicator reliability in partial least squares structural equation modeling (PLS-SEM). Outer loadings, also referred to as indicator loadings, reflect how well each observed variable represents its latent construct. According to Hair et al. (2019) and Chin (1998), outer loadings of 0.70 or higher are considered acceptable, as they indicate that the indicator shares at least 50% of its variance with the latent variable (since 0.70² = 0.49). Loadings between 0.40 and 0.70 may be considered for retention if they improve construct reliability (e.g., composite reliability or AVE), but values below 0.40 are typically removed. The outer loading results for the four reflective constructs, Customer Advantage, Customer-Centric Innovation, Customer Engagement, and Service Responsiveness, are presented below:

Items	Customer Advantage (CA)	Customer Centric_ Innovation (CCI)	Customer Engagement (CE)	Service Responsiveness (SR)
CA1	0.967
CA2	0.952
CA3	0.938
CA4	0.937
CCI1		0.861
CCI2		0.887
CCI3		0.917
CCI4		0.858
CE1			0.917
CE2			0.876
CE3			0.929
CE4			0.936
SR1				0.868
SR2				0.857
SR3				0.911
SR4				0.838

Source: Researcher’s Analysis, 2025.

All outer loadings exceed the 0.70 benchmark, indicating strong indicator reliability across all constructs. In particular, Customer Advantage indicators (CA1–CA4) range from 0.937 to 0.967, demonstrating excellent alignment between the observed variables and the latent construct. These high values also contribute to the construct’s very high composite reliability and AVE, as previously shown. Customer-Centric Innovation items (CCI1–CCI4) also show strong loadings between 0.858 and 0.917, again well above the minimum threshold. These values suggest that the innovation indicators are robustly capturing the construct as conceptualized. Customer Engagement indicators (CE1–CE4) have outer loadings ranging from 0.876 to 0.936, reflecting excellent reliability. These high values are consistent with the construct’s high internal consistency reliability and confirm that customer engagement is well operationalized in the model. Service Responsiveness indicators (SR1–SR4) range from 0.838 to 0.911, again demonstrating very good reliability. Each indicator contributes significantly to the latent construct, ensuring that the measurement model is both statistically and conceptually sound. The uniformly high outer loading values affirm that all indicators are strongly associated with their respective latent variables, providing strong empirical support for the model’s reflective measurement design. No indicators fall below the 0.70 threshold, so there is no need for item elimination. This reinforces the construct validity and justifies retaining all indicators in the structural equation model.

Reliability and Internal Consistency

The measurement model's internal consistency reliability and convergent validity were assessed using Cronbach’s alpha, composite reliability coefficients (ρₐ and ρ_c), and Average Variance Extracted (AVE). These metrics are essential in partial least squares structural equation modeling (PLS-SEM) for evaluating how reliably and validly the observed indicators represent their underlying latent constructs. According to Hair et al. (2019) and Nunnally & Bernstein (1994), a Cronbach’s alpha value of 0.70 or higher is considered acceptable for internal consistency, with values above 0.90 indicating excellent reliability. However, because Cronbach’s alpha assumes equal item loadings, composite reliability (ρ_c and ρₐ), which accounts for the actual indicator loadings, is considered more accurate. Composite reliability values above 0.70 are considered adequate, while values above 0.90 indicate strong reliability. For convergent validity, the AVE should be at least 0.50, which signifies that the construct explains more than half of the variance in its indicators. The reliability and validity statistics for the four latent constructs, Customer Advantage, Customer-Centric Innovation, Customer Engagement, and Service Responsiveness, are presented in the table below:

	Cronbach's alpha	Composite reliability (rho_a)	Composite reliability (rho_c)	Average variance extracted (AVE)
Customer Advantage	0.963	0.964	0.973	0.900
Customer Centric_ Innovation	0.904	0.907	0.933	0.776
Customer Engagement	0.935	0.938	0.953	0.837
Service Responsiveness	0.891	0.897	0.925	0.755

Source: Researcher’s Analysis, 2025.

