> For the complete documentation index, see [llms.txt](https://docs.nested-knowledge.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.nested-knowledge.com/review-workflow/tagging/configure-tagging.md). # Configure Tagging Learn how to configure your tag hierarchy for data extraction. Configure tags in a tag hierarchy to set you up for data extraction. Your setup determines the reviewer extraction process, how well AI tools perform and the structure of exports. A tag hierarchy is built from root tags — broad, folder-like concepts — that break down into parent and child tags as specific as your protocol needs. Everything captured in Tagging/Extraction, and analyzed later in Synthesis Outputs, depends on what this hierarchy defines. You can change your tagging mode, add tags, or introduce automation at any point in a review.

What is a tag?

In Nested Knowledge, a tag isn't just a screening label. It is also the core unit of data capture in a nest: either a single datapoint, or a container for a set of datapoints you extract from a study's abstract or full text. The shape you build is the shape your outputs take, so the hierarchy should reflect the information you intend to extract and report. The mental model: question tags are the column headers of your exported dataset and input of cells are child tags and/or evidence, where each study is a row.

Example tag hierarchy with root, parent, and child tags — Tag hierarchy

Spreadsheet with tags represented as column headers — Tags represented as column headers in a spreadsheet

{% hint style="info" %} **Root** tags (your most overarching concepts) break down into **parent** and **child** tags for detail. e.g. Root = Study Characteristics; Parent = Location; Child = USA. {% endhint %}

When are tags used?

Tagging has no fixed start point in a review. You can create and apply tags as soon as records are in your nest. Both stages in the table below use the same Tag Hierarchy and the same tools described in this module — there's no separate "screening tags" system. Most reviews use tags twice, in two different ways: | Stage | What most reviews do | Typical tools | | ------------------------------------------------------------------------------------------------------ | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------- | | [During Screening](/review-workflow/screening/configure-screening.md#optional-apply-tags-in-screening) | A light pass: quick manual tags for categorization, often alongside extraction of PICOs, Study Type, Location and Size — useful as a preliminary filter well before Screening finishes | Manual tags, Core Smart Tags | | Once studies are included | A fuller pass: a purpose-built hierarchy, extracted to capture everything for final exports — tables, reports and insights | Manual tags, Smart Config: Tagging, Adaptive Smart Tags |

{% stepper %} {% step %} ### Choose Tagging Mode Toggle in [Nest Settings.](/your-nests/nest-settings.md)

Form-Based Tagging (default)

This default mode presents each tag as a question with answer options, supporting richer content — tables, preset option lists, numeric values — and the specific question wording that Adaptive Smart Tags needs to extract accurately.

Form-Based Tagging questions grouped by root tag — Form-Based Tagging groups questions by root tag

In Tagging, presented questions are grouped in the form by root tag. If you'd prefer one single form of questions toggle this form group mode in Nest Settings.

Standard Tagging

This mode presents tags as a lighter dropdown selection instead, without the question-and-answer structure, suited to reviews that only need to mark whether a concept is present.

You can switch between the two at any point without losing existing tag data. If you're planning to use Smart Config: Tagging or Adaptive Smart Tags, **Form-Based Tagging** is the better fit — both are built around tags read as questions. This guidance page uses this mode, follow the same instructions for Standard mode minus the question configuration. {% endstep %} {% step %} ### Decide How To Build Your Tag Hierarchy Nested Knowledge supports four ways to put a hierarchy in place, and they aren't mutually exclusive — combine and edit them freely in the same nest. See [Configure Tag Hierarchy](#configure-tag-hierarchy) below for step-by-step instructions on tag creation.

If you want to…	Choose	Compatible with AI Extraction?
Reproduce a data extraction spreadsheet	Import Tag Hierarchy Templates then create/edit tags yourself. Also see Match Tags to your Export Spreadsheet	Yes — with Adaptive Smart Tags
Get standard hierarchy of PICOs, Study Type, Size & Location tags extracted immediately from abstracts, often early in your review	Core Smart Tags	Yes — with Adaptive Smart Tags after edits are made
Get a custom multi-level hierarchy tailored to your research question, often for extraction of included studies	Smart Config: Tagging	Yes — with Adaptive Smart Tags
Build entirely from scratch	Configure Tag Hierarchy below	Yes — with Adaptive Smart Tags

