Skip to main content
How to A/B test my scanners?
Vadym O avatar
Written by Vadym O
Updated over a week ago

Working with scanners requires constant improvements across various aspects such as filters, keywords, templates, and profiles. To identify areas for improvement and determine effective strategies, conducting testing, specifically A/B testing, is essential. Testing various hypotheses is the key to successful lead generation.

To perform A/B testing, the following steps are recommended:

  1. View the performance of your scanners on the Dashboard, specifically the Proposal View Rate (PVR) and Lead Reply Rate (LRR) results as follows:

    1. From your GigRadar account, navigate to the Dashboard.

    2. In the Scanner Breakdown section, identify the scanners with poor PVR and LRR results.


  2. Perform A/B testing using any of the following options. Expand the option for instructions:

Option 1: A/B test various cover letter templates for your scanners


This option will allow you to evaluate which template performs better and which one should be kept in use.

  1. Select the scanner identified in step 1 above and duplicate it. For more information about duplicating a scanner, see How to duplicate a scanner?.

  2. In the original scanner:

    1. Click the Auto Bidder tab.

    2. Under Auto Bidder Schedule, make sure that the Auto Bidding Enabled button is ON.

    3. Select the Custom bidding setting and specify the recurring weekly Auto Bidder schedule by selecting the days you wish to run it, such as Monday, Wednesday, Friday, and Sunday.


    4. Keep the Cover Letter Template unchanged for this scanner.


    5. Click Save.

  3. In the duplicate scanner:

    1. Click the Auto Bidder tab.

    2. Under Auto Bidder Schedule, make sure that the Auto Bidding Enabled button is ON.

    3. Select the Custom bidding setting and specify the recurring weekly Auto Bidder schedule by selecting the days you wish to run it, such as Tuesday, Thursday, and Saturday. The schedule must be different from the original scanner schedule.


    4. Update the Cover Letter Template for this scanner to be different from the original scanner.

    5. Click Save.

      Watch a video


Option 2: A/B test various Freelancer Profiles


This option will allow you to evaluate which profile works better and receives more responses.

  1. Select another scanner identified in step 1 above with poor PVR and LRR results and duplicate it. For more information about duplicating a scanner, see How to duplicate a scanner?.

  2. In the original scanner:

    1. Click the Auto Bidder tab.

    2. Under Auto Bidder Schedule, make sure that the Auto Bidding Enabled button is ON.

    3. Select the Custom bidding setting and specify the recurring weekly Auto Bidder schedule by selecting the days you wish to run it, such as Monday, Wednesday, Friday, and Sunday.

    4. Keep the same Freelancer Profile that you have selected under Bidding Options for this scanner. Do not make any change to this setting.


    5. Click Save.

  3. In the duplicate scanner:

    1. Click the Auto Bidder tab.

    2. Select the Custom bidding setting and specify the recurring weekly Auto Bidder schedule by selecting the days you wish to run it, such as Tuesday, Thursday, and Saturday. The schedule must be different from the original scanner schedule.


    3. Under Bidding Options, select another freelancer name in the Select freelancer drop-down and select a different profile in the Select profile drop-down. Alternatively, you can use the same freelancer as the original scanner, but select a different profile for the duplicate scanner.

    4. Click Save.

      Watch a video

👉 Next up:

Once you've finished creating duplicate scanners and adjusting their settings, you can assess their performance in the Dashboard > Scanner Breakdown section.

The scanners with the best results should be kept and the scanners that are showing poor results could be deleted or further fine-tuned, if you want to perform more testing.

GigRadar recommends assessing the results of both scanners over an equal timeframe. This will help you analyze the results better.


Did this answer your question?