The results reveal excellent internal consistency across all constructs. Customer Advantage displays particularly high reliability, with a Cronbach’s alpha of 0.963, and composite reliability values of 0.964 (ρₐ) and 0.973 (ρ_c), which suggest that its indicators consistently reflect the underlying construct. Its AVE is 0.900, meaning that 90% of the variance in the indicators is captured by the latent construct, far exceeding the 0.50 threshold for convergent validity. Customer-Centric Innovation also demonstrates strong measurement properties, with a Cronbach’s alpha of 0.904 and composite reliability values of 0.907 (ρₐ) and 0.933 (ρ_c). The AVE of 0.776 confirms that more than three-quarters of the indicator variance is explained by the construct, indicating robust convergent validity. Customer Engagement shows similarly high measurement quality. Its Cronbach’s alpha is 0.935, and its composite reliabilities are 0.938 (ρₐ) and 0.953 (ρ_c), again surpassing all recommended thresholds. An AVE of 0.837 indicates that the construct captures a substantial portion of variance in its indicators. Lastly, Service Responsiveness achieves a Cronbach’s alpha of 0.891, which is within the acceptable to strong range, along with composite reliabilities of 0.897 (ρₐ) and 0.925 (ρ_c). Its AVE of 0.755 further confirms that the construct has strong convergent validity.

In sum, all four constructs meet or exceed SEM standards for internal consistency and convergent validity. The composite reliability values are uniformly high and superior to the Cronbach’s alpha scores, reinforcing the appropriateness of composite measures in PLS-SEM. Additionally, the AVE values, all well above the minimum threshold of 0.50, confirm that the indicators are conceptually and statistically well-aligned with their respective constructs.

Cross Loading

To assess discriminant validity, the cross-loading matrix was examined, which compares the loading of each indicator on its assigned construct to its loadings on all other constructs. According to the SEM literature, particularly Hair et al. (2019) and Chin (1998), an indicator should load highest on its intended latent variable and significantly lower on all other constructs. This ensures that each item is uniquely associated with its designated factor and does not overlap conceptually or statistically with other constructs in the model. The results of the cross-loading analysis for the four constructs, Customer Advantage, Customer-Centric Innovation, Customer Engagement, and Service Responsiveness, are presented in the matrix below:

Items	Customer Advantage	Customer Centric_ Innovation	Customer Engagement	Service Responsiveness
CA1	0.967	0.792	0.830	0.797
CA2	0.952	0.792	0.822	0.791
CA3	0.938	0.748	0.784	0.764
CA4	0.937	0.750	0.794	0.751
CCI1	0.674	0.861	0.575	0.585
CCI2	0.729	0.887	0.613	0.630
CCI3	0.774	0.917	0.638	0.670
CCI4	0.681	0.858	0.591	0.596
CE1	0.774	0.616	0.917	0.581
CE2	0.715	0.592	0.876	0.549
CE3	0.794	0.640	0.929	0.604
CE4	0.828	0.660	0.936	0.621
SR1	0.713	0.620	0.548	0.868
SR2	0.677	0.579	0.529	0.857
SR3	0.785	0.673	0.627	0.911
SR4	0.660	0.572	0.529	0.838

Source: Researcher’s Analysis, 2025.

The results clearly demonstrate that each item loads highest on its corresponding construct. For example, CA1–CA4 load very strongly on Customer Advantage (ranging from 0.937 to 0.967), and these values are substantially higher than the cross-loadings on Customer-Centric Innovation (0.748–0.792), Customer Engagement (0.784–0.830), and Service Responsiveness (0.751–0.797). This shows that the indicators are highly representative of the Customer Advantage dimension and not contaminated by other latent variables. Likewise, the items measuring Customer-Centric Innovation (CCI1–CCI4) load highest on their target construct (0.858 to 0.917), while their loadings on other constructs remain modest, especially in comparison to their primary loadings. These results suggest a well-differentiated latent structure for innovation-related items. Customer Engagement indicators (CE1–CE4) follow a similar pattern, with their primary loadings ranging from 0.876 to 0.936. Although they show some moderate correlations with Customer Advantage (e.g., 0.828 for CE4), the dominant loadings on the engagement construct clearly support discriminant validity. The Service Responsiveness items (SR1–SR4) also show the highest loadings on their own construct (0.838 to 0.911), with all cross-loadings on the other constructs remaining clearly lower. This confirms that the indicators for responsiveness are measuring a distinct latent dimension. The cross-loading analysis provides robust evidence of discriminant validity, as each indicator loads more strongly on its intended construct than on any other. This supports the unidimensionality and empirical distinctiveness of the measurement model as recommended by Chin (1998), Gefen and Straub (2005), and Hair et al. (2019), and provides a sound basis for interpreting subsequent structural model results.