Options for building a tag hierarchy — Choose a method for building your tag hierarchy

{% endstep %} {% step %} ### Apply Tags Once your tag hierarchy is set, use [Adaptive Smart Tags](/review-workflow/tagging/ai-tools-in-tagging/adaptive-smart-tags.md) for AI extraction of created tags, apply tags manually in **Tagging** module, or both. See [Apply Tags.](/review-workflow/tagging/apply-tags.md) {% endstep %} {% endstepper %} ## Configure Tag Hierarchy The Configure Tagging page is where you create, edit, and organize the tags that capture your review data. Use it to reproduce an extraction spreadsheet, add concepts that AI tools do not cover, or refine an existing hierarchy. ### Create Tags Create tags manually when you need precise control. You can also start with [Core Smart Tags](/review-workflow/tagging/ai-tools-in-tagging/core-smart-tags.md), [Smart Config: Tagging](/review-workflow/tagging/ai-tools-in-tagging/smart-config-tagging.md), or [import tag hierarchy templates](/review-workflow/tagging/configure-tagging/import-tag-hierarchy-templates.md). {% stepper %} {% step %} ### Open Configure Tag Hierarchy In the navigation menu, select **Configure Tagging** .

Configure Tagging navigation option — Open Configure Tagging from the navigation menu

{% endstep %} {% step %} ### Add New Tag Select **New Tag**, the blue circle icon or start typing a Tag Name. If you have existing tags, drag and drop the blue circle icon to create a new tag and attach them. {% endstep %} {% step %} ### Add Root Tags Create root tags for major, distinct sections of extraction. Think sheets in a workbook. Root tags require only a Tag name, this is how it'll also appear in exports.

Root tag created in the tag hierarchy — Create root tags for major extraction sections

{% endstep %} {% step %} ### Add Question Tags Drag a tag to move it in the tree. Drop it onto another tag to make it a child. Create question tags and, for those with fixed answers, answer/sub-tags. When a question type is selected, the tag becomes a circled Q icon. {% endstep %} {% step %} ### Choose A Question Type Option 1: Choose **Apply** for an answer entered on the tag. For example the question tag: **Study Objective.** Input a question. No child tags created.

Single Apply question tag configuration — Configure a Single Apply question tag

Option 2: Choose **Select** for answers stored as child tags. For example the question tag: **Study Design**, with child tags: RCT, Observational, Case Study. Like root tags, child tags require only a tag name, which is how they'll appear in exports. Turn on **Multiple Responses** when more than one child can apply.

Configure a Select question tag with child-tag answers

See below for more on [question types](#choose-a-question-type-1) and how to decide what to use. {% endstep %} {% step %} ### Configure Tag Contents By default all tags extract evidence as text. You can configure content mode to better format and extract the content you wish to export. See below on [content type](#choose-your-tag-contents) for more. {% endstep %} {% endstepper %} ## Format Tags ### Choose A Question Type Choose the question type that matches the answer you need: * **Single Apply** captures one answer on the tag. * **Single Select** requires one child-tag answer. * **Multi-Select** allows several child-tag answers. Use **Single Apply** with text options for one fixed label, no attached text annotation. Use **Select** or **Multi-Select** when answers need their own notes, evidence, or data.

Question type options for a tag — Choose a question type for each tag

### Choose Your Tag Contents A **Single Apply** tag captures one answer. Choose the content type that fits that answer: * **Text** or **Numeric** for one value. * **Text Options** for one choice from a fixed list. * **Table** or **Index Table** for related structured data. Tables can be created and combined together in your exports. Select the content type you wish to configure for step-by-step instructions.

