How to Test Personalized Videos With Sample Recipients

Test a few sample recipients before you scale

Before you send a personalized-video campaign, generate a sample with realistic recipient values and inspect the result. Check that each variable sounds natural, appears at the right time, and works with your subtitles, backgrounds, and video framing. Fix problems in the template or settings before you generate videos for a full list.

Choose representative sample values

Pick a small set of test cases that reflects the kinds of values in your recipient list. Include names with different lengths, company names, and any row-specific links or media your video uses. This is practical testing advice, not a required number of samples.

Use only the part of a phrase that should change. For example, if the template already says “Hi” before a first-name variable, enter “Erada,” not “Hi Erada.” Including fixed words again can make the video repeat them.

• A short and a longer first name
• A short and a longer company name
• A realistic value for each link, image, or background variable

Generate a sample in BHuman

In BHuman, sample generation happens in AI Studio, where you can check a template before launching a campaign. The template needs a valid source video and at least one variable. Add realistic values for the template’s variables, then generate a sample.

You can optionally choose a background type for each variable and enable subtitles. BHuman submits one preview request and returns you to the template editor while the output finishes. Sample generation may be blocked by free-plan limits.

• Open the template in AI Studio and make sure its source video is ready.
• Enter a test value for every listed transcript or manual-timing variable.
• Optionally choose backgrounds and enable subtitles.
• Generate the sample and wait for its status to update.

Review the finished video

When a sample succeeds, open it from Sample and Other Generated Videos and watch it. Listen for pronunciation and voice problems, and check that each replacement fits the timing and subtitles. Look for issues with the source video, presenter framing, and any backgrounds you selected.

BHuman sample results can show as preparing, queued, processing, succeeded, failed, or needs review. If a sample fails or needs review, use the status and any available details to guide your next check.

• Does each name and company value sound right?
• Does the personalized speech fit the available timing?
• Do subtitles reflect the inserted values?
• Do the background and presentation mode look right?
• Does the source video load and frame the presenter as expected?

Fix issues before campaign generation

If the sample looks wrong, correct the relevant transcript variables, voice settings, manual timing, source video, subtitles, backgrounds, or presentation mode, then generate another sample. A successful sample checks more than the text: it can help reveal problems that might otherwise affect a batch.

For campaign generation, BHuman lets you use CSV or Excel data, as well as supported connected or API-based inputs. Map each template variable to its matching field and make sure every row has a value for mapped variables. Campaign setup validates these requirements before sending the render request.

Choose Mini or Max based on the framing

BHuman offers Mini and Max sample modes. Mini places the presenter over a background and is the safer default for a first sample, especially when the presenter is close to the camera. Max uses full-screen HD, but it can be unavailable because of plan limits or face-distance checks. Choose the mode that fits the footage and your account’s available options, then check the framing in the sample.

If you use Mini in a campaign, personalized moments need background coverage. Check that coverage before generating the batch.