Banco Santander and its business value of 1,000 million in AI

Table of Contents

Forecasts €1 billion in value from AI

  • Banco Santander estimates it will generate €1 billion in additional enterprise value over the next two years thanks to investments in data and artificial intelligence.
  • The bank wants to increase its customer base from 180 million to more than 210 million by the end of 2028.
  • AI is integrated as a central lever of its ONE Transformation strategy, with a focus on hyper-personalization, commercial productivity, and process automation.
  • Santander has signed an agreement with OpenAI to roll out ChatGPT Enterprise, initially for 15,000 employees.
  • In 2024, AI initiatives would already have generated more than €200 million in savings, with copilots supporting more than 40% of contact center interactions.

Enterprise value with data and AI
– What “enterprise value” means here: in the coverage it is described as a combination of more revenue and lower costs, attributable to investments in data and AI.
– Time horizon: Finextra frames it as additional value over the next two years; in parallel, other coverage based on Reuters frames it as an ambition to exceed €1 billion annually by 2028 (a target, not a result).
– Source of the figure: Finextra (editorial note on Santander’s plan) and Reuters (via TradingView) report the expectation/target in those terms.

Enterprise value generation

Banco Santander has put a number on its bet on artificial intelligence: it expects to obtain that additional enterprise value through investments in AI. According to what was published by Finextra, the expectation is set in the next two years and combines revenue growth and cost reduction. The objective combines two classic impact pathways: boosting revenue and reducing costs.

The institution presents itself as a “global digital bank with branches” and argues that efficiency improvements will come from simplifying products and processes, greater collaboration among global businesses, and scaling common technology platforms.

Enterprise value levers
How enterprise value is usually “made up” in data and AI initiatives (and how it fits with what Santander describes):
1) Revenue (growth)
– Levers: hyper-personalization, better conversion, greater

linkage, better service.
– Signals: more completed digital sign-ups, more products per customer, greater channel usage, better retention.
2) Costs (efficiency)
– Levers: end-to-end automation, process simplification, common platforms, frontline productivity.
– Signals: less time per transaction, less rework/errors, more self-service, more interactions resolved at first contact.
3) Deployment speed (scale)
– Levers: component reuse, data governance, collaboration across geographies.
– Signals: time to production release, adoption by teams, consistency across countries/units.

Expansion of the customer base to 210 million

Santander’s growth plan involves expanding its base from 180 million customers to more than 210 million by the end of 2028. The thesis is that a more digital—and more personalized—experience can accelerate acquisition and engagement, especially if friction is reduced in sign-ups, service, and day-to-day operations.

In parallel, the bank aims to sustain mid-single-digit revenue growth and to reduce total costs every year, with a view to exceeding €20 billion in profit in 2028.

From goal to action
A practical path to connect the “+30M customers” goal with actions (and checkpoints):
1) Acquisition
– Action: segmentation and more relevant messaging (offers/context).
– Checkpoint: acquisition cost and conversion rate by channel.
2) Onboarding (sign-up)
– Action: reduce steps and time; smart assistance for frequent questions.
– Checkpoint: % of sign-ups completed and average sign-up time; main drop-off points.
3) Engagement (recurring use)
– Action: hyper-personalized journeys (alerts, recommendations, next-best-action).
– Checkpoint: monthly activity, products per customer, 90/180-day retention.
4) Service and trust
– Action: faster and more consistent service (copilots + analytics) without losing human control.
– Checkpoint: first-contact resolution, complaints about “misunderstandings”, satisfaction/NPS.

Contribution of artificial intelligence to revenue growth

Santander positions data and AI as a pillar of its ONE Transformation strategy, “fully integrated” into the businesses. The stated focus is on:

  • Hyper-personalized customer journeys, with recommendations and offers better tailored to the context.
  • Frontline productivity (sales and service teams) through AI-assisted tools.
  • End-to-end automation of processes, to reduce time, errorsand manual tasks.

In retail and business banking, the promise of AI is not just efficiency: it is also better conversion, greater use of digital channels, and a more continuous relationship with the customer based on behavioral signals and needs.