Text (default)

Free text is the default content type for all tags — a reviewer reads the study and records the answer in their own words, with the source excerpt kept alongside it. Use it for anything open-ended, or when you want the flexibility to capture nuance a fixed list or number can't. See below to link tag in hierarchy to format for data extraction to format in export. **Tag in hierarchy**

Text tag in the tag hierarchy — Text tag in the hierarchy

**Format for data extraction**

Text field used during data extraction — Text field during data extraction

**Format in export**

Exported text-tag data — Text-tag data in an export

Text Options

A fixed, pre-configured list a reviewer picks from instead of typing free text. Only the options you've configured appear in the dropdown, and exactly one can be selected. Selecting an option records that label directly, in place of a text excerpt — there's no separate annotation to add. Use it as the default for any fixed list where a study gets one answer — it exports as a single, clean column. Switch to a Select tag with child tags only if a study can have more than one answer, or each answer needs its own evidence or follow-up questions; see Choose A Question Type. See below to link tag in hierarchy to format for data extraction to format in export. **Tag in hierarchy**

**Format for data extraction**

Text Options dropdown during data extraction — Text Options during data extraction

**Format for export**

Exported Text Options data — Text Options data in an export

Table

A configured grid attached to a tag, for collecting multiple rows of the same kind of data on one study — best when the number and identity of rows aren't predictable ahead of time, so a fixed set of separate tags won't fit. Select **Table** as the tag's content mode, then select **Configure Table Columns** to add the fields you want: drag and drop to reorder them, select the pencil icon to edit a column's header, or the trash icon to remove one. Leave a column as free text, or give it a preset list of options the same way you would on a Single Apply tag. While tagging, the table adds new rows automatically as a reviewer fills in more data — there's no fixed row limit to plan around. The tag icon forms a square when table content is enabled. **Tag in hierarchy**

Table content configuration for a tag — Table content configuration

**Format for data extraction**

Table used during data extraction — Table during data extraction

**Format in export**

Exported table data — Table data in an export

Index Table

A two-layer structure for stratified data: a fixed set of **index columns** (the "spine") that every row shares, with one or more **sub-tables** attached underneath, each capturing a different domain against that same spine. An arm identified once in the index carries its identifying columns into every sub-table entry against it, so it isn't re-entered for each domain. Unlike a plain table, an index table's row structure is fixed once you're extracting — there's no adding an extra index row, or an extra sub-table row beyond what the index defines, mid-extraction the way a plain table auto-adds rows as you go. Set up every index row you'll need — one per study arm, typically — before you start filling in sub-table values against them. Keep the index itself lean, use text options for index columns wherever the values are a fixed, short list, and design each sub-table around what you actually want to see in your export. The tag icon forms a square labelled with "i" when index table is enabled. **Tag in hierarchy**

Index Table configuration for a tag — Index Table configuration

**Format for data extraction**

Index Table sub-table during data extraction

Completed Index Table during data extraction

**Format in export**

Exported Index Table data — Index Table data in an export

Numeric

A single number, digits only — for a value like total sample size that you'll want to sort or chart rather than read as text. **Tag in hierarchy**

**Format for data extraction**

**Format in export**

Exported numeric-tag data — Numeric-tag data in an export

### Match Tags To Your Export Spreadsheet Start with the columns you need in your final export. Create root tags for major sections, parent tag questions for your column headers then add child tags for each field where needed. Use a table when one study can contain repeated rows, such as outcomes, timepoints, or treatment arms. See [Match Tags to your Export Spreadsheet.](/review-workflow/tagging/match-tags-to-your-exports.md) {% hint style="info" %} Bibliographic data usually imports with each record. You don't need separate tags for it unless this data should be structured in another way. See [Export](/review-workflow/exports/custom-tables.md) for available bibliographic columns. {% endhint %} ### Write Questions For AI For **Smart Config: Tagging** or **Adaptive Smart Tags**, write each tag question as an instruction. **What was the intervention?** gives the AI an extraction task. **Intervention** does not. Add guidance when the expected answer could be ambiguous. See [Prompt guidance.](/ai-transparency/ai-in-nested-knowledge/prompt-guidance.md)

Can I use an AI chatbot to plan my tag hierarchy from my spreadsheet?

