Conversational Analytics

Conversational Analytics lets you ask questions about your recruiting data in plain English and get instant answers with charts, tables, and explanations. Instead of building reports or navigating complex dashboards, you simply type a question — like "How many candidates applied last month?" — and get an immediate, visual response.

Conversational Analytics is powered by AI and works with your existing Recruit reporting data. No special training or Looker expertise is required.

Who can access Conversational Analytics

Conversational Analytics is available to users with a Tier 3 Reporting License. Your Administrator must also enable the feature at the site level.

If you don't see Conversational Analytics in your left navigation, contact your Administrator about your reporting tier assignment.

Navigate to Conversational Analytics

In Rival Recruit, click Reporting & Analytics in the left navigation menu, then select Conversational Analytics.

When you land on the page, you can:

• Start a new conversation — Begin a fresh session with any available Explore (data set).
• Access recent conversations — Pick up where you left off with a previously saved conversation.
• Choose your Explore — Select which data set you'd like to query.

Choose an Explore

Before asking questions, select which Explore (data set) you'd like to query. Click the Explores tab to see what's available.

Explore	Description
Jobs and Resumes	Your core recruiting data — jobs, candidates, applications, hiring stages, dispositions, sources, time-to-fill, time-to-hire, offers, and more. This is where most questions are answered.
Requisitions	Requisition approval workflows — who requested, who approved, approval dates, and status.
Fee Agencies	External recruiting agency data — agency names, submissions, placements, and related job information.
Job Templates	Reusable job template configurations — template names, associated settings, and usage.
Data Transfer Logs	Technical data transfer activity between systems — primarily useful for Administrators monitoring data pipeline health.
Users	User account information — names, roles, departments, locations, and reporting tier assignments.

Tip: For most users, Jobs and Resumes is the best starting point. It contains the broadest set of recruiting data.

Ask a question

After selecting an Explore, you'll see the conversation interface. Type your question in the "Ask a question" bar at the bottom of the screen. No special syntax or formulas are needed — just write naturally.

You can also click one of the Suggested Questions the AI offers as starter prompts.

Question modes

Use the mode dropdown next to the "Ask a question" bar to choose between two modes:

• Thinking mode (default) — Best for complex, multi-step analysis. The AI plans its approach and shows its reasoning. Use this for trend comparisons, "why" questions, and multi-dimension analysis.
• Fast mode — Best for quick, factual lookups. Use this when you need a specific number, a single metric, or a simple list.

The mode stays the same throughout your conversation unless you change it.

Multi-turn conversations

The AI remembers what you've already asked and builds on it. Each follow-up question refines the previous answer — just like a real conversation. Start broad, then drill down.

Example:

1. "Show me total applications by source this quarter"
2. "Now filter that to just the Engineering department"
3. "Which of those sources had the highest conversion to hire?"

The AI may also ask clarifying questions before answering (for example, "Did you mean total candidates, or candidates currently active in a pipeline?"). Answer the question and the conversation continues.

Visualizations

The AI automatically creates the most appropriate visualization for your question — bar charts, line charts, pie charts, scatter plots, or data tables. You can ask it to change the type (for example, "Show that as a pie chart instead").

Example questions

Here are example questions you can ask, organized by topic. Results depend on the data your organization has configured in Recruit.

Hiring funnel & speed

• "What is the average days to hire across different departments over the last year?"
• "Which hiring stage currently has the longest average duration?"
• "How has our time-to-fill trended month-over-month for the past 6 months?"

Source & quality effectiveness

• "Which recruitment source has the highest conversion rate from applicant to hire?"
• "What percentage of our total hires came from employee referrals this year?"
• "What is the total resume count by source for the last quarter?"

Team performance

• "Which recruiters have managed the highest number of successful hires in the last 90 days?"
• "What is the job count and filled job count for each business unit?"
• "Compare the average days to hire between different hiring managers."

Diversity & compliance (EEO)

• "What is the breakdown of candidate ethnicity for all resumes received in the last 6 months?"
• "Show me the gender distribution of candidates who reached the final offer stage."
• "What is the count of hired candidates categorized by veteran status?"

Salary & job trends

• "What is the average budgeted salary by job level and department?"
• "List the most common job titles currently being recruited for."
• "Show me all active jobs that have been open for more than 60 days."

Manage conversations

• Auto-save: Every conversation is automatically saved. Just close the page or navigate away when you're done.
• Resume a conversation: Go to the Conversational Analytics page and click any conversation in the Recent section. Conversations are listed by title (automatically named based on your first question).
• Start fresh: Click + New Conversation and select your Explore to begin a new conversation.
• Delete a conversation: Deleted conversations go to the Trash folder and stay for 30 days before being permanently removed. You can restore them from Trash.

Tips for better results

1. Be specific with time ranges — Instead of "recently," say "in the last 30 days" or "this quarter."
2. Name what you want to see — "Show me a bar chart of applications by source" gives better results than just "show me applications."
3. Use follow-ups to refine — Start broad, then narrow down with follow-up questions.
4. Try both modes — If Fast gives you a number but you want to understand the trend, switch to Thinking and ask a follow-up.
5. Don't worry about exact wording — The AI understands natural language. "How long does it take to fill a job?" works just as well as "What is the average time-to-fill?"

Known limitations

Limitation	Details
Maximum rows per query	5,000. Add filters to narrow large data sets.
Visualization types	Line, area, bar, scatter, and pie charts. Maps and heatmaps are not available.
Conversation context	Context is per-conversation only. You cannot reference questions from a previous conversation.
Data availability	Results depend on the data your organization has configured. EEO and salary queries require those fields to be populated.
Privacy	Your conversations are private. Other users — including managers and administrators — cannot see your queries or results.

Conversational Analytics only accesses data you already have permission to see. All existing security and access rules remain in place.