Revenue impact indicators
“Grounded” indicators to see whether AI is driving revenue (beyond the talk):
– Hyper-personalization
– ☐ Increases offer conversion (by segment/channel)
– ☐ Boosts activation of digital functionalities (recurring use)
– ☐ Improves retention (churn) in cohorts exposed to personalized journeys
– Front-line productivity
– ☐ Less time per interaction (without quality dropping)
– ☐ More sales/cases handled per hour in commercial teams
– ☐ Fewer escalations and less “rework” due to incomplete information
– End-to-end automation
– ☐ More processes completed without manual intervention
– ☐ Fewer operational errors and fewer repeated incidents
– ☐ Measurable reduction in unit costs per transaction

Agreement with OpenAI to implement ChatGPT Enterprise

In August, Santander reached an agreement with OpenAI to roll out ChatGPT Enterprise. The launch began with an initial group of 15,000 employees, framed within the ambition to become an “AI-native” bank—that is, with AI structurally integrated into operations and decision-making.

The move aligns with an industry trend: moving from isolated pilots to enterprise tools with cross-cutting reach (operations, technology, business, customer service, and internal support).

Internal rollout of ChatGPT Enterprise
– Reported initial scope: deployment of ChatGPT Enterprise for a first group of 15,000 employees.
– Why it matters for “AI-native”: it is not just a customer-facing chatbot; it is often used as an internal tool for search and synthesis, drafting, analysis support, assistance with procedures, and task acceleration across multiple areas.
– Context of the statement: Finextra’s coverage frames it within the ONE Transformation strategy; and Santander has also publicly communicated its collaboration with OpenAI as part of its data and AI approach.

Savings generated by AI initiatives in 2024

According to the bank, in 2024 AI initiatives generated more than 200 million euros in savings (a figure attributed in the coverage to statements by the chief data and AI officer, Ricardo Martín Manjón). The figure suggests that

part of the return is already materializing before the 2028 horizon, especially in automation and operations support.

The data also reinforces the argument that AI is not limited to “flashy” generative AI use cases: savings usually come from process optimization, applied analytics, and automation of repetitive tasks.

Savings from Data & AI 2024
– Reported metric: “more than 200 million euros in savings” in 2024, attributed in the coverage to Ricardo Martín Manjón (Chief Data & AI Officer), a role directly responsible for data and AI initiatives.
– Where those savings typically materialize (consistent with what is described in the article):
– automation of repetitive tasks and reduction of cycle times,
– fewer errors/rework thanks to better information capture,
– assistance to agents/teams to resolve faster.
– Important for interpreting the figure: it is presented as a reported internal result in press coverage; it does not publicly break down what portion corresponds to each initiative.

Use of AI copilots in the contact center

Santander states that AI copilots already support more than 40% of interactions in its contact center, one of the areas where automation and real-time assistance often have a direct impact on response and resolution times.

In Spain, the institution uses Speech Analytics to process 10 million voice calls a year. Among the cited operational effects: the system auto-completes records in the CRM, improves service, and frees up more than 100,000 hours per year for higher-value tasks.

Integrated Customer Service Flow
Typical “copilot + voice analytics + CRM” flow (and checkpoints for it to work in production):
1) Interaction intake
– Call/inquiry enters the contact center.
– Checkpoint: correct identification of the reason (intent) and the customer.
2) Real-time assistance
– The copilot suggests responses, steps, and relevant data during the conversation.
– Checkpoint: the agent validates; if confidence is low, procedural guidance and escalation are prioritized.
3) Speech Analytics and extraction
– The conversation is analyzed to detect topics, compliance, and key data.
– Checkpoint: correct extraction rate (avoid erroneous “auto-fills”).
4) CRM logging
– Auto-completion of fields and a summary of the interaction.
– Checkpoint: quality audit (samples), reduction of rework, and consistency of the record.
5) Operational outcome
– Less after-call time and more hours freed up (such as the +100,000 hours/year cited).
– Checkpoint: real improvement in resolution and satisfaction, notonly in speed.

AI training plan for employees

To sustain the scaling leap, Santander has begun to adapt AI training by role and market. In addition, it plans to launch this year a mandatory AI training plan for the entire workforce, with the aim of standardizing the use of tools and raising internal capabilities.

