Hi! This is our template at Centro to building content experiments. Hope this helps :).

PS: If you’re a person building or working at a Shopify or Amazon brand, shoot me an email ([email protected])! We build software to help you scale!

Goal:

  1. Build a content engine that can repeatedly generate views and reach in a topic either adjacent to our audience or directly related to our audience.

1. Platform Selection:

  1. Rank the platforms where your ideal audience lives, based on your best knowledge. This worked for us, since we knew our audience by demographics and heuristics.
    1. TikTok
    2. YouTube
    3. LinkedIn
    4. Facebook
    5. Twitter
    6. TV
    7. Reddit
  2. Choose the highest ranking channel where we do experiment exploration. You could explore multiple channels but it was easier for my small team to focus on 1.

2. Content Trend Selection:

  1. Make a list of 5 content trends that’s ideally <6 months old on your channel of choice. List each platform the content trend has lived on and 1-2 great examples of that trend on each platform. This means you HAVE to find 1-2 examples of each platform. This was good for Inspo! Why do I want to find trends? If it’s a content style that’s worked across multiple channels, multiple generations, there’s something inherently entertaining about it that I can leverage with my own spin. That also means, it is probably something I can make “an engine” out of for repeat content.
    1. Eg; Like this:
      1. DiY Recipes (Home Channels → Youtube Videos (Bon Appetit) → Facebook (Tasty) → TikTok (DoobyDoBap))
      2. Street Interviews (Late Night Shows (Jimmy Kimmel) → Youtube Interviews (Konbini Confessions) → TikTok (@rizwan )
      3. 1v1 Sports Clips (Roman Coliseums → Radio → Cable TV → Youtube →
      4. Inspirational Speaking (Churches and Religions → etc etc
      5. Dramatic Interviews with Cut Scenes (Reality TV → True Crime → etc etc)
  2. Factors we evaluate the content idea on:
    1. How old is this content on my channel of choice?
      1. Anything <1 year is probably good, the fresher the better. Or if it’s veryyy old (>3 years). Nothing in between, won’t hit.
    2. Can be retrofitted to my ideal audience
      1. I AM my ideal viewer: this kind of content is something I already watch!
      2. If I know my ideal viewer: this trend is something they HAVE watched or DO watch.
    3. What’s one NEW thing we can do on this that makes it either TIMELY or CAPTIVATING
      1. Write any ideas asap!
      2. Search for it in your platform of choice etc. If it exists, it’s a GOOD sign.
    4. Within my team’s strengths to do well
      1. For us, we prioritized things we could do QUICK.
      2. We have the product uniquely outfitted for this.
      3. We have access to people who would be great IN the content we want to produce. (eg; the interviewer, direct star, etc).
  3. Pick 1 trend that satisfies all the factors in 2. If none do, ideate another trend.

3. Tactical Experiment Parameters

  1. Hypothesis
    1. Content: This content will generate 1 viral video in my first experiment. Define your own version of viral. We said 100,000 views is viral given cold start.
    2. Business: We wanted to test if this had any direct business impact.
      1. Direct site traffic increase?
      2. Direct sales impact?
      3. Direct lead or pipeline impact?
  2. Sample
    1. We chose 20 videos as sample size of N as sufficient. This worked because we chose something simple (street interviews)
  3. If Hypothesis Confirmed, So What?
    1. We said we’d produce another 40 videos if it went well.
      1. Our mindset is we will keep experimenting at increasing scales and budget. If scale is 2x, budget is 1.5x.
    2. Goal would be to maintain the same hypothesis parameters. Basically, can we recreate success.
    3. Also, if you figure out a good content trend that aligns and gets you content/biz value, can probably repurpose for ads or blogs.
  4. If Hypothesis Denied, So What?
    1. Try a different content trend (step 2) OR a different channel (step 1).

4. Project Steps

Prep Phase

  1. Learn about the platform of Choice’s nuances
    1. When to post
    2. How the algorithm works
    3. If I have to “Warm up” the algorithm by engaging in existing content
    4. What is the local platform’s SEO like
      1. Tiktok, I should use their autogenerated subtitles, etc.
    5. How long should I observe results
      1. Most social platforms is within 1 week
    6. https://blog.hootsuite.com/how-the-linkedin-algorithm-works-hacks/#What_is_the_LinkedIn_algorithm for linkedin, etc.