Yes. A general AI chatbot, such as Claude or ChatGPT, can read your data extraction spreadsheet and propose a hierarchy, tag types, and AST questions. You still build the hierarchy in Nested Knowledge yourself and review every question before extraction. Check your organization's policy before uploading client or confidential spreadsheets to an external tool. Copy the prompt below, fill in the bracketed parts, and attach your spreadsheet. > I'm building a tag hierarchy in Nested Knowledge (NK) so its AI extraction tool, Adaptive Smart Tags (ASTs), can extract study data into the attached data extraction spreadsheet. NK can't import a spreadsheet as a hierarchy, so I need a plan I can build by hand. > > First, read these NK documentation pages: > > * Configure Tagging (tag types and how they export): > * Match Tags to your Exports: > * Adaptive Smart Tags: > * Prompt Guidance (how to write AST questions): > > Then, for each of these sheets: \[list sheets, or "every sheet except ..."] > > 1. Propose a root tag. > 2. Map each column to one tag type: Text, Numeric, Text Options, Select with child tags, Table, or Index Table. Use an Index Table when several groups of columns share the same identifying columns, such as arm or intervention: name the index columns and group the rest into sub-tables of related columns. Use a Table for repeating rows of one kind with no shared identifying columns. > 3. Write the AST question for every tag, and a header that stands on its own for every table column. Follow Prompt Guidance: define the concept, state the units or summary statistic, and say what to return when data isn't reported. > 4. Show the hierarchy as an indented outline, in the order tags should appear. > > Ask me before assuming anything the spreadsheet doesn't make clear. Plan the hierarchy only; don't extract data.

{% hint style="info" %} [**Copy and paste tags**](#copy-and-paste-tags) to avoid manual creation of similar tags. {% endhint %} ## Navigate Tags ### Reorganize Your Hierarchy Drag a tag to move it. Drop it next to/under another tag to make it a child. Drop it between tags to reorder it. Moving a tag doesn't remove existing answers. The order of tags, left to right and top to bottom, is the order displayed in the Tagging form/drop-down list when you're ready to extract. ### Edit Existing Tags Select a tag to open its settings. You can change its name, question, content type, description, and parent. Changes apply going forward. They don't erase applied data. After changing a question type, check existing answers. ## Toolbar As you build and tag, a set of tools on the toolbar help you keep the hierarchy clean and accurate. Each has a specific job. By default you'll land on the Edit page.

Copy and paste tags

Copy a tag and its child tags to reuse part of your tag hierarchy in the same nest or across nests. 1. On **Configure Tag Hierarchy**, select **Copy & Paste**. 2. Select the tag to copy. 3. Copy the tag with the appropriate shortcut. 4. Select a destination tag. 5. Paste the copied tag with or without its child tags. If you don't select a destination tag, the copied tag becomes a root tag. | Function | Mac | Windows/PC | | -------------------------------- | ------------------- | ---------------------- | | Copy selected tag and child tags | `⌘` + `C` | `Ctrl` + `C` | | Paste copied tag only | `⌘` + `Shift` + `V` | `Ctrl` + `Shift` + `V` | | Paste copied tag and child tags | `⌘` + `V` | `Ctrl` + `V` |

Copy and Paste controls in Configure Tag Hierarchy

Examine applied tags at a glance

To inspect a tag alongside its extracted contents, select the **Examine** icon — most useful after Core Smart Tags or Adaptive Smart Tags have run, so you can review what was captured straight from the hierarchy. Select a tag, start typing a tag name, or step through tags with the arrows. A selected tag shows a snapshot of the studies reporting it. By default the **recursive** option is on, so the contents include the tag and any child tags (most relevant for tags higher up the hierarchy). A shortcut to Study Inspector is provided with the tag filter already applied, and selecting an individual tag’s contents opens that study for editing.

Examine view showing a selected tag and its applied contents — Examine applied tag contents

{% hint style="warning" %} Examine tags will only be populated once tags are applied. {% endhint %}

Hide tags in Synthesis Outputs

If your hierarchy contains tags you don’t want shown in Synthesis — admin or organisational tags, say — you can hide them. They’re hidden **only** in Synthesis and remain usable everywhere else in AutoLit. Select the **Hide** icon and click the tag; selecting a parent hides its children too, and you can hide multiple tags at once. Hiding only changes the tag’s shape (to distinguish hidden from shown), and toggling back is non-destructive.

Hide tags controls in the tag hierarchy — Hide tags from Synthesis Outputs

Merge tags into one

To combine multiple tags into one — whether they’re duplicates or concepts you want to consolidate — select the **Merge** icon and use the two dropdowns on the right to choose your tags. The **Source** tag is merged into the **Target** tag, so the Target’s position is kept. (For example, data collected for “Mean Age” might be merged into a general “Age” tag.)

