Microsoft Energy Global Company newsroom

Repsol

Document Processing

Repsol found that everyday knowledge-work tasks such as summarizing content, pulling key information, and managing emails were consuming time that employees could otherwise spend on higher-value strategic work, so it wanted to improve productivity and work quality with new AI-enabled ways of working.

121 minutes per week saved per average employee
16.2% boost in output quality
25,000 employees targeted for rollout
550 employees in the pilot study

Reported impact

Business outcome

Repsol reported measurable productivity and quality gains from the pilot, along with stronger employee adoption, improved confidence and creativity, and a planned expansion from the pilot to enterprise-wide deployment.

  • Productivity Improvement

Solution

Repsol worked with Accenture, Microsoft, and Avanade to deploy Microsoft 365 Copilot across the organization through an early access program and validate the rollout with a 550-person pilot. The implementation focused on day-to-day information work rather than a standalone AI app: employees record meetings, then use Microsoft 365 Copilot to extract tasks, deadlines, and responsibilities from that meeting content. Users continue the workflow in Word, where prompts turn notes into structured summaries, and then generate presentation drafts in PowerPoint. Finance teams also use Copilot to analyze RFP responses and create RFP content from scratch. The source does not describe a separate storage or model-hosting layer; artifacts remain embedded in Microsoft 365 work products such as meeting notes, summaries, presentations, emails, and RFP documents, with training, support, incident handling, and change management wrapped around the deployment.

Data flow

Employees capture meeting content and other work materials such as emails and RFP responses, then Microsoft 365 Copilot extracts key tasks, deadlines, and responsibilities. Those outputs are refined in Word into summaries and turned into PowerPoint drafts, while finance teams use the same Copilot-assisted workflow to analyze incoming RFP responses or generate new proposal content.

Solution architecture

3 components · 1 layer
  1. Serving
    • Microsoft 365 Copilot Generative AI assistant used to extract tasks from meetings, summarize content, support email work, and help finance teams analyze or draft RFP materials.
    • Word Authoring surface where employees prompt Copilot to turn meeting notes into clear written summaries.
    • PowerPoint Presentation surface used to turn Copilot-assisted summaries into draft slide decks.

Architecture clues

  • Deployed through the Copilot for Microsoft 365 early access program
  • Future roadmap includes new AI agents for planning, forecasting, application maintenance, and incident resolution
  • Pilot ran for four months of experimentation, testing, and analysis
  • Rollout supported by training, community-building, technical incident management, and change management

Evidence from the source

"After four months of experimentation, pilot testing, and analysis... the average employee was saving 121 minutes per week."
"Before the transformation, Repsol recognized that everyday tasks like summarizing content, pulling key information, and managing emails took up time that could be better used on more strategic work."
"Looking ahead, Repsol is set to expand Copilot to all 25,000 employees, starting with an initial phase involving 3,000 users."
"Now, thanks to generative AI, employees can record meetings and ask Copilot to pull out key tasks, deadlines, and responsibilities."
"Repsol's AI-driven workflows boosted output quality by 16.2%"
"Then, with just a few prompts in Word, they can transform those notes into a clear summary and then draft a PowerPoint presentation."
"When the chance to join the Copilot for Microsoft 365 early access program came along, Repsol jumped at the opportunity."
"teams in finance can better analyze responses to Request for Proposals (RFPs) or even create them from scratch."
Transparency Evidence & methodology 100/100 reference value · 94% confidence · 1 source

Review how this reference was scored and trace each normalized claim back to its documented source.

01

Scoring methodology

How scores are calculated

Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.

Customer reference value100/100
  • Problem clarity20/20
    Measures whether the customer's operational problem is specific enough to compare.
  • Implementation depth25/25
    Rewards named components, data flow, integrations, and architecture detail.
  • Outcome specificity20/20
    Measures whether the reported benefit is concrete rather than a generic claim.
  • Quantified impact25/25
    4 metrics describe measurable impact.
  • Deployment scale10/10
    Rewards evidence of a live or scaled deployment and meaningful integration scope.
Evidence confidence94%
  • Source authority25/25
    The deployment is described by the adopting organization or a government source.
  • Claim traceability21/25
    10 of 12 key fields link to a named source.
  • Direct evidence20/20
    8 direct source quotes retained.
  • Date provenance10/10
    The publication date comes from the source page or publisher.
  • Source transparency8/10
    1 source reference document publisher, URL, authority, retrieval, and supported fields.
  • Implementation specificity10/10
    Rewards source-backed components, data flow, quantified impact, and a specific outcome.
02

Traceability

Sources & provenance

Field coverage shows which facts each source supports. Dates are shown only when documented by the source.

  1. Repsol

    Repsol customer story

    First-party
    https://www.accenture.com/en/case-studies/energy/repsols-workforce-now-runs-ai-fuel
    Source type
    Customer-published story
    Published
    Mar 13, 2025
    Supports
    • Customer
    • Business problem
    • Solution
    • Products
    • Solution components
    • Data flow
    • Integrations
    • technical area
    • use case category
    • Business outcome
    • outcome category
    • Metrics
    • Architecture
    • evidence quotes
    • Publication date

Date evidence

  • Published on source page Mar 13, 2025 · Accenture · First-party
    Open date source
Not separately sourced Industry, Region, Company size. Treat these normalized attributes as lower-confidence until a dedicated source is added.