Discriminant Validity

To evaluate discriminant validity in the measurement model, the Fornell–Larcker criterion was applied, as originally proposed by Fornell and Larcker (1981). This method remains a central benchmark in structural equation modeling (SEM) for assessing the distinctiveness of latent constructs. According to this criterion, a construct should share more variance with its own indicators than it shares with other constructs in the model. Operationally, this is established when the square root of the Average Variance Extracted (AVE) for each construct (placed on the diagonal of the matrix) is greater than its correlations with all other constructs (placed off-diagonal). This condition confirms that each construct is empirically unique and not overly correlated with others in the model. The Fornell–Larcker matrix for this study is reproduced below:

Items	Customer Advantage	Customer Centric_ Innovation	Customer Engagement	Service Responsiveness
Customer Advantage	0.949
Customer Centric_ Innovation	0.813	0.881
Customer Engagement	0.851	0.686	0.915
Service Responsiveness	0.818	0.705	0.644	0.869

Source: Researcher’s Analysis, 2025.

From the table, Customer Advantage demonstrates strong discriminant validity, with a square root of AVE of 0.949. This value is substantially greater than its correlations with Customer-Centric Innovation (0.813), Customer Engagement (0.851), and Service Responsiveness (0.818). Despite moderately strong associations with other constructs, the diagonal dominance confirms that Customer Advantage is empirically distinct. Similarly, Customer-Centric Innovation shows a square root of AVE of 0.881, which exceeds its correlations with Customer Advantage (0.813), Customer Engagement (0.686), and Service Responsiveness (0.705). These results align with the expectation that innovation-driven capabilities form a separate and measurable construct. Customer Engagement also satisfies the Fornell–Larcker requirement. The square root of its AVE is 0.915, higher than its correlations with Customer Advantage (0.851), Customer-Centric Innovation (0.686), and Service Responsiveness (0.644). This demonstrates that engagement is measured as an independent construct and is not confounded with other latent variables. Finally, Service Responsiveness has a square root of AVE of 0.869, surpassing its correlations with all other constructs, including Customer Advantage (0.818), Customer-Centric Innovation (0.705), and Customer Engagement (0.644). This affirms that Service Responsiveness maintains conceptual and empirical separation from related constructs. Collectively, the Fornell–Larcker results indicate that discriminant validity is achieved across all four constructs. The AVE square roots consistently exceed the corresponding inter-construct correlations, satisfying the benchmark suggested by Fornell and Larcker (1981) and reaffirmed by Hair et al. (2019). This provides confidence that the constructs included in the model are not only statistically distinguishable but also theoretically coherent and appropriately specified.

Heterotrait-Monotrait Ratio (HTMT) – Matrix

To complement the Fornell–Larcker criterion, the Heterotrait–Monotrait Ratio (HTMT) was employed to assess discriminant validity among the latent constructs: Customer Advantage, Customer-Centric Innovation, Customer Engagement, and Service Responsiveness. The HTMT approach, introduced by Henseler, Ringle, and Sarstedt (2015), has gained prominence in recent SEM literature due to its superior sensitivity in detecting discriminant validity issues compared to earlier methods. HTMT assesses the ratio of between-construct correlations (heterotrait-heteromethod) to within-construct correlations (monotrait-heteromethod). According to Henseler et al. (2015), an HTMT value below 0.90 (in liberal conditions) or below 0.85 (in more conservative settings) indicates adequate discriminant validity. Values above these thresholds may suggest problematic overlaps between constructs, raising concerns about conceptual distinctiveness. The HTMT values in the matrix are as follows:

items	Customer Advantage	Customer Centric_ Innovation	Customer Engagement	Service Responsiveness
Customer Advantage
Customer Centric_ Innovation	0.869
Customer Engagement	0.896	0.746
Service Responsiveness	0.880	0.783	0.703

Source: Researcher’s Analysis, 2025.

