Auto-benchmarking uses AI to automatically match your employees to Comptryx market data, instead of manually looking up one job at a time. This guide walks through the full workflow.
Step 1: Navigate to auto-benchmarking
In the left nav, click Benchmark employees under the Benchmarking section. This opens the "Map your team to market benchmarks" page. Then click "Start mapping".
Step 2: Configure your percentile settings
Click the Market percentile button in the top right. A modal opens with three settings:
Target market percentile (Default 50th percentile)
Choose the percentile you want to anchor your pay strategy to (e.g. 50th = targeting median market pay)
Comptryx supports a range of 10-90
Benchmark thresholds (Defaults 25th and 75th percentiles)
Consider pay low when under [percentile] - flags employees whose pay falls below this threshold
Consider pay high when over [percentile] - flags employees whose pay exceeds this threshold
These are used to highlight pay outliers in the insights dashboard
Comptryx percent adjustment (optional)
A multiplier applied to all displayed percentile values
Example: setting 0.91 reduces all benchmark values by 9%
Use case: accounting for regional or market-specific adjustments
Step 3: Auto-map your employees
Click the Auto-map employees button (with the AI icon). Comprehensive will analyze your workforce and automatically suggest a job family, level, and geography match for each employee. Once complete, AI-suggested values appear in the table with a blue AI icon. You can review and adjust any of these before approving. You can also skip this step and map employees manually using the drop-downs in the table.
Step 4: Review the employee table
The table shows all your employees with these key columns:
Employee Name and Job Title - for reference
Status - Mapped, Unmapped, or Unapproved
Job Family - dropdown to select or change the matched Comptryx job family
Level - dropdown to select or change the matched level
Benchmark Geo - dropdown to select geography (e.g. Tier 1, Tier 2, or a specific metro)
Benchmark Type - choose Base Salary or Total Cash depending on what you want to benchmark against
Additional columns you can toggle on:
Market Percentile - where the employee's current pay falls relative to the benchmark
Percentile columns (10th %, 25th, 50th, 75th, 90th %)
Sample Size - number of employees in the benchmark dataset for that job + level
Compensation, OTE, Employee Geo, Department, and more
Step 5: Make adjustments
Single employee: Click any dropdown (Job Family, Level, Geo) to change the selection. Click View all in the dropdown for a full list of options.
Bulk edit: Select multiple employees using the checkboxes, then use the bulk action buttons:
Change family / Change level / Change geo / Change pay type - update the selection for all selected employees at once
Reset - revert to the original AI recommendation
Re-run for selected - re-run the AI matching for specific employees
Step 6: Approve mappings
Each employee's mapping must be approved before it appears in insights. You can:
Click Approve on individual rows
Select multiple employees and click Approve in bulk
A mapping must have all three fields filled (Job Family, Level, and Geo) before it can be approved.
Step 7: Save and view insights
Click Save & view insights in the top right. If any mapped employees haven't been approved yet, you'll see a confirmation modal with two options:
Approve all and view insights - approves all fully mapped employees in one click (recommended)
View only approved insights - shows insights for only the employees you've already approved
Step 8: Review the insights dashboard
After saving, you're taken to the Benchmark Insights page, which shows:
Cost to Market - the total compensation adjustment needed to bring your team to your target percentile
Employees Below Low Percentile - count of employees flagged as underpaid based on your threshold
Employees Above High Percentile - count of employees flagged as overpaid
Average Percentile - your workforce's average position relative to the market
Detailed employee table - each employee's benchmark with their market percentile position
Tips for accurate auto-benchmarking
Review AI suggestions before approving. Auto-map does a good job, but always verify that the suggested job family and level match the actual scope of the role.
Use Benchmark Geo intentionally. Choose the geography that reflects where you compete for talent. If you hire nationally, a tier-based approach works well; if you hire locally, pick the specific metro areas, where available.
Watch sample sizes. Small samples mean percentile values are directional, not precise. Check the Sample Size column if something looks off.
Adjust thresholds to your philosophy. Conservative orgs might flag pay below the 40th and above the 70th; aggressive orgs might use 25th and 90th. Set these to match how your team thinks about comp.