Reporting that holds up.

Meridian Analytics turns scattered files and recurring reports into clear dashboards, reliable checks, and a documented refresh process.

Example: An online retailer with eleven staff January to December
This year Last year

Fabricated demonstration data. Use the chart or table to inspect each month.

A fixed price in writing

You own every file

Documentation with every build

Applied statistics behind every number

Try the demos.

Working demos with fabricated data. Open one, change the filters, and inspect the detail.

Choose a business size

The questions a report should answer, and the demo that answers them.

Also in every build

Regional performance map

Every state and every county

Start with the national pattern, select a state, and review its counties. Every view updates with the selected metric.

Open the map →

All example values are fabricated. Nothing shown here is client data.

Ways to work together.

Choose a focused review, a fixed-scope build, or ongoing reporting support.

How an engagement runs

  1. A short note

    Describe the report that is causing trouble. If it is a fit, you get a fixed price in writing before anything starts.

  2. Definitions first

    We reconcile the sources and write down what each metric means before anything gets charted.

  3. The build

    Clear views for leadership, the detail underneath, and quality checks that run with every refresh.

  4. The handoff

    A training session and a written document. The files are yours, and the refresh runs without me.

The cafe demo above was built exactly this way, from three files: a register export, a supplier invoice sheet, and a loyalty list.

Reporting Diagnostic

About two weeks

For teams that know the reporting needs improvement but need a clear starting point. The result is a prioritized plan that any qualified analyst can use.

  • A review of the reports, files, and workflows you use now
  • A written list of what to fix first, what it will take, and what can wait
  • One working prototype of the view that matters most
  • A walkthrough call to go through both

Dashboard Sprint

Four to eight weeks

A fixed-scope build with the price and schedule agreed in advance. You own the result, and your team receives the documentation needed to maintain it.

  • Source data cleaned, reconciled, and documented
  • Written metric definitions
  • Leadership views, with the supporting detail underneath
  • Data quality checks that run with every refresh
  • A training session and a written handoff document
  • Two weeks of questions answered after handoff

Reporting Partnership

Month to month, cancel anytime

For teams whose reporting works but keeps drifting out of date as questions change and systems get upgraded underneath it.

  • Scheduled refreshes and new metrics as they come up
  • Troubleshooting when a source changes shape or a number looks wrong
  • A short written brief for leadership each cycle
  • A standing call to decide what is worth building next

What I will need from you

Each engagement requires read access to the data and one person authorized to confirm metric definitions. A Diagnostic includes a 45-minute user interview, a Sprint includes a brief weekly review, and a Partnership includes a standing check-in. Partnership availability is limited.

Pricing is quoted per engagement.

Final pricing depends on source-data condition, the number of metrics requiring definition, the review group, and training needs. You receive a fixed price in writing before work begins. It changes only if the scope changes. If a request does not warrant a paid engagement, I will say so.

Which one fits?

Three questions, scored right here on the page. Nothing is recorded.

Where is your reporting today?
What would a good outcome look like?
How much of your time can this get?

Built for continuity. Every deliverable includes documentation for the next refresh.

Reporting built to last.

Most reporting problems start before the chart. I fix the definitions, source trail, and refresh process first.

Start with the decision

Define the decision, the audience, and the timing first. Every chart should support a specific action.

Trace every number to its source

Document where each figure comes from, which records it excludes, what can break it, and how far back it can be trusted.

Write the definitions down

Record what counts as a client, a case, or an active month so each metric keeps the same meaning next quarter.

Build the view people will actually open

Keep the main view clear for leadership, with supporting detail available for anyone who needs to verify the calculation.

Make the refresh repeatable

Document the steps and checks so the next reporting cycle can run without relying on the original developer.

The tools I work in

I choose the tool that fits the data and the way your team will maintain the work.

  • Power BI
  • Looker Studio
  • Excel
  • Custom HTML
  • Python
  • R
  • JavaScript
  • React
  • CSS

About Juliette.

Portrait of Juliette Gudknecht

I have spent about four years turning complex data into useful reporting across science, health, communications, and government operations. My work combines applied statistics with practical dashboard design.

I particularly value the point when two systems reconcile and a team can trust the number in front of them.

I am honest about timelines up front. If a project needs more hours than I have, I say so instead of taking it on and delivering late.

BasedTampa Bay region, Florida. Remote anywhere in the US.

TrainingMS in Applied Statistics, from Columbia University. BS in Physics and Data Science.

AvailabilityLimited project availability. One active build at a time.

A firm boundary. I do not take U.S. Department of Education-funded clients, represent clients before federal agencies, work on federal grants, or accept IDEA Section 618, EDFacts, or SPP/APR projects. This practice is entirely separate from my federal employment.

See it working before you commit.

Open a demo, then send a note about the report you want to fix. You get a straight answer on fit, timeline, and cost.

FAQ.

How long does this take?

Each offer has its own schedule range. Delays most often come from data access or unresolved metric definitions. You receive a realistic delivery date before work begins.

Can a non-technical audience use it?

Yes, and that is usually the point. A report that needs a walkthrough before anyone can read it will stop getting opened.

Does the data need to be cleaned first?

No. Cleaning, reconciling, and defining the data is the bulk of the job, and it is included. If your numbers currently disagree across two systems, finding out why is part of what you are paying for.

Do we have to switch software?

No. I recommend changing software only when the current tool cannot meet the reporting need. Otherwise, the goal is to improve the system your team already uses.

What happens once it is finished?

Questions are answered free for two weeks after a Sprint. After that you can move to a Partnership or simply run it yourself. Running it yourself is the outcome I am building toward. Nothing is held back: the files are yours, and any competent analyst can pick them up without calling me.

What about data privacy?

Working copies stay on my machine, never go into third-party AI tools, and are deleted at the end of the project. I will sign your NDA or use a simple mutual one. If your data includes health, education, or other regulated records, tell me early so the handling terms are settled before any file changes hands.

What if our data lives somewhere unusual?

Usually, yes. Spreadsheets, CRM or case-management exports, databases, and PDFs are all workable. Systems with no export capability require an early feasibility review before either party commits.

What work is outside scope?

The full list is under About. The short version: nothing federally funded, and nothing touching the special education data collections I work on in my day job. It is a firm line rather than a preference, and you will hear about it in my first reply rather than partway into a project.

Tell me what your data reporting needs.

A few details are enough for me to assess the fit, timeline, and likely cost.

  • What report is causing trouble?
  • What data sources are involved?
  • Who uses the report?
  • When is it needed?
(813) 279-1982

Email is the fastest way to reach me. I reply within two business days.