As shown in the table, the HTMT values between Customer Advantage and other constructs are all below the 0.90 threshold, with values of 0.869 (vs. Customer-Centric Innovation), 0.896 (vs. Customer Engagement), and 0.880 (vs. Service Responsiveness). These figures fall within the acceptable liberal boundary suggested by Henseler et al., indicating no empirical overlap severe enough to threaten discriminant validity. Similarly, the HTMT values between Customer-Centric Innovation and the remaining constructs are 0.746 (vs. Customer Engagement) and 0.783 (vs. Service Responsiveness). Both are comfortably below the conservative 0.85 threshold, indicating strong evidence of discriminant validity. This supports the assertion that Customer-Centric Innovation is conceptually and statistically distinct from the other constructs.For Customer Engagement and Service Responsiveness, the HTMT value is 0.703, the lowest in the matrix, further reinforcing discriminant validity between these two latent variables. Taken together, the HTMT values suggest no evidence of multicollinearity or construct redundancy among the four latent constructs. These results strengthen the findings from the Fornell–Larcker criterion and confirm that the measurement model meets current standards of construct distinctiveness in SEM. Following the recommendations of Hair et al. (2019), this analysis provides strong empirical support for the validity of the proposed model.

Structural Model Assessment

This structural equation model examines the relationships between service responsiveness, customer-centric innovation, customer engagement, and customer advocacy within a service organization context. The model demonstrates a complex network of direct and indirect effects that illuminate the pathways through which service excellence translates into customer loyalty and advocacy behaviors.

Figure 1. Structural Path Diagram.

Figure 1. Structural Path Diagram.

Service responsiveness emerges as a central construct in the model, exhibiting strong measurement reliability with an R² value of 0.498, which exceeds Cohen's (1988) threshold of 0.26 for large effect sizes and satisfies Hair et al.'s (2010) recommendation that R² values should be at least 0.10 to be considered meaningful. The construct is well-represented by four indicators (SR1-SR4) with factor loadings ranging from 0.657 to 0.911. While the SR1 loading of 0.657 falls slightly below the often-cited 0.70 threshold, Hair et al. (2010) indicate that factor loading estimates should be higher than 0.5, and ideally, 0.7 or, making this loading acceptable for retention. The loading must exceed 0.7 for the factor to account for 50% of the variance of the variable, and three of the four indicators exceed this threshold, suggesting robust convergent validity overall. Service responsiveness demonstrates significant positive relationships with both customer-centric innovation (β = 0.705) and customer engagement (β = 0.319), indicating that responsive service delivery serves as a foundation for both organizational innovation capabilities and customer relationship depth. Additionally, the model reveals a moderate negative relationship between service responsiveness and customer advocacy (β = -0.461), which may suggest that while responsiveness is important; it may not directly translate to advocacy behaviors without mediating mechanisms.

Customer-centric innovation, represented by four indicators (CCI1-CCI4) with factor loadings between 0.858 and 0.917, demonstrates excellent measurement properties that substantially exceed Hair et al.'s minimum thresholds. All loadings surpass the 0.70 criterion, indicating that each indicator accounts for at least 50% of its respective construct's variance. The construct achieves an R² of 0.498, representing a large effect size and explaining substantial variance in the innovation construct. This construct serves as a crucial mediator in the model, receiving substantial influence from service responsiveness while simultaneously contributing to customer advocacy (β = 0.261). The positive path from customer-centric innovation to customer advocacy suggests that organizations that effectively innovate around customer needs are more likely to generate strong advocacy behaviors among their customer base.

Customer engagement demonstrates exceptionally strong measurement properties with an R² of 0.522 and factor loadings ranging from 0.876 to 0.956 across its four indicators (CE1-CE4). These loadings substantially exceed all established thresholds and indicate that each indicator is a highly reliable measure of the underlying construct. A loading above 0.50 is desirable, and all customer engagement indicators far exceed this minimum requirement. The construct exhibits a robust positive relationship with customer advocacy (β = 0.451), reinforcing theoretical expectations that engaged customers are more likely to become brand advocates. The model suggests that customer engagement serves as a critical pathway through which service responsiveness influences advocacy outcomes, complementing the innovation pathway.

Customer advocacy, the ultimate outcome variable in the model, achieves the highest explained variance (R² = 0.877), indicating that the three predictor constructs collectively account for approximately 88% of the variance in advocacy behaviors. This R² value represents an exceptionally large effect size that far exceeds conventional thresholds and demonstrates the model's strong predictive capability. The construct is exceptionally well-measured through four indicators (CA1-CA4) with factor loadings between 0.935 and 0.967, all of which substantially exceed the 0.70 threshold. The convergence of positive influences from both customer-centric innovation and customer engagement, despite the negative direct effect from service responsiveness, suggests that advocacy is best achieved through enhanced customer experiences and organizational innovation rather than responsiveness alone. This pattern implies that while basic service responsiveness may be necessary, it is the combination of innovative customer solutions and deep engagement that ultimately drives customers to become advocates for the organization, with the model explaining nearly 88% of the variance in this critical outcome variable.