Training becomes critical in an “AI-native” model: it is not enough to deploy technology; routines, criteria, and ways of working must change across commercial, operational, and support areas.

Training and adoption by roles
“By role” training checklist (what it should cover and how to measure adoption):
– Customer service/contact center teams
– ☐ Use of the copilot (search, scripts, summaries)
– ☐ Verification criteria before logging in CRM
– ☐ Metric: reduction in after-call work without a drop in quality
– Sales teams
– ☐ Interpretation of recommendations (next-best-action) and limits
– ☐ Assisted writing and proposal preparation
– ☐ Metric: conversion and productivity per hour
– Operations/back office
– ☐ Task automation and exception handling
– ☐ Quality control and change traceability
– ☐ Metric: cycle times, errors, and rework
– Technology/data
– ☐ Data best practices, model evaluation, and monitoring
– ☐ Security and access control in corporate tools
– ☐ Metric: incidents, deployment times, and compliance with internal standards

Santander in the AI era

Impact on the banking sector

Santander’s roadmap reflects a broader transformation of the sector: banking is increasingly competing on digital experience, operational efficiency, and the ability to turn data into decisions. In that context, AI becomes infrastructure: a cross-cutting component that affects everything from customer service to back-office automation.

If the bank manages to scale common platforms and standardize processes, it can gain deployment speed and consistency across geographies, a key factor for a global group.

Opportunities and challenges

The opportunity is clear: more productivity, better service, and growth supported by personalization. The challenge is clear too: moving from use cases to sustained transformation, with real adoption by teams, integration with existing systems, and rigorous measurement of impact.

The goal of €1 billion in business value ultimately serves as a litmus test: if AI is integrated into the business, the return should

to be seen both in the income statement and in the customer experience.

Scaling AI: opportunities and challenges
Opportunities vs. practical challenges when scaling AI in a global bank:
– Opportunity: efficiency and recurring savings
– Trade-off: if you automate “too fast,” exceptions and rework increase; data/CRM quality becomes critical.
– Opportunity: better experience and personalization
– Trade-off: personalization requires strong data governance and consistency across channels; otherwise, the customer perceives inconsistent messages.
– Opportunity: team productivity (copilots)
– Trade-off: real adoption depends on training, incentives, and trust; without metrics, usage may remain “curiosity.”
– Opportunity: common platforms and scale
– Trade-off: integrating with existing systems can be the bottleneck; value arrives when the change is reflected in end-to-end processes.

Transformation in the financial sector

Investment in AI as a Growth Engine

Santander’s strategy suggests that investment in AI is no longer justified only by innovation, but by profitable growth: acquisition, greater engagement, and efficiency. The bank aims for AI to be a scale multiplier, not a set of isolated tools.

Personalization and Efficiency Strategies

The combination of hyper-personalization and automation points to a model where the customer receives faster, more relevant responses, while the institution reduces unit costs. In banking, that equation usually translates into improvements in the efficiency ratio and the ability to compete on price and service.

The Future of Banking: A Data-Driven Approach

The underlying message is that the banking of the future will be data-driven: more predictive decisions, more automated processes, and more conversational channels. Santander is betting that this change, well executed, can become a competitive advantage measurable in customers, revenue, and costs.

“Banco Santander and its €1 billion business value in AI” shows how AI stops being a pilot and becomes a measurable lever for efficiency, automation, and a better experience. At Suricata Cx we closely follow these types of deployments because they confirm that real value appears when AI is integrated into omnichannel operations with human oversight and clear metrics, not just in the technology itself.

Scope of this note: the article summarizes what Finextra reported on objectives, figures, and lines of work (ONE Transformation, the rollout of ChatGPT Enterprise, and the cited operational metrics). The reading is

focuses on which specific levers are mentioned (automation, frontline productivity, and voice analytics) and on how they connect to measurable outcomes.

The figures and targets mentioned are understood as publicly shared goals or expectations, not as guaranteed results. Some operational metrics are cited as they have been communicated and may vary or be updated with new disclosures. In corporate AI, the final impact is uncertain and often depends on actual adoption, integration with processes, and data quality.