CHAVERY

Solutions and products

Products built the way we build solutions

Chavery builds its own products as well as solutions for others. They show how we think: start from the process people actually follow, and design the system around it.

Featured product

Chavery Research Companion

Technology designed around the research process—not just another collection of disconnected tools.

A research technology platform for postgraduate researchers, their guides and their institutions. It supports the research journey from the first question to the finished document, in one place, so nothing is retyped and nothing is copied between files by hand.

Chavery Research Companion is available now as a web application. Planning, the master chart and data entry are free; the analysis and the finished document are paid, per research project. Institutions begin with a three-month pilot.

Explore Chavery Research CompanionPricing

The application runs separately from this website, at app.chavery.in. This site explains Chavery; the application is where the work happens.

The Master Chart screen of Chavery Research Companion: a table of 60 invented participants with haemoglobin, dosing regimen and age columns.
Master Chart. Synthetic demonstration data: a fictional study with invented participants.

The research journey

One path from question to finished document

  • Learning and research planning

    Short explanations of study design and statistical tests, and a planner that turns a research question into objectives, a design and the variables to record.

  • Study design and structured workflows

    Sample-size calculation for the design chosen, with the power, alpha and effect size recorded so the assumptions can be reported.

  • Master Chart and data organisation

    A living dataset: observations recorded as the study runs, values validated as they are entered, and a locked value amended only with a reason that stays on the record.

  • Statistical analysis

    Assumptions are checked against the data, the test that fits is recommended, the alternative is named when an assumption fails, and the reasoning is shown.

  • Research writing and documentation

    Rule-based checks for style and for statistical reporting, a reference manager, and guide review of sections while there is still time to act on it.

  • Report generation

    One formatted dissertation document with a table of contents, numbered chapters, captions and a bibliography built from the works actually cited.

In the product

What it looks like

Two screens from the running application. The data is synthetic: a fictional study with invented participants, entered for demonstration. No real participant or researcher is shown.

The Master Chart screen: a table of 60 invented participants with haemoglobin, dosing regimen and age columns, and a count of how many cells are filled in each column.
Master Chart. Values are checked as they are entered, and empty cells are shown rather than hidden.
The assumption checks before a t-test: Shapiro-Wilk results and a normal plot for each of two groups, a Levene's test result, and a button to run the test.
Assumption checks. The data decide whether the test suits them, the researcher makes the choice, and it is recorded.

A design decision

Calculated, not generated

Chavery Research Companion does not use generative AI. Every number, recommendation and reported sentence comes from a formula or a written rule, the same input gives the same output, and research data is not sent to an external AI service. That is a design decision for a product whose output is examined; it is not a limit on what Chavery builds for clients.

Explore the product

The pages below describe Chavery Research Companion in detail and carry the policies that apply to it.

Have a different problem?

Discuss Your Challenge

Research Companion is one answer to one problem. If yours is different, we start by understanding it.