AI Personalization Boosts $150M Revenue, Cuts Time-to-Market by 30% for Telecom

Industry: Telecommunications, Technology

Client

Telstra: Australia’s largest telecommunications and technology company.

Goal

To transform the client’s customer experience across all channels by driving a consistent AI-first approach in digital, retail, and email, building a standardized, industrial-scale AI engine that delivers significant revenue growth, retention, and improved customer decision quality across client products.

Challenges

  • A lack of translatable evidence of AI’s value made it difficult for the C-suite to justify ongoing investment in data and AI capabilities. Previous implementations were siloed and underpinned by inaccurate or over-complicated academic practices.
  • Marketing and Product teams competed for the same customers using volume-based tactics, resulting in inconsistent customer experiences, offer cannibalisation, and low ROI. AI models remained in experimental phases with no path to permanent implementation.
  • Moving from the POC stage to production required performance-by-design and operational guardrails to build trust and deliver AI at scale.
  • Technical AI delivery was fragmented across consumer teams and slowed speed-to-market. Specialists worked in silos with no central “ways of working” that could scale simultaneously across Mobile, Fixed, and Loyalty divisions.

Solution

Our Fractional Head of AI developed the business case for a flagship “Lighthouse AI Mission” and presented it to the executive team. The proposal outlined the structural changes required to enable AI from data to execution, shifts in data management, and the expected immediate and long-term financial uplift from AI-driven personalization. This secured leadership buy-in across the necessary functional areas to build a mission structure, define remit, and execute.

ML models across multiple dimensions automatically ranked customers daily by likelihood to convert, churn risk, and expected lifetime value across products. After AI models proved their value in the POC phase, ML targeting became the standard top-ranked approach, replacing broad demographic- or volume-based targeting. These auto-segments were closely monitored as part of performance tracking, reducing wasted spend on low-intent customers and increasing conversion rates for priority segments.

A live revenue dashboard was built to compare AI-selected strategies against existing rules-based campaigns, using clear metrics on conversion, revenue per customer, and cost per acquisition. Performance was reviewed daily for BAU strategy optimisation and reviewed fortnightly with executives for significant changes to customer experiences and journeys. Campaigns where AI outperformed legacy logic were automatically prioritised for scale, while underperforming variants were descaled or paused.

Unified a cross-functional team of 20 data science specialists under the Lighthouse mission and standardized how AI features were built and monitored—from data ingestion and transformation through to production. As a result, data scientists spent less time on data engineering and cleaning, accelerating model production by up to 30% and improving the agility, reliability, and effectiveness of AI across customer touchpoints.

Impact:

The program delivered an annual $150M revenue uplift through a scaled, omnichannel AI-driven customer experience across Product, Marketing, and Retail.

Unified ways of working for data and AI, improving speed-to-market for personalized campaigns by 30% through standardization of technical delivery, data ingestion, and feature execution.

Established a high-transparency reporting model that demonstrated AI’s superior ROI compared with traditional marketing methods across key metrics such as conversion, revenue, LTV, and CPA.

Context

A leading Australian telecommunications and technology company sought to transform its consumer customer experience across all channels. The objective was to drive a consistent AI-first approach across digital, retail and email touchpoints by building a standardized, industrial-scale AI engine. The program aimed to deliver significant revenue growth, improved retention and higher-quality customer decisions across consumer products by connecting data, machine learning and execution into a single, repeatable operating model.

Challenges

Marketing and Product teams were competing for the same customers using volume-based tactics, which produced inconsistent customer experiences, offer cannibalisation and low return on marketing spend. Previous AI efforts were siloed, academically complex, or founded on inaccurate practices, leaving models stuck in experimental phases without a path to permanent implementation. A lack of translatable evidence of AI’s commercial value made it difficult for the executive team to justify continued investment in Data & AI capability. Technical delivery was fragmented across consumer divisions: specialists were isolated, there was no central “ways of working,” and delivery could not scale simultaneously across mobile, fixed and loyalty channels. These combined issues created slow speed-to-market, wasted spend on low-intent audiences, and resistance to moving models from proof-of-concept into production.

Implementation

The Fractional Head of AI developed a business case and launched a flagship “Lighthouse AI Mission” presented to the executive team. The proposal focused on structural change required to enable AI from data to execution: shifts in data management, repeatable engineering patterns, operational guardrails, and a clear roadmap for immediate and long-term financial uplift from AI-driven personalization. Leadership buy-in was secured across Product, Marketing, Retail and Technology functions to form a mission-led structure with an accountable remit.

A cross-functional team of 20 data science specialists was unified under the Lighthouse mission. Standardized ways of working were introduced across data ingestion, transformation, feature engineering and production deployment so data scientists spent less time on plumbing and more time on modeling. This standardization accelerated speed to model production by up to 30% and reduced operational friction when scaling models across channels.

Machine learning models were built to automatically rank customers daily across multiple dimensions: likelihood to convert, churn risk and expected lifetime value across products. Once models proved their value in proof-of-concept, ML-based targeting replaced broad demographic and volume-based segments as the primary treatment. Auto-segments were closely monitored to protect customer experience and limit offer overlap.

To deliver transparent, accountable outcomes, a live revenue dashboard was established comparing AI-selected strategies to legacy rules-based campaigns with clear metrics for conversion, revenue per customer, lifetime value and cost per acquisition. Campaign performance was reviewed daily for business-as-usual optimisation and presented fortnightly to the executive team for any material customer experience decisions. Campaigns where AI outperformed legacy logic were automatically prioritised for scale; underperforming variants were paused or descaled. Operational guardrails and performance-by-design principles were applied to ensure production reliability and build executive trust.

Results

The program delivered an annual $150M revenue uplift through a scaled, omnichannel AI-driven customer experience across Product, Marketing and Retail. Standardizing technical delivery, data ingestion and feature execution improved speed-to-market by approximately 30% for personalized campaigns and accelerated model production times. The live reporting model demonstrated AI’s superior ROI compared to traditional marketing across key metrics—conversion, revenue, lifetime value and cost-per-acquisition—enabling continuous prioritisation of effective treatments and rapid shut-down of ineffective spend. Collectively, the Lighthouse AI Mission converted experimental models into dependable, production-grade capability that reduced wasted marketing spend, increased conversion among high-value segments, and established an industrialized path for future AI initiatives.

*Case studies reflect work undertaken by our Heads of AI either during their tenure with Head of AI or in prior roles before they were part of the Head of AI network; they are provided for illustrative purposes only and are based on conversations with our Heads of AI.