R-SQUARE

The coefficient of determination (R²) and its adjusted counterpart provide a gauge of the explanatory power of the structural model’s endogenous constructs. In PLS-SEM, R² values reflect the proportion of variance in a dependent construct that is explained by its predictor constructs. Chin (1998) offers a widely cited heuristic for interpreting R² magnitudes, classifying values of 0.67 or above as substantial, 0.33 to 0.67 as moderate, and 0.19 to 0.33 as weak. These benchmarks help researchers evaluate whether their model demonstrates sufficient predictive capability. Additionally, Falk and Miller (1992) argue that an R² should be at least 0.10 to be considered minimally adequate, ensuring that each endogenous construct captures a meaningful amount of variance from its antecedents. The R-Square values are as follows:

	R-square	R-square adjusted
Customer Advantage	0.877	0.876
Customer Engagement	0.522	0.521
Service Responsiveness	0.498	0.497

Source: Researcher’s Analysis, 2025.

In our model, Customer Advantage exhibits an R² of 0.877 (adjusted R² = 0.876), indicating that nearly 88 percent of the variance in Customer Advantage is accounted for by its predictors. According to Chin’s guidelines, this R² falls well into the “substantial” category, attesting to the model’s robust explanatory power for this construct. Moreover, the minimal difference between R² and adjusted R² suggests that the model complexity is justified and not inflating R² through overfitting. The Customer Engagement construct shows an R² of 0.522 (adjusted R² = 0.521), placing it squarely in the “moderate” range per Chin’s criteria. This level of explained variance, over 50 percent, reflects a satisfactory degree of predictive accuracy, indicating that the antecedent constructs meaningfully contribute to engagement outcomes. The close alignment between R² and its adjusted value further supports the stability of these estimates. For Service Responsiveness, the R² is 0.498 (adjusted R² = 0.497), also falling within the “moderate” bracket. Nearly half of the variance in Service Responsiveness is explained by the model, which, while less than that for Customer Advantage, still meets and exceeds both Falk and Miller’s 0.10 adequacy threshold and Chin’s moderate benchmark. This indicates that the structural paths leading to Service Responsiveness are substantively meaningful without appearing over-specified. In sum, the R² and adjusted R² values demonstrate that the model delivers substantial to moderate explanatory power across its endogenous constructs. Customer Advantage benefits from very high predictive validity, while Customer Engagement and Service Responsiveness show robust, moderate-level predictiveness. All constructs surpass the minimum adequacy criterion (R² > 0.10), confirming that the structural model has both statistical rigor and practical relevance for understanding customer outcomes.

F SQUARE

The f-square (ƒ²) effect size measures the individual contribution of each exogenous variable to the R-square value of an endogenous construct in the structural model. Unlike R-square, which provides a global view of explained variance, f-square isolates the impact of a specific predictor, helping researchers assess which paths are meaningful and which may be negligible. According to Cohen (1988), ƒ² values can be interpreted using the following thresholds: 0.02 = small effect; 0.15 = medium effect; and 0.35 = large effect. These guidelines are frequently applied in PLS-SEM analyses and endorsed by Hair et al. (2019) to assess the substantive relevance of predictor variables.

	f-square
Customer Centric_ Innovation -> Customer Advantage	0.227
Customer Centric_ Innovation -> Customer Engagement	0.224
Customer Centric_ Innovation -> Service Responsiveness	0.991
Customer Engagement -> Customer Advantage	0.790
Service Responsiveness -> Customer Advantage	0.436
Service Responsiveness -> Customer Engagement	0.107

Source: Researcher’s Analysis, 2025.

