Skip to content
improve usability and outcome introducing new recommendations ( thumbnail recommendations, tags, and recommend title). encouraging users to set a search term for the video to give them better results. allowing collaboration and sharing. In most video agencies that use the product there is usually 2 people optimizing the video.
Product owner: Matthias responsible for the whole project roadmap and the delivery. He also gathered feedback from customers why they did not use the tool. Developers: There was 1 frontend and 1 backend developer, also an AI engineer who developed the recommendation algorithm. UX/UI designer: I was responsible for ideation, wireframes, validating ideas with the PO and the AI engineer, as well as prototyping and user testingMarketing: They did a good job promoting the new feature, and managed the alpha release to selected customers while monitoring their usage closely.
The product owner was a YouTube consultant himself for 10 years, he helped alot in the ideation and he was getting me in contact with his agency customers. He is also expert with analytics (mixpanel)Me, as a product designer. I did customer interviews, competitor analysis, research, prototyping and monitored the results after release on Mixpanel AI engineer and devssenior youtube consultant: helped in improving the quality of the recommendations
Share
Explore
Tubics Video Optimization Tool 2021
The video optimization page is the most used feature in Tubics, 79% of the time spent on Tubics users are trying to optimize the meta data of their old videos.
Company
Tubics Gmbh
Duration
4 Months
Role
UX/UI Designer & Researcher
Watch the feature announcement video 👀
About Tubics:
Tubics is a video optimization and ideation tool powered by data from YouTube API, what makes it very powerful is the data driven recommendations for titles, descriptions and tags. As well as suggesting you new video ideas based on your industry, the app can even generate full script using AI based on your YouTube channel meta data
About the project:
The project started on Oct 2020, the first release of the video optimization page was on Feb 2021. The aim of the project was
This is the old video optimization page
Project Team and responsibilities
Research
The Problem
“Our users wants quick wins, they want to optimize their videos in 5 minutes with little effort, but our AI can not generate good recommendations without knowing the video topic/ the search term which the user want to optimizw the video for. And most users skip selecting a search term for the video”
The project team and the roles
Getting customer feedback:
We wanted to know which softwares our users normally use to optimize for SEO in general. and what are their other problems with out current software. For this we interviewed 4 expert users who worked in YouTube agencies for years and use Tubics.
I am missing Rich recommendations, give me examples and why do I need to do this. not just tell me change the video title
this is an example of Tubics recommendations at that time
I am using grammarly pro, and usetopic content optimization tool, I want a mix of the ease of use of grammarly, and the richness of use topic
Example from grammerly doc optimization page
Example from use topic SEO rich recommendation
I used TubeBuddy video optimization page, it is very slow, their recs are better than Tubics but still not so much helpful
Thought Process:
we thought of how might we solve these problems to achieve the best outcome for our customers.
Problem1:
Without a search term the AI can not generate quality recommendations for the users. Selecting a search term should be the first step. and we need to justify for the user why do they need to select a search term upfront by watching a video and knowing what is it about. Otherwise the YouTube API gives a list of suggested search terms which YT think the video is about it.
Just as easy as an API call. 😀
user first has to select a search term
Or select a suggested one
Problem 2:
We wanted to give our customers quality recommendations, but also tell them why they need to do this and show them examples.
Problem 3:
We want to give the users guidance, give them a step by step instruction on how to fix their video content. Grammarly does a good job in this, by showing your current article on the right side, and the grammarly recommendations on the left side.
1st release
For the first release we wanted to focus on helping the customers getting more views and subscribers through optimizing their video. Thus the quality of the meta data was the most important thing.
To achieve this we first we made this step by step video optimization process, that starts first with choosing a high traffic search term so that the AI can give our customers the best recommendations.
1st step: Select a search term
select a search term to optimize your video metadata for. user can either select one from the YT suggested search term or manually search for a search term. After this user will be taken to the video optimization page
2nd step: Optimize the video SEO
once you click on the video title you will see the corresponding recommendation on the left side. in this case the user is trying to edit the description and is recommended to add the video chapters to the description
Feedback
We did a figma protoype and tested it with 5 customers in an unstructured user interview. We just showed them the prototype and asked them to speak their thought loud. We put them in a scenario and asked them to optimize a video speaking about “Winter holidays in Austria” and the feedback was as follows.
“now it makes more sense to me, and the video title examples are nice”
“It is nice that you show me a list of suggested search terms, didnt know that youtube offer this feaure. now I can make a data driven decision based on numbers”
“before I publish this video, i want to share it with a colleague of mine first”
“how can I revert my changes?”- somehow it was hiddent from the users how to undo their changes by clicking on the dropdown beside the save draft button to select delete draft.
“Sometimes I want to search for keywords while editing my video metadata, in this case I am opening Google Adwords in another tab, would be nice if tubics game this feature”
“i dont want to rewrite the video description everytime, all my videos descriptions have the same template. would be nice if their a way to create templates on tubics and access them if needed”
last release
In this release we included 3 other features based on the feedback we got from the first release.
Feature 1: Sharing a draft with a colleague before publishing. User can click on this link icon to copy the draft link to the clipboard. The icon itself might not be very obvious so we made an onboarding tour with intercom to highlight it once the user lands on this page
Feature 2: Search for keywords while working on your video.
Feature3: add and import templates in your description.
What we learned and next steps:
With now the video optimization page getting bigger and having more features, it is important to offer a different experience for novice users compared to expert users. we learned that only the expert users use the templates, and the keywords search feature, also the draft feature. But own normal users didnt really use this feature, and based on a survey we launched to 200 users through our app. Those users didnt use these features because they didnt need it actually. So the next step we did, is that in the onboarding we asked the user 4 questions to classify them to expert or novice users and based on this we show them 2 different views of the video optimization page.
Want to print your doc?
This is not the way.
This is not the way.
Try clicking the ⋯ next to your doc name or using a keyboard shortcut (
CtrlP
) instead.