Turning operational data into business insight.

I combine technical service experience with data preparation, KPIs and dashboards to understand operations and support business decisions.

Explore analytical cases ↗View CV ↗

Technical Service Supervisor, IberiaJohnson & Johnson Vision

Data tools in use

Analytical cases

Applications built

Years leading technical service

01 / SELECTED ANALYTICAL CASES

Questions, data, decisions.

Three examples of analysis applied to service operations. Explore the business question, the data work and the decision each case supports.

Install Base — single source of truth

Synthetic data · Figures obscured

How do you turn equipment and contract records held in separate enterprise systems into one source the service organisation can plan from?

Data work: Consolidate the sources in Alteryx, reconcile records against each other and check coverage and data quality before anything reaches a dashboard.
What it changed: One install base view for EMEA showing what is installed, where, and which contract covers it — and surfacing records that disagree between systems.

SQL
Python
Alteryx
Tableau

Explore the case ↗

Cost per System

Synthetic data · Figures obscured

Where does service cost actually sit, once you can compare it by equipment, service type and period?

Data work: Prepare cost data in Alteryx, then compare labour, travel and spare-part cost across platform, country, cluster and contract type.
What it changed: A cost-per-system view for EMEA showing which platforms and which service types drive cost, and how that moves year over year.

SQL
Python
Tableau

Explore the case ↗

Contract Profitability — ASP

Synthetic data · Figures obscured

Which maintenance contracts are actually profitable, and on which platforms and for which customers?

Data work: Bring contract, revenue and cost together into comparable metrics, including average sales price per contract type.
What it changed: The ASP analysis is used to reprice maintenance contracts: it shows which contract types return least against their cost, by platform and by customer. A companion dashboard flags contracts approaching renewal for the sales team.

SQL
Python
Tableau

Explore the case ↗

02 / CAPABILITIES

From raw data to clear decisions.

Data preparation & quality

Clean and transform data, combine sources, handle missing values and duplicates, and create derived variables and datasets.

SQL
Python
Alteryx

Descriptive & exploratory analysis

Compare trends and segments, calculate metrics and examine data quality to answer operational questions.

SQL
Python
Excel

Visualisation & communication

Build KPIs, dashboards and clear summaries that connect analytical findings with business decisions.

Tableau
Power BI
Excel

My approach: understand the question, prepare the data, analyse and validate, then communicate the findings. I use AI assistance for coding and review the outputs against source data and business rules.

-- the question: what is covered, and what is exposed, by platform
SELECT   c.cluster,
         c.platform,
         count(*)            AS contracts,
         sum(c.annual_value) AS exposure

-- prepare the data: one source, joined on the asset
FROM     contracts    AS c
JOIN     install_base AS b USING (asset_id)
WHERE    c.status = 'active'

-- analyse
GROUP BY c.cluster, c.platform

-- validate: drop nothing silently, keep only what reconciles
HAVING   bool_and(b.reconciled)

-- communicate: biggest exposure first
ORDER BY exposure DESC;

The same method, written out. Invented schema.

The experience behind the analysis.

OPERATIONS · PEOPLE · DATA

I don’t analyse service operations from the outside. I run them.

I coordinate technical service at Johnson & Johnson Vision in Spain and Portugal, working with teams and ophthalmic equipment.

This operational perspective shapes the questions I bring to data: how to understand service costs, follow contracts and make information useful for planning. My work combines data preparation, descriptive analysis and visualisation.

That is the difference. I know which questions the data has to answer before I build anything, and I can tell when a number is wrong because I know what the operation behind it looks like.

SELECT operational_data,
       technical_experience,
       data_prep + kpis + dashboards AS business_insights
FROM   "Miguel Castillo"
WHERE  goal = 'Better Business Decisions'
  AND  coffee_level > 0;

View CV LinkedIn ↗

CAPABILITIES IN PRACTICE