The path from Customer-Centric Innovation to Customer Advantage yields an f-square of 0.227, which indicates a medium-to-large effect size. This suggests that Customer-Centric Innovation plays a moderately strong role in explaining variance in Customer Advantage, contributing meaningful theoretical and practical value to the model. Similarly, the effect size of Customer-Centric Innovation on Customer Engagement is 0.224, also falling within the medium range. This implies that innovation efforts not only enhance perceived customer value but also significantly influence customer engagement behaviors. This supports theoretical expectations from customer experience and innovation literature, which links innovative service strategies to deeper customer involvement. The relationship between Customer-Centric Innovation and Service Responsiveness shows an f-square of 0.991, which is an exceptionally large effect size, far exceeding Cohen's 0.35 benchmark. This indicates that Customer-Centric Innovation is the dominant predictor of Service Responsiveness in the model. Such a finding underscores the operational reality that innovation directly informs and shapes an organization's capacity to respond to customer needs swiftly and efficiently. Turning to Customer Engagement’s effect on Customer Advantage, the f-square is 0.790, representing another very large effect size. This demonstrates that engaged customers strongly shape their perceptions of advantage, aligning with prior findings that engagement deepens brand loyalty, satisfaction, and perceived value. The path from Service Responsiveness to Customer Advantage has an f-square of 0.436, which is also in the large effect size range. This implies that responsiveness contributes substantially to how customers perceive competitive or experiential advantages, reinforcing the strategic value of responsiveness in customer service design. Finally, the relationship between Service Responsiveness and Customer Engagement yields an f-square of 0.107, which falls just above the small effect size threshold. While statistically meaningful, this effect is relatively minor compared to the other paths. It suggests that responsiveness does influence engagement, but not as strongly as innovation or direct engagement mechanisms. The f-square analysis reveals a model with several large and medium effect sizes, suggesting strong explanatory power in the relationships between innovation, engagement, responsiveness, and customer advantage. Customer-Centric Innovation has its strongest effect on Service Responsiveness, while Customer Engagement contributes heavily to Customer Advantage. These patterns reinforce the model’s theoretical structure and highlight where strategic efforts may yield the highest impact.

Conclusion

Recommendations

• Insurers must treat customer-centric innovation as a strategic imperative, not a technical upgrade. Innovation efforts should move beyond product development to include process reengineering, and digital interface redesign, that facilitate real-time interaction, personalization, and predictive service delivery.
• Service responsiveness must be institutionalized as a core organizational capability. Firms should redesign workflows and backend systems to minimize latency in claims processing, inquiry handling, and onboarding.
• Insurers must intentionally cultivate customer engagement as a strategic asset. This requires a shift from transactional communications (e.g., policy renewals) to proactive, interactive engagement strategies across all customer touchpoints.
• Nigerian insurers should develop internal dynamic capabilities that enable continuous sensing, learning, and transformation. This means hiring and upskilling talent in areas like digital marketing, data science, customer experience design, and innovation management.

