Personalized Video Outreach Campaign Launch Checklist
Plan, build, test, and launch a personalized video outreach campaign with a practical checklist and a clearly hypothetical worked example.
On this page6 sections
The launch checklist
Use this checklist to prepare your message, recipient data, video template, and campaign settings before generating personalized outreach videos. Test a small set of rows first, then review the results before preparing the full batch.
- Define the audience and the one next step you want recipients to take.
- Write a short base script. Decide which words or media should change for each recipient and keep the fixed wording in the template.
- Record a source video and prepare an AI Studio template with at least one variable.
- Create a clean recipient file. Include a column for every variable, with one recipient’s values per row.
- Choose a campaign data source, such as CSV or Excel, Zapier, Leadr, Pabbly, or an API, based on how you’ll provide the rows.
- Map each template variable to its matching column or field. Check that every required variable has a mapping and every mapped row has a value.
- If using personalized moments in Mini mode, make sure each moment has background coverage. If using timed background segments, provide valid URLs or uploads for each segment.
- Generate a few test rows. Check names, pronunciation, timing, subtitles, and any background assets.
- Review the generated results and resolve failures or items needing review before using the videos in outreach.
- Confirm generation is available and the account has enough quota, then submit the campaign and monitor its results.
Build the template and recipient data
In BHuman, a campaign connects an AI Studio template to a data source. The template’s variables define what can change for each recipient; campaign rows supply those values. A campaign sends the rows to an asynchronous render pipeline, where you can monitor statuses such as queued, processing, succeeded, failed, or needs review.
Use stable, readable variable names, preferably lowercase snake_case, such as first_name, company, or booking_url. Keep each variable’s meaning consistent between the template and the data source. Map each one to its corresponding column or field.
- Put only the replacement words in a spoken-variable cell. If the template already says “Hi,” use “Erada” as the first_name value—not “Hi Erada.”
- Use separate variables for different kinds of values. For example, use booking_url for a scheduling link rather than a generic url field.
- Keep variable names stable after a campaign or automation is live; changing them can break mappings.
Choose a data source and viewing mode
For a manual batch or test, use CSV or Excel; their headers become mapping options. For connected workflows, documentation also lists Zapier, Leadr, Pabbly, and API inputs. Pick the source that fits how your team supplies campaign rows.
Choose Mini or Max for the viewing mode. Mini places the presenter over backgrounds and may add a default generic background; personalized moments need background coverage. Max creates full-screen HD output, but it may be unavailable because of plan limits or face-distance checks.
Hypothetical worked example
For example, imagine a sales team inviting operations leaders to book a product conversation. Its template says, “Hi {first_name}, I made this for {company}. You can choose a time here: {booking_url}.” The team’s source file has first_name, company, and booking_url columns, with each row containing values for one recipient.
Before generating the full campaign, the team maps those three variables, generates a few rows, and listens for correct pronunciation and natural timing. It also checks that each booking link points to the intended destination and that any personalized background moments have suitable assets. This is an illustration of the workflow, not a claim about campaign results.
Final checks before generation
Campaign generation can stop before rendering if a required variable is unmapped or its source column is missing. Check the source video, attached data source, variable mappings, and row values before submitting.
Also check mode-specific requirements: Mini needs background coverage for personalized moments, and timed background segments need valid URLs or uploads. If voice cloning is disabled, manual audio must be recorded when the campaign uses manual audio. Generation also requires availability and enough account quota.
After you submit the campaign
Generation is asynchronous. Use the generated videos area to monitor progress, review failures or needs-review results, and retry failures when appropriate. The campaign workflow can provide generated status and, when available, share and download links.
Treat review as part of the launch, not an optional polish step: verify the rendered variable replacements and destination links before sending videos to recipients.