- Home
- /
- By your needs
- /
- Communicate results
You have results. Now others need to be able to use them.
The work ends when whoever receives the result can understand it, use it and verify it. This is that final stretch: turning a result into something that can be read, explored, audited and maintained.
A good analysis that is poorly communicated gets lost. Not because the method fails, but because the result stays locked in a script that only its author understands, in a table nobody knows how to read or in a PDF that goes out of date as soon as new data arrive. Rigorous statistical work deserves a finish that matches it.
That finish means moving from a technical result to a deliverable that its recipient (a committee, a reviewer, a manager, a funder) can understand, explore on their own and reproduce. It is not about dressing up the data to make them look better, but about making sure the evidence reaches the decision-maker without losing rigour along the way.
Depending on who will receive the results and what they need to do with them, that finish can take different forms: a reproducible report that documents the analysis and makes it auditable, a visualisation that makes it understandable at a glance, or an interactive application for exploring it. Most projects end up combining more than one.
This is for you if…
You have a finished analysis and need to present it to someone who is not a statistician: a committee, a funder, a clinical team, a board.
You produce the same report by hand every quarter and want to be able to regenerate it automatically, in a controlled way, when the data change, without rebuilding it from scratch.
You need other people to explore the results on their own (filter, segment, compare) without asking you for every cut.
A reviewer or an institution has asked you to make the analysis reproducible and traceable from start to finish, not just a final result with no trail.
How we can help
Not sure which of these fits your situation?
Tell us what stage you are at and we will tell you what work is needed, or whether none is. No commitment, and to the point.
Let's talk about your projectDeliverables
Statistical report
PDF, Word or slides, documented and with the methodology made explicit. The usual format, done properly.
Web report
navigable by section, with the figures and tables built in. Shared with a link, not as a 40-page attachment.
Dynamic dashboard
multidimensional exploration of the results. Whoever receives it filters, segments and compares on their own.
Infographic
a visual summary tailored to the recipient. For when the message has to reach people who are not going to read the full report.
How we have approached it in other projects
Before you take the step
How do I know whether I need a report, a visualisation or an interactive application?
It is decided by the recipient and by what they need to do with the result, not by the technique used in the analysis. This arises when you already have a result and need to get it to someone: a committee, a reviewer, a manager, a funder, a clinical team. The guidance is simple: if the recipient is going to read and needs traceable documentation, a reproducible report fits; if they need to grasp a finding at a glance to support a decision, a visualisation built with sound judgement; if they need to explore the data on their own (filter, segment, compare), an interactive application. Most projects end up combining more than one. You don't need to choose the format in advance; it is enough to know who receives the result and what they need to do with it.
Can a report that I have to repeat at regular intervals be automated?
Yes: a report that is redone periodically can be generated from code so that it is regenerated in a controlled way when the data change, instead of being assembled by hand each time. This arises when you produce the same report every quarter, every survey wave or every update, and the manual work is slow and prone to inconsistencies. The way to do it is the reproducible report: it is generated from code (Quarto, R Markdown) in which every number comes from a traceable line, so that when new data arrive the document is regenerated with the same procedure and without surprises. The limit: automation requires the new data to arrive with a stable structure; if the shape of the data changes, the template is reviewed. The result is a report that is re-run in a controlled way with each update, traceable and auditable.
Does an interactive tool replace the report, or are they different things?
They are different things and often complementary: one serves exploration, the other documentation. This arises when you are unsure whether to deliver an application that the team operates or a report that the team consults. The difference lies in what the recipient does: an interactive application lets them change parameters, filter and recalculate on their own; a reproducible report gives them a traceable document that they consult but do not manipulate. Neither replaces the other (a tool that is explored leaves no documentary record, and a report that is audited does not allow the data to be queried live), and many projects deliver both: the application for day-to-day use and the report for the record. You don't need to decide which one before starting; it is decided according to who uses the result and for what.
A reviewer or an institution requires the analysis to be reproducible and traceable. Is that what you do here?
Yes: reproducibility and traceability from start to finish are central properties of what is delivered, not an add-on. This arises when a reviewer, a funder or an institution asks for the analysis to be re-runnable and for every result to have its trail, not just a final result with no origin. The way to meet it is the reproducible report: an analysis generated from code, with the versions and the environment documented, in which every number comes from an identifiable line and the whole process can be repeated on the same data. That is what makes it possible to meet the requirement with an auditable artefact, not with a claim that the analysis "was done properly". The deliverable is reproducible and auditable according to technical requirements defined in advance; compliance with a formal regulatory framework, where one applies, is a different matter.
Can you work on an analysis we have already done, or does it have to be redone from scratch?
We can work on an existing analysis: if it is well defined, the communication or the tool is built by encapsulating what you already have, without redoing it. This applies to anyone who already has a finished analysis or model (their own or their team's) and needs to turn it into something that others can read, explore or audit. A reproducible report, a visualisation or an interactive application can be built on top of a previous analysis, validating each view against its original results so as not to distort what it says. The honest caveat: moving an analysis into an operable tool is not always just a matter of "wrapping" it (it may require reorganising its logic so that it recalculates or regenerates reliably), but the analysis is preserved. If the review shows that the analysis itself needs to be redone, that is a matter for **I have data and don't know where to start**.
I don't have the results yet, but I want to know what the delivery will look like. Is it too early to talk?
It is not too early: thinking about the delivery before having the results often improves the analysis itself. This arises when you are still analysing data, or even designing the study, and want to anticipate in what form the result will be communicated to its final recipient. Knowing in advance whether the finish will be an auditable report, a visualisation for a committee or a tool that others will explore helps to structure the analysis from the outset so that the finish is possible without rework. This is the natural end of the journey (after designing, analysing, evaluating or monitoring), but knowing its logic beforehand guides the earlier phases. If you are at an earlier point, the conversation can start there: what you want to analyse or what study you want to set up, with the delivery already in view.
Communicating is the end of the journey. If you have reached this point with a result that needs a finish, or if you are still at an earlier stage (with data to analyse or a study to design) and want to know what the delivery will look like, let's talk.
Shall we talk?
Tell us what you are working on and we will give you a straight answer on how we can help.