I got off the phone yesterday with my dad (Jason Wilkey) and he was explaining a conversation he had with a Merch by Amazon seller yesterday. He said:

“The person could not believe that the shirts they created were not selling. They said that their family members loved them and they figured everyone else would as well.”

As we were talking, I agreed that it is not an uncommon practice for someone to only create designs THEY would like.

When I first started, I did the same thing. I started by creating a bunch of shirts that I personally thought were funny or awesome and never considered other buyers.

For being such an intelligent guy, I can be extremely stupid at times.

Think about it. I was creating t-shirt designs for an audience of one.

Let’s face it. If you are trying to build a t-shirt empire, you are going to need designs that appeal to more than just one person.

Chris. This sounds great, but how in the world do I know what other people like?

The answer is simple: Research.

If you have been uploading Merch by Amazon designs over the past year (or how ever lone you have had your account), hopefully you have been doing some initial research to help you determine your niche.

If you have not, don’t fear because I’ll give you the steps you need to take.

Let me warn you right now. I just finished up a rough draft of a research proposal for my doctoral class and this might be more technical than usual. 

When it comes to research, I want you to think about one of my favorite shows of all time (note: this is not the first time it has been referenced on this blog).

Mythbusters.

The show lays out a nice formula for research and testing that I think most people can understand on some level (way better than the jargon and methods in my research design class).

So I’m going to layout out a very similar, easy to follow formula designed to help you research print on demand design ideas.

1 – Focus on data, not your gut.

The first major hurdle we have to face is the dreaded gut feeling.

My gut could tell me that I am a badass NCIS agent, but data can prove to me that I couldn’t run and catch a suspect if I tried.

And when it comes to research:

Data > Gut Feelings

2 – Start with a literature review

The first part of any research study is a review of related studies. You want to find out what has previously been tested, what has worked, and what has failed miserably.

When it comes to Merch By Amazon, there are a few different ways this can be done:

A – A search through Amazon

What better way to determine what sells well on Amazon than to look at Amazon. This isn’t an exact science, but it will at least get you started in the right direction.

Simply go onto Amazon and search for your keyword or niche.

Once the results populate, drill down on the left hand side to get to Clothing > Novelty & More > Clothing > Novelty

At this point, you should have drilled down to a level where Merch by Amazon shirts can be found. Since we are wanting to learn about their performance, we want to eliminate the other data points from our analysis.

(Side note: If one of the shirt bullet points reads: “Lightweight, Classic fit, Double-needle sleeve and bottom hem” that’s a dead giveaway it is a Merch by Amazon shirt).

Then you simply can choose a shirt and scroll down to the sales rank on the page. It should look something like this:

If you don’t see a sales rank, that means the shirt has not yet sold on Amazon.

The lower the sales rank number, the more often this shirt has sold, just like a normal Amazon product.

B – A search through Google Keyword Planner

If you are a little more skilled in digital marketing, you can actually use an awesome free tool to help you gather information about different keywords and niches outside of Amazon (yes – I know our audience is on Amazon, but multiple sources of data can help support our research as a large number of people use Google to search and find things on Amazon).

Meet the Google Keyword Planner. If you want a print on demand specific keyword planner, check out WordCandy or MerchInformer (these are paid services, but they are specifically designed for Merch sellers).

I won’t spend a bunch of time explaining this right now, but it’s pretty straightforward:

Type in your keyword(s) and hit search.

Once you hit search, you will get results back.

All we care about are the first two columns (for beginners that is). This is going to show you the average number of search for that keyword per month and then it is going to show you the competition level for ads.

It’s up to you to determine what levels are acceptable to you, but the data is there for the mining.

C – Searching Google

If we want to know what people are looking at related to a specific topic or niche, we should go to the largest search engine in the world: Google.

As I am searching for the keyword or niche, I try to drill down on things like phrases, saying, quotes, funny images, etc to find out what people in this niche enjoy.

Once I have something that is a possible idea, I throw it into one of the first two methods and see if there is an audience for the idea. This is an awesome way to quickly validate ideas before moving to the next step in the process.

D – Searching Pinterest

My favorite place to start my research has been Pinterest. Personally, I do not use it on a regular basis, but my wife is an avid user. She had to teach me how to use it (it wasn’t hard, but I asked for a little help anyway).

Since most people buy after seeing a design, Pinterest is a great platform to test and gather data from a visual perspective.

I like to judge the design ideas by the number of pins or re-pins on a design as someone took the time to say they liked a design (please remember that showing interest and making a purchase are two completely different things). Once I find a design idea or niche, I take a look at it in Google or Amazon to see the performance there as well.

3 – Test a hypothesis (aka a design or niche) on a small scale first

If you find a niche that you think has potential, you should spend the month making thousands of designs…

Seriously? You should know better than that.

Start off small with 1 or 2 different design ideas and give them a small scale test run. See if they make any sales. If they do, great – move on to the next step! If they don’t, you should go back and look at your research method and see where you went wrong (a great way to do this is to compare the data between the shirts that do sell and the shirts that do not sell).

4 – Take it to a large scale test

If the small scale test provides positive results, then it is time to go large scale.

Continue making new designs and leverage the knowledge you have gained from the research and experience. Repeat the process and find your next niche (over and over again).

And more importantly.

Make it rain.

CW