Merge tags controls with Source and Target selections — Merge source tags into a target tag

Delete tags

Select the **Bin** icon, choose the tags to delete (they turn red), then check **Confirm** and **Delete**. The **recursive** option deletes a tag together with all its children. If you’re bulk deleting tags by selecting parent tags, **make sure you select recursive!** Otherwise, orphaned child tags makes for a tedious select and delete process. {% hint style="danger" %} **Warning:** Deleting a tag removes it from the hierarchy **and from every study it’s been applied to**, and it cannot be undone — so don’t delete a tag unless you’re certain you won’t need it. Take particular care deleting a parent tag on its own: its child tags revert to Root Tags and move to the bottom of the hierarchy. {% endhint %}

Zoom

Use the right-hand view panels to zoom in and out on a section or view the full hierarchy. Select a parent tag and zoom in to view only that branch. These controls only change the view. **Zoom**

Hierarchy views

The tag hierarchy is shown in horizontal orientation by default. You can also switch to vertical orientation. The **Tagging QA table** also lists each tag, its question type, and its completion count. Counts show studies marked Relevant, Not Relevant, or Unanswered. Each count opens **Study Inspector**, filtered to matching studies. The table is available only in **Form-Based Tagging**. Use it to identify unanswered tags during a review. Before exporting, confirm each required tag has no unanswered studies. **Vertical orientation**

Vertical view of the tag hierarchy — Vertical orientation

**Tagging QA table**

{% hint style="warning" %} The Tagging QA table is only populated once tags are applied. {% endhint %}

### FAQs

How do I decide on a tag hierarchy structure so my exports come out the way I want?

Structure with your export in mind from the start — root and child tags become the columns and groupings in your exported dataset, so the same decisions that make tagging clear also decide what your spreadsheet looks like afterward. See [Match Tags to Your Export Spreadsheet](/review-workflow/tagging/match-tags-to-your-exports.md) for mapping an existing sheet's columns to tags, and the "Choosing a tag's question type" tree, above, for the specific question type and content type each individual tag should use.

Can I just create one large tag table rather than several smaller ones?

You can, but several smaller, purpose-built tables are usually the better choice. Splitting your data by domain makes filtering easier — a smaller table's rows are all one clear kind of data, rather than something you have to pull out of a larger mix — and each gets its own Smart Insights generation, where one large combined table only produces a single, less focused summary. If you want a single combined view for a report, combine the separate tables at export instead of building one large table from the start.

How do I tell whether a tag was created manually or by a Smart Tags tool?

Check for the sparkle icon — see [edit existing tags](#edit-existing-tags) above.

Can I change a tag's question type after studies are already tagged?

Yes. Captured data remains available. Review tagged studies after the change to confirm each answer still fits. See [Apply Tags](/review-workflow/tagging/apply-tags.md).

Should I use a Select tag or Single Apply text options?

Default to **Single Apply** with **Text Options** whenever a study gets one answer from a fixed list — it's faster to tag and exports more cleanly. Use **Select** when a study can have more than one answer, when each answer needs its own evidence, or when an answer needs follow-up questions beneath it. See Choose A Question Type above.

Can Text Options capture more than one answer per study?

No. A Text Options tag records exactly one choice. If more than one answer can apply, use a **Select** tag and turn on **Multiple Responses**.

Can I combine manual tags with generated or imported tags?

Yes. You can combine manual tags, **Core Smart Tags**, **Smart Config: Tagging**, and imported templates in one hierarchy. You can also merge tags. See [above.](#merge-tags-into-one)

Does reordering or reparenting affect applied data?

No. Reparenting changes a tag's location and position in the Tagging form. It doesn't change data applied to studies.

### Next Steps


Match Tags to Your Export Spreadsheet	Plan an export that matches your tag structure.	Match Tags to your Exports
Import Tag Hierarchy Templates	Start with a hierarchy template and adapt it for your review.	Import Tag Hierarchy Templates
AI Tools in Tagging	Use AI-assisted tools to build and apply tags.	AI Tools in Tagging
Apply Tags	Capture data from your included studies.	Apply Tags
Import Applied Tags	Import tag data you captured outside Nested Knowledge.	Import Applied Tags

--- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.nested-knowledge.com/review-workflow/tagging/configure-tagging.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.