Data Health Check

You have the data. You don't have the answers.

Data spread across tools that never talk to each other  Â·  Reports that disagree  Â·  A simple question that takes two spreadsheets to answer

The Data Health Check is a three-week audit that tells you exactly what your data can support, what it cannot, and what to build first.

Start Here
From 2,700 EUR. No call required.

  • What is the Data Health Check?

    Data Health Check is a three-week, fixed-scope audit of the data your business already holds. At the end you have a written picture of what data exists across your systems, what it can and cannot answer, agreed definitions for the metrics that matter, and a prioritised plan for what to build.

  • Who is this for?

    Businesses that run on more tools than they can join up. Metrics that are assumed rather than agreed. Decisions that stall because nobody quite trusts the number underneath them.

    From a fifty-person operation running an ERP, a CRM and a warehouse system, to a one-person business running several online stores, an invoicing tool and two advertising platforms. Scale is not the point. If the complete picture has to be assembled by hand, this is for you.

If you already have reporting in place, three questions:

1

Do you still add numbers across reports to get a complete picture?

2

Do the same metrics mean different things in different departments?

3

Are there data sources nobody uses, because nobody is sure what is in them?

The Data Health Check answers all three in writing. Not by building something completely new, but by establishing what you already have, what it can support, and what each number officially means.

It is probably not for you if:
  • You know which dashboards you want to build and why
  • Your data is already connected and gives you the complete picture
  • You cannot commit time to granting access and quick alignment

What do you actually get?

Three full documents, one summary, each covering a different part of the picture.

Data inventory and health

For whoever owns the systems.

For every system on your list:

  • What it is and what your business uses it for
  • What reportable data it holds
  • How that data can be extracted: API, scheduled export, manual export, or not at all
  • Whether that extraction is sustainable or means someone doing it by hand every month
  • What is clean, what is inconsistent, what is missing, what is broken
  • Whether it can be joined to your other systems, and on which key

And across all of them: how data currently flows, which systems can talk to each other and which cannot, and every problem I found, named in writing.

Metric definitions

For anyone who needs a number to mean the same thing in every report.

  • The metrics that matter for your business, with the reasoning for each
  • A precise definition per metric: what it counts, what it excludes, over what period, from which source
  • The business question each one answers
  • How they relate to each other
  • Which are available today, and which need something to change first

This document is agreed, not proposed. It comes out of a definitions meeting and is signed off. That is what turns it into a standard rather than a suggestion.

The roadmap

For whoever decides what to build, and what to leave alone.

  • The business questions your data can answer, in priority order
  • What can be built now with the data as it stands
  • What would become possible with a specific change: an API tier, one added field, a process fix
  • What you could reasonably do yourselves, and what needs help
  • What is not worth doing

From scattered data to a clear inventory, agreed definitions, and a plan for what to build. Three weeks.

Who you would be working with

Geta Viasu-Räisänen, founder of Light On Analytics.

I am Geta Viasu-Räisänen, Geta for short. I have worked in data and business intelligence since 2016, as a BI and CX analyst for Sky Deutschland and for Personio, one of Germany's tech unicorns. In 2025 I founded Light On Analytics, to help SMEs understand and use their own data.

I have worked with data at every stage: building data sources, fixing what was broken, writing queries, and designing dashboards people actually used. My own tool of choice is Klipfolio, where I am one of a small number of certified partners worldwide. Enough systems, in enough states, to know exactly what to look for.

You work with me directly. No agency, no account manager, no third party. I am on every call and I do all of the work. The person who reads your systems is the person who writes the documents and sits across from you at handover. Ten years of doing this, condensed into three weeks.

How much does it cost?

Up to 3 data sources
2,700 EUR
4 to 6 data sources
3,700 EUR
7 or more
Book a call

Payment in full at purchase. Nothing open-ended, no scope creep, no hourly surprises.

How long does it take, and when does it start?

Three weeks of work. The clock does not start at purchase.

1

You buy. Your calendar slot is blocked immediately.

2

You have 10 days to grant access to your systems. I tell you exactly what each one needs.

3

The three weeks begin the day the last access is in place, starting with a session on your data, your tools, your business and how you use it.

What I need from you

Before the clock starts:
  • A named contact for questions about the business and its systems
  • Access to each tool on your list, at whatever level that specific tool requires
  • A short conversation with the people who actually use each tool
During the three weeks:
  • Short working sessions, spread across the three weeks
  • Replies within two working days
  • Access to the people who can agree the metric definitions
What you do not need:
  • Clean data. Finding what is broken is part of the delivery.
  • To know what you want built. That is the output, not the input.

What if my data turns out to be unusable?

Then that is the finding, and it is a valuable one.

"Your data cannot currently support the reporting you want, and here is specifically what would have to change" is a legitimate outcome. It is also considerably cheaper to learn in three weeks than six months into a build.

The Data Health Check has to be worth buying on its own. If these documents were only useful as a lead-in to hiring me for something bigger, this would be a paid sales call, and you would be able to tell. If you never buy anything from me again, you should still be better off for having done it.

What happens after the audit?

You own the documents, and what happens next is your decision.

If you want the roadmap's top five questions actually built, that is the Dashboard Sprint: fixed price, fixed scope, and a separate decision made afterwards with the roadmap already in your hand.

See the Dashboard Sprint

Ready to start?

Three questions. They set your price and tell me whether I can do a good job for you.

I read each form personally before accepting. If something in your setup would make this a bad use of your money or my time, I will tell you before you pay, not after.

FAQs

Do I need to use a specific tool?

No. The Data Health Checki s tool-agnostic. It looks at the data you have, wherever it lives, and the roadmap tells you what would be worth building and what would suit it. Nothing in the audit commits you to a platform.

Is this just a sales call in disguise?

No, and the roadmap is where you can check. It includes what is not worth doing, and what you could reasonably do yourselves without hiring anyone. A document that recommends everything is a pitch. This one is designed to rule things out.

What if I already have a data analyst?

Then this still works, and often works better. An analyst does not settle the question of what a metric officially means across departments. The definitions document does, because it comes out of a meeting where it gets agreed and signed off.

Can I get a refund?

Within three days of purchase, in full. From day four until the clock starts, half. Once the work has started, no. Before any refund I will offer you a free reschedule, because most cancellations this early are a timing problem rather than a change of mind.

Why is it priced by number of tools?

Because that is the only thing either of us can know before the work starts. The real cost drivers are how data comes out of each system and how messy it is, and neither of us can see that yet. Source count is the fairest available proxy, and fixing the price means the risk of it being harder than expected sits with me, not you.

How is this different from a dashboard project?

A dashboard project builds something. This tells you what is worth building. Doing them in this order means the build is scoped against what your data can actually support, rather than against what everyone hoped it could.

Three weeks. Fixed price. A written answer.

You will know what data you have, what it can answer, what each number officially means, and what to build first.

Start with three questions