Service
Personalized video marketing
We build systems that generate video for each viewer, changing the name, place or product automatically at the scale of your audience. Before any of that, we work out what data you can lawfully use and whether the personal detail will actually make the message more useful.
What's included.
- use case and audience definition
- data and consent mapping
- script and template design
- personalization fields and fallback rules
- automated rendering setup
- AI voice or avatar, where agreed
- AI disclosure placement, where required
- email, SMS, landing page or CRM delivery
- batch testing and sample review
- tracking and reporting setup
- documentation and handover
What personalized video is.
Personalized video is a video that changes based on information about the person watching it. A customer might see their own name on screen, the store closest to them, the product they looked at or a summary of their account.
There are two main ways to make it. The older method uses a template, where text, images and pre-recorded clips get swapped in by a video rendering system based on your data. AI video personalization adds generated elements, such as a synthetic voice that says each name or an avatar whose speech is generated from a script. Templates are predictable and easier to review. Generated elements offer more variety, and they bring more risk of errors and more questions about consent and disclosure.
It's also worth separating two levels of personalization. Segment-level video changes by city, language, product category or customer type, and it can often run with little or no personal information. Name-level video speaks to one identified person, so it needs lawful personal data behind every version.
When personalized video makes sense.
It tends to suit moments where the viewer already expects to hear from you and the data makes the message more useful. Onboarding, renewals, account summaries, appointment follow-ups and post-purchase messages are typical examples.
It's unlikely to help if your customer data is patchy or out of date, because errors become very visible on video. It's also a poor fit when the only personal touch is a first name on a message nobody wanted. A well-made standard video would serve you better.
Consent and personal data.
Name-level personalization needs lawful data. That covers where the data came from, what people were told when they gave it and whether this use fits within that permission. Sending the video has its own rules too. Email and text messages are subject to consent requirements in many markets, including Canada's anti-spam law, and privacy laws such as Quebec's Law 25 and Europe's GDPR can apply to how the data gets processed.
We map the data before we build. We decide which fields the video actually needs, which outside services will receive them, how long rendered videos and data are kept and how someone's request to opt out or have their data removed gets handled. If a detail would feel intrusive, we leave it out even where the law would allow it.
A personal video also needs to be delivered safely. Each one can reveal information about the viewer, so we plan links that are hard to guess, along with an expiry date where that makes sense. Legal questions stay with your qualified reviewer.
AI disclosure and likeness.
Where a video uses a generated voice, avatar or face, some markets and platforms require you to say so. Your qualified reviewer decides what applies to your campaign, and we build the disclosure into the video or the surrounding message as agreed.
Using a real person's voice or face, such as a founder or a spokesperson, needs that person's documented consent for this specific use. We don't generate anyone's likeness without it.
Quality at scale.
No one can watch thousands of versions, so the checking has to be designed in. We test unusual names, missing fields, long place names, accents and special characters, and each field has a fallback for when the data is missing or doesn't look right. A video that drops the name is fine. One that greets someone by the wrong name or by a raw database field is a real problem.
We also review a sample from each batch before sending. A person signs off on the template, the script and the sample, and automation only produces what they approved.
How it works.
- 01
Pick the moment
We choose the message where personal detail will help the viewer.
- 02
Check the data
We review what data you hold, how complete it is and what your permissions cover. Your qualified reviewer looks at the consent and disclosure questions at this stage.
- 03
Design the template
We write the script and design the video with the personal elements in mind, then define a fallback for each one. Your team approves it before any rendering starts.
- 04
Build and test
We set up rendering and delivery, then run difficult test records through the system. A person reviews the output and checks that the right video reaches the right person.
- 05
Send and learn
We start with a smaller batch, check the results and watch for errors, then widen it. We report on what people watched and did.
How we measure personalized video.
We test it against the alternative. Where the audience is large enough, one group receives the personalized video and a comparable group receives the standard version, so you can see what the personalization added. We then compare opens, plays, watch time, clicks and the action the video was asking for.
We also track the problems, including render failures, data errors and opt-outs.
How we price it.
We scope the build in hours and confirm the applicable engineering rate before work starts. Strategy and advisory are CAD $175/h. Hosting, AI usage, software subscriptions, creative production and ad budgets are identified separately, because some of them can continue after launch.
We agree on the deliverables and review points in advance, and your team receives the agreed code, accounts and documentation. If new work appears, we discuss the hours before expanding the scope.
Questions before we start.
What is personalized video marketing?
It's the use of videos that change for each viewer based on data, such as their name, location or the product they bought, usually sent by email, text or a landing page.
How much does personalized video cost?
It depends on the template, the number of personal elements and how the videos get delivered. Rendering and any AI usage add an ongoing cost that grows with your audience, and we identify that separately in the scope.
Is it legal to use someone's name in a video?
That depends on how you obtained the data, what people agreed to and where they live. Your qualified reviewer needs to confirm it, and we build within what they approve.
Do we have to tell people the video uses AI?
Some markets and platforms require it where generated voices or faces are used. Your qualified reviewer decides what applies, and we build it into the video or the message.
Can we use personalized video in paid ads?
Often at segment level, such as by city or product. Name-level detail is harder, because ad platforms have their own rules on what an ad can imply about a person. We check this for your plan.
Related services.
Start a conversation.
Tell us what you are trying to move. We will tell you which disciplines are worth engaging, and which are not.