E-commerce firms that invest in technology readiness and AI-driven personalization report stronger customer satisfaction and better financial results; customer satisfaction appears to be the key channel linking AI adoption to profit growth.

Main Finding

Higher technology readiness (TR) and adoption of AI in marketing (AIM) are strongly associated with better customer satisfaction (CS) and customer performance (CP) in Indonesian e‑commerce firms; CS (and CP) mediate these relationships and translate improvements into superior financial performance (FP). The Sobel test confirmed CS as a key mediator; personalized AI marketing tools (recommendation systems, chatbots, predictive analytics) had a substantial effect on customer engagement and downstream profitability.

Key Points

• Concepts integrated: Technology readiness (TR, via TRI dimensions), AI in marketing (AIM), customer experience (CE), customer satisfaction (CS), customer performance (CP), and financial performance (FP).
• Core relationships:
  • TR → CS (H1): organizations with greater readiness report higher CS.
  • CE → CS (H2): better overall customer journeys raise CS.
  • AIM → CS (H3) and AIM → CP (H4): AIM increases satisfaction and behavioral engagement.
  • CS and CP mediate the path from TR/AIM/CE to FP (Sobel test supports CS mediation).
• Practical recommendations in the paper: prioritize staff/organizational readiness, adopt AI personalization tools, and monitor CS metrics as leading indicators of FP.
• Limitations reported: convenience sampling, demographic concentration (mostly 25–34, male, bachelor’s, small firms), single cross‑sectional self‑report survey—raising concerns about generalizability and common method bias.

Data & Methods

• Design: Cross‑sectional quantitative survey of e‑commerce business owners who are actively using AI tools (chatbots, recommender systems, AI content).
• Sample: 314 valid responses collected online (Google Forms) in August 2024. Target was 334; authors argue 314 exceeds PLS‑SEM '10× rule' for this model.
• Measures: Multi‑item scales for TR (TRI dimensions: optimism, innovativeness, discomfort, insecurity), AIM, CE, CS, CP, FP (FP indicators referenced: profitability, ROA, ROE).
• Analytic approach: Partial least squares structural equation modeling (PLS‑SEM) to estimate measurement and structural models; mediation tested with Sobel test.
• Bias control: Procedural remedies for common method bias (anonymity, randomized item order); single‑source/self‑report and sampling strategy still noted as threats.
• Statistical details provided qualitatively (reported "strong positive relationships") but specific path coefficients, effect sizes, and model fit indices are not reproduced here.

Implications for AI Economics

For firms and managers - Investment in technology readiness (training, leadership support, user onboarding, reducing discomfort/insecurity) increases ROI from AI marketing investments because TR strengthens how AIM translates into CS and FP. - Prioritize AI personalization (recommendation engines, predictive targeting, conversational agents) to lift engagement (CP) and satisfaction—these produce measurable financial gains. - Track customer satisfaction metrics routinely as proximate indicators of the financial returns from AIM initiatives; use CS/CP as intermediate KPIs to evaluate AI deployment effectiveness.

For researchers - The integrated model supports a mediation pathway: TR/AIM → (CE) → CS/CP → FP. Future work should: - Use longitudinal or quasi‑experimental designs to support causal claims. - Employ representative sampling (across firm sizes, demographics, countries) to improve generalizability. - Disentangle measurement overlap between AIM and CE (the paper notes construct ambiguity). - Report and compare effect sizes and cost–benefit analyses of specific AI interventions (e.g., recommender vs. chatbot) to inform microeconomic models of AI investment.

For policy and ecosystem actors - Policies and programs that raise firm-level TR (training subsidies, technical assistance for SMEs, trust & privacy standards) can amplify private returns from AI and accelerate e‑commerce productivity gains. - Supporting measurement infrastructure (benchmarks for CS, standardized FP reporting for digital firms) helps firms and regulators assess the economic impact of AI adoption.

Caveat: findings are context‑specific (Indonesian e‑commerce, convenience sample of early adopters). Apply conclusions cautiously across different industries, firm sizes, and national contexts.