Early Access
This article has been published prior to Ask’s GA date. Content may be subject to updates before its official release.
Introduction
Workflow Steps
Ask allows you to explore your documents using natural language questions. While it’s easy to use, following this recommended workflow will help you get the most accurate and actionable results:
Accessing Ask
To start using Ask, click the Ask AI button on the right side of the Search box.
This opens the Ask popup window. You can close it at any time by clicking the “X” in the upper-right corner. Closing the window does not end your Ask session — click the “Ask AI” button again to reopen the popup and resume exactly where you left off. The window can also be freely moved around the screen for your convenience.
I. Set Scope
Ask allows you to define the range of documents it will search:
Search Entire Dataset – Ask reviews every document in your dataset.
Search a Subset of Documents – Ask limits results to a specific slice of the dataset, such as filtered search results, selected custodians, or communications within a time period.
For best results, we recommend defining a focused search range whenever your question references metadata (e.g., “emails from John Doe during September 2025”).
Search Entire Dataset
Selecting All Docs at the bottom of the Ask window instructs Ask to analyze your full dataset. This is the broadest option and ensures you don’t miss any potentially relevant materials.
Search a Subset of Documents (Current Results)
To take a more targeted approach select Current Results at the bottom of the Ask window. This focuses Ask only on that filtered set of documents.
Think of this as your deep dive — giving you sharper answers and more relevant supporting documents.
Examples of Narrowing Your Dataset
Here are three common ways to focus your search before using Ask:
By Metadata and Timeframe
Example: Limit queries to email communications between specific people or during certain dates.
How: Use metadata fields like
REV FROM,REV TO,REV FROM DOMAIN, andREV TO DOMAINor filter by file extension and date ranges in the Dashboard.
By Keyword List
Example: Search only documents containing terms from a curated list.
How: Use Term List Search to add up to 1,000 terms (words, Boolean expressions, or field-specific queries). Preview document counts, adjust as needed, and run the list to generate a focused dataset.
By Custodian
Example: Focus only on documents belonging to a particular custodian.
How: Apply a custodian filter via the Custodian chart or directly through search.
II. Ask Your Question
Once you’ve defined your search range, it’s time to write your question and submit it to Ask. You can also provide additional instructions to shape the response.
Writing an Effective Question
Ask can process almost any natural-language question. However, the way you phrase your query strongly influences the quality of the response.
Best practices:
Ask one question at a time – Break down complex or multi-part queries into separate questions.
Be explicit and specific – Avoid vague terms or pronouns; the clearer your wording, the clearer the response.
Match your query to your goal – Use broader queries for high-level overviews, and narrow ones for detail-oriented answers.
Less effective approaches:
Relying on metadata queries – Ask only analyzes document content, not metadata. Use LGK search to filter metadata first, then submit to Ask.
Asking for “all” information – Ask is designed for precision, not exhaustive recall.
Expecting intuition – Ask cannot infer beyond the evidence in your dataset.
Requests spanning massive document sets – Ask builds narratives from a limited number of highly relevant documents, not full-scale aggregation (e.g., across hundreds of spreadsheets).
Adding Additional Instructions
You can guide Ask’s output further with custom formatting or content instructions. Use this feature to adapt responses to your workflow.
Examples:
“Create a summary?”
“Translate the response into German.”
“Create a list.”
To add instructions, click More Details in the Ask window, then type or paste your preferences.
Submitting Your Question
When your question and (optional) instructions are ready:
Press Enter, or
Click the up arrow to the right of the query box.
III. Refine
Ask is designed to be iterative. You ask a question, review the answer, then decide whether to keep it, tweak it, or reframe it. This refine–evaluate cycle helps you get to the most useful results quickly.
When No Adjustment Is Needed
Sometimes your first question delivers exactly what you need. In that case, move forward to validating with citations.
Minor Adjustments
If the answer is close but not perfect, a small adjustment is often enough.
Rephrase your query with slightly different wording.
Clarify terms or avoid ambiguous phrasing.
Even subtle changes can shift Ask’s interpretation and improve results.
Major Adjustments
If the answer seems incomplete, off-target, or incorrect, consider broader adjustments. Common reasons include:
The information isn’t in your dataset (or in the subset you’re searching).
The dataset is too broad or too narrow.
The query is poorly aligned with the kind of answer you want.
There are ways to improve it flow:
Adjust scope – Try searching a more focused subset (e.g., emails in a date range) or expand to the full dataset.
Include duplicates – Ask normally indexes unique/pivot docs only. Adding duplicates may surface more relevant material.
Reframe the question – Phrase it in a different way to match how information might appear in documents.
IV. Validate
Just as with any research method, it’s essential to validate the answers Ask provides. AI-generated summaries are helpful starting points, but they must be checked against the actual source material to ensure accuracy and completeness.
Ask organizes its findings using Sources (the documents) and References (specific passages within those documents that closely match your question). Multiple References can come from a single Source, allowing Ask to provide concise summaries while pointing you to the most relevant text.
Even with this structured approach, AI-generated answers can sometimes be incomplete or misleading. For example, if a document discusses an employee’s performance over multiple years, Ask may highlight only part of the story, potentially missing key context.
Validating with Sources and References
To help you validate results, Ask provides access to the top 100 Supporting Docs and References for your query.
Supporting – A document that contains relevant information.
Reference – A short excerpt from a Source that closely matches the question you asked. Each Source includes at least one Reference.
Ask displays the Supporting docs in descending order of the sematic similarity score of the first Reference from each Supporting docs. Ask shows an abbreviated version of each Reference; you can click on More to see the full Reference.
Some Supporting Docs contain more than one Reference. When that happens, Ask displays the References for the Supporting Docs in descending order of sematic similarity.
You can validate Ask’s results in several ways:
Review References – Start with the short text snippets (References) provided. Expand with More to see the complete passage when needed.
Read Sources – Access the full documents (Sources) to check context and ensure Ask’s summary matches the original through the button
next to the Reference score.Use Other Search Tools – Leverage Logikcull’s keywords, filters, or concept searches to confirm results or uncover additional relevant information.
Check External Information – When necessary, cross-check facts against outside resources such as interview notes, meeting records, or documents not included in your dataset.
Note
Think of validation as a critical checkpoint. Ask helps you find relevant information quickly, but reviewing the underlying documents ensures that your conclusions are accurate, complete, and actionable.
V. Review
Once you’ve validated Ask’s results, the next step is to decide what to do with them. Ask is not just about getting answers — it’s about moving your matter forward.
After reviewing the responses, you can:
Tag documents – Apply tags to relevant documents for organization, privilege review, or production prep.
Save into a search set – Group the Sources you validated into a saved search so you can easily return to them later.
Export or share – If appropriate, export results or share them with your legal team for further analysis.
Using Ask History
Ask History helps you keep track of your past questions, so you can build on previous searches without starting from scratch. It’s a quick way to revisit insights, continue exploration, and stay organized.
View Past Questions: Access the Ask History sidebar to see all previous queries.
Reopen: Click any question to reopen it, or add follow-up questions.
Retain Context: Each question in history keeps the context of your prior search, ensuring continuity.
Note
Use Ask History to experiment with different queries and compare results without losing track of what you’ve already explored.
Best Practices
When using Ask, it’s important to approach the results thoughtfully. The feature is designed to help you surface insights quickly, but human review and judgment remain essential. Therefore:
Always verify Ask’s results; AI-generated responses should be confirmed against the source.
Start broad, then narrow your questions as insights emerge.
Use Ask for investigations, early case assessments, or fast fact-finding in large datasets.