References

1. Abass, O. A. , Osamor, I. P., Olubusade, T. J., & Shomuyiwa, A. B. (2025). Risk Management Practices And Financial Performance Of Non-Life Business In Nigeria. Nigerian Journal of Banking and Financial Issues (NJBFI)-Vol.
2. Abass, O. A. , Shomuyiwa, A. O., Olubusade, T. J., & Sulaimon, R. A. (2025). Impact Of Reinsurance Utilisation On The Claims’efficiency Of Non-Life Insurance Companies In Nigeria. Nigerian Journal of Banking and Financial Issues (NJBFI)-Vol.
3. Abdulkadir, J. Enterprise Risk Management (ERM) Practices and Financial Performance: Evidence from Listed Insurance Firms in Nigeria. Yildiz Social Science Review 2024, 10, 99–109. [Google Scholar] [CrossRef]
4. Agboola, A. B. , & Aduloju, S. A. (2025). Nigeria’s insurance industry’s contribution to the nation’s economic growth. International Journal of Entrepreneurship, Management and Social Sciences (IJEMSS), 3026. [Google Scholar]
5. Agboola, O. O. , & Epetimehin, F. M. Microinsurance and microinsurance risk control impact on small and medium enterprises (SMEs) performance in Nigeria. African Scholar Journal of Management Science and Entrepreneurship 2021, 2, 49–67. [Google Scholar] [CrossRef]
6. Ahmad, S. , Karim, R., Sultana, N., & Lima, R. P. (2025). InsurTech: Digital Transformation of the Insurance Industry. In Financial Landscape Transformation: Technological Disruptions (pp. 287-299). Emerald Publishing Limited. [CrossRef]
7. Altomonte, C. , FAVOINO, D., MORLACCO, M., & SONNO, T. M. (2024). Liquidity as Competitive Advantage: The Role of Intangibles.
8. Apostolakis, G. , van Dijk, G., & Drakos, P. Microinsurance performance–a systematic narrative literature review. Corporate Governance 2015, 15, 146–170. [Google Scholar] [CrossRef]
9. Aragani, V. M. Unveiling the magic of AI and data analytics: Revolutionizing risk assessment and underwriting in the insurance industry. International Journal of Advances in Engineering Research 2022, 24, 1–13. [Google Scholar]
10. Aslan, P. (2024, September 3). How Customer Experience (CX) Drives Business Growth in 2025. 30 Examples, 3 September 2025. [Google Scholar]
11. BCG. (2025, January 9). Game Plan for Customer-Centric Growth in Insurance, /: https, 9 January 2025.
12. Capgemini. (2024, December). Capgemini_Top Trends 2025_P&C Insurance, /: from https, 2024.
13. Chennamsetty, V. S. Customer-centric insurance solutions: AI-powered claims processing and fraud prevention. International Journal of Intelligent Systems and Applications in Engineering 2024, 12(23s), 573–583. [Google Scholar]
14. CHILEKEZI, O. (2025). DIGITAL MARKETING OF INSURANCE IN NIGERIA: CHALLENGES AND ISSUES. International Journal of Entrepreneurship and Forensic Management Research.
15. Chin, W. W. (1998). Commentary: Issues and opinion on structural equation modeling. MIS quarterly.
16. Cohen, J. (2013). Statistical power analysis for the behavioral sciences. [CrossRef]
17. Dataforest. (2025, March 31). Digital Transformation Trends in Insurance, Customer Expectations, /: https, 31 March.
18. Dong, F. , & Doukas, J. (2025). The Role of Intangible Assets in Shaping Firm Value. European Financial Management. [CrossRef]
19. Emily, A. (2024, ). Omnichannel Communication in 2025: The Ultimate Strategy Guide. Retrieved from https://ewebinar. 4 September.
20. Etuk, I. U. , Nneoyi, N. A., & Edim, E. J. Factors Influencing Consumers’ Adoption of Mobile Financial Technology (FinTech) Services in Calabar, Nigeria. African Journal of Management and Business Research 2025, 18, 131–153. [Google Scholar] [CrossRef]
21. Exploding Topics. (2025, April 16). Average Customer Retention By Industry (2025), /: https, 16 April.
22. Fornell, C. , & Larcker, D. F. Evaluating structural equation models with unobservable variables and measurement error. Journal of marketing research 1981, 18, 39–50. [Google Scholar] [CrossRef]
23. Gallagher Insurance. (2025, May). Global InsurTech Report - Q1 2025, /: from https, 2025.
24. Gathua, R. N. (2018). Innovation practices and performance of insurance companies in Kenya.
25. Gefen, D. , & Straub, D. A practical guide to factorial validity using PLS-Graph: Tutorial and annotated example. Communications of the Association for Information systems 2005, 16, 5. [Google Scholar] [CrossRef]
26. Goftar, M. N. , Heydariyeh, S. A., & Faezi Razi, F. Presentation of a digital transformation model for customer experience in insurance products (Case study: Asmari Insurance). International Journal of Innovation Management and Organizational Behavior 2024, 4, 44–55. [Google Scholar] [CrossRef]
27. Hekmat, S. N. , Rahimisadegh, R., Mehrolhasani, M. H., & Sirizi, M. J. (2022). Understanding Health Insurance from the Perspective of an Ecosystem: A Systematic Review and Meta-Synthesis. Evidence Based Health Policy, Management and Economics.
28. Henseler, J. , Ringle, C. M., & Sarstedt, M. A new criterion for assessing discriminant validity in variance-based structural equation modeling. Journal of the academy of marketing science 2015, 43, 115–135. [Google Scholar] [CrossRef]
29. IJNRD. (2025, January 1). Enhancing Health Insurance Claim Processing through Artificial Intelligence and Data Analytics, /: https, 1 January 2501.
30. Kolapo, F. T. , Asiyanbi, H. B., & Dada, S. O. (2025). Globalisation and insurance sector penetration: Macro evidence from Nigeria. Journal of Economics, Finance and Management Studies (JEFMS). [CrossRef]
31. Madhani, P. M. (2012). Intangible assets: Value drivers for competitive advantage. In Best practices in management accounting (pp. 146-165). London: Palgrave Macmillan UK. [CrossRef]
32. Madrid, R. L. P. (2014). Innovations in microinsurance: evidence from the Philippines.
33. Maximizer. (2025, April 23). Cross Selling Opportunities in Insurance with CRM: A 2025 Guide, /: https, 23 April.
34. McKinsey. (2023, September 21). Elevating customer experience: A win–win for insurers and customers, 21 September.
35. Méndez-Aparicio, M. D. , Jiménez-Zarco, A., Izquierdo-Yusta, A., & Blazquez-Resino, J. J. Customer experience and satisfaction in private insurance web areas. Frontiers in Psychology 2020, 11, 581659. [Google Scholar] [CrossRef] [PubMed]
36. Mike, H. (2024, October 30). Customer Experience in Insurance: Tips & Examples, 30 October.
37. Oyedokun, G. E. (2024). Assessment of Regulatory Framework for Insurance Business in Nigeria. Lead City University Law Journal.
38. Patin, L. (2024). Customer Retention Strategies in Insurance Agencies.
39. Purnamawati, I. , Jie, F., Hong, P. C., & Yuniarta, G. A. Analysis of maximization strategy intangible assets through the speed of innovation on knowledge-driven business performance improvement. Economies 2022, 10, 149. [Google Scholar] [CrossRef]
40. PwC. (2025). AI in Nigeria – Opportunities, challenges and strategic pathways, /: from https.
41. Samsiah, S. , Nik Muhammad, N. M., Rodiah, S., & Ramayani, C. The Impact of Tangible and Intangible Assets on Financial Performance: Evidence from Private Higher Education Institutions in Indonesia. International Journal of Entrepreneurship and Management Practices 2025, 8 (29), 97-116. [CrossRef]
42. Sarstedt, M. , Ringle, C. M., & Hair, J. F. (2021). Partial least squares structural equation modeling. In Handbook of market research (pp. 587-632). Cham: Springer International Publishing. [CrossRef]
43. Singireddy, S. (2023). Multi-policy synergies: A data-driven analysis of consumer behavior and cross-selling strategies in auto, home, life, and umbrella insurance. Journal for ReAttach Therapy and Developmental Diversities. [CrossRef]
44. Sosa, I. , & Sosa, S. Building an InsurTech Ecosystem Within the Insurance Industry. Risks 2025, 13, 108. [Google Scholar] [CrossRef]
45. Steve, O. C. Digital transformation in Nigeria’s supply chain: Opportunities and challenges. International Journal of Academic Accounting, Finance & Management Research (IJAAFMR) 2025, 9, 36–45. [Google Scholar]
46. Suvashree, B (2024, February 12). 12 Customer Retention Strategies in the Insurance Industry (2025), 12 February.
47. van der Hulst, F. J. , Huijgen, S., Brabers, A. E., & de Jong, J. D. Exploring Trust in Health Insurers: Insights from Enrollees’ Perceptions and Experiences. Journal of Market Access & Health Policy 2025, 13, 29. [Google Scholar] [CrossRef]
48. Wagan, S. M. , & Sidra, S. (2025). Insurance Companies' Activities to Maintain Customer Loyalty in a Competitive Market Environment. Malaysian Journal of Business, Economics and Management. [CrossRef]
49. Wahua, L. , Kenea, G. F. Supervisory Regulation and Performance of Insurance Industry. African Journal of Economics and Sustainable Development 2025, 8, 65–79. [Google Scholar] [CrossRef]
50. Zamharir, M. A. (2025). Digital Transformation and Customer Loyalty in Insurance: The Mediating Roles of Trust, Satisfaction, and Personalization. Retrieved from: https://www.preprints. [CrossRef]
51. Zamharir, M. A. (2025). Untapped Potential of AI in Insurance: A Novel Framework for Enhancing Customer Experience through Personalization, Trust, and Ethical AI Practices Author Information. [CrossRef]
52. Zarifis, A. , Holland, C. P., & Milne, A. (2023). Evaluating the impact of AI on insurance: The four emerging AI-and data-driven business models. Emerald Open Research. [CrossRef]