How To: Market Research for Your App Idea
Doing market research is essential to build your own indie app. Based on the market research you do, you can make better decisions about design, development, marketing and your app’s business.
Traditionally, market research has often been done by giving surveys to focus groups. Invite a few people, let them fill out some forms or discuss a promo video, and you’re done. Based on those surveys, marketers then made decisions about the products they were trying to sell.
Surveys are fraught with selection bias, so you never know what you’re really testing. Do they just like the product or will they buy it? As it turns out, most people in survey liked a product – but wouldn’t ever spend their own money on it, outside the survey. Oops!
In this tutorial, we’re going to take a different approach to market research. Instead of creating new data with surveys, we’re going to find the data that already exists. You’re going to find out if there’s a demand for your app, without ever needing to leave your (home) office.
Ready? Let’s go.
- Kill Your Darlings?
- Comparing Ideas with Google Trends
- Market Research Insights with Keyword Planner
- Finding Stories, Problems and Questions
- Further Reading
This is an article from the series about my app Crest. In the series, I show you how I’m creating this app from scratch. I’ll guide you through the tools I used, the decisions I took, the mistakes I made, and the insights I found. See all articles
Kill Your Darlings?
When I studied user interaction design, I received a bit of advice.
Kill your darlings ASAP
When you’re building your app, you’re also choosing from dozens of other app projects you could potentially build. Many app ideas haven’t been discovered yet; many markets go untapped. How do you pick the right project to work on?
Great apps solve a problem, so unless you’re out of problems, not all apps have been built yet. You always build your app for a specific audience, and that means you can solve the same problem for different audiences.
When you start with many app ideas, you go through a process of elimination. This is where Kill Your Darlings comes in. Instead of sticking with 1 supposedly great app idea, you start your market research with 10 decent app ideas and find out which one is the best. You kill the other ideas.
When I started working on the Crest app, I had a few app ideas:
- A journaling app, like Bear
- A to-do list app, like Wunderlist
- A to-do list app specifically for entrepreneurs
- A day planner app, like a calendar or paper planner
- Something entirely different: an app with inspiring motivational quotes
I have a thing for to-do list apps – my baby darling to-do list app idea – so I was emotionally attached to the idea of building a gorgeous to-do list app myself. But I also knew that a to-do list app without a clear differentiator wouldn’t do well in the App Store.
What comes next is important: Instead of looking for proof that one of these ideas was the best, I started looking for proof that every idea was the worst. I wanted to get rid of these ideas as quickly as possible!
Instead of secretly hoping that my nicest, sweetest app idea would win, I scrutinized each of the app ideas. Are they viable? Technologically possible? Would my people, my audience, benefit from this app?
If not, kill the idea and move on. Let’s discuss how you can find out the difference between go/no-go for yourself.
Comparing Ideas with Google Trends
When I’m researching an idea, my first stop is always Google Trends. With it, you can see the relative popularity of searches in the Google Search engine. It’s a great market research tool!
You can find trends like:
- Spotting disease epidemics – when are people Googling for flu symptoms?
- Finding news-related trending topics, for instance about popularity of iPhone models
- Comparing the rise(s) and decline(s) of technologies, like SMS vs. VR
Quick Note: You won’t find how many people Googled a search term in Trends. Instead, it shows what percentage of people searched for a term relative to when this search term was the most popular (i.e., from 0% to 100%).
How do I test my app ideas? I type in my search queries into Google Trends and measure their relative popularity (see image above).
There’s two things you should know here:
- I chose to format the search term as “best X app”, because people Googling for such a phrase are clearly looking to download an app. When you’re unsure whether your search term reflects what you’re looking for, simply Google the term, and check if you expected the results that you got.
- I chose a wide time range for the graph (2008 to now), so I can clearly see the popularity over time. The popularity of apps in the App Store is fickle, so I wanted to see the full range since the beginning of the App Store in 2008.
Then, looking at the data, this is what you see:
- Planner apps are the most popular, and show a August-January seasonal pattern
- Journaling apps are the least popular, but their popularity grows steadily
- To-do apps were the most popular in 2011-2012, but their popularity is now declining
- Quote apps are the late risers here, but after 2014 they appear to flatten out, and even decline, too
It’s easy to make a mistake interpreting this graph! When working with trends, it’s important to consider the trend directions instead of shallow data points. Should I choose the planner app idea because it’s more popular than the journaling app idea?
A few examples:
- I wouldn’t pick the to-do list app idea, because it’s declining after it rose to popularity. It’s a sinking ship! (I’m glad I didn’t build this without checking.)
- I also won’t pick the quotes app, among other reasons, because it’s appearing to flatten out. I can’t say for sure, it might be temporary, but I don’t want to bet on it.
- I might choose the planner app idea, because it has a pattern: in January and in August, planners are more popular. Why? I’d have to research a hypothesis here, but an uneducated guess is: people buy planners in January and August! I could tap into that trend by giving them an alternative purchase: my app.
It’s also a good idea to check what you see in the graph with what you know about that particular period. Don’t just use numbers in your market research!
For instance, Wunderlist was created in 2011, in the period to-do apps were popular. The first version of Bear was published in 2016, which coincides with the rise of journaling apps. Keep in mind that this doesn’t prove that the graph is right, it’s just data to support assertions you make about the trends.
What about the journaling app idea?
As you’ll discover in future articles about the work I did for the Crest app, I chose to combine a planner app with a journaling app, for a specific niche market. The assertion I made based on the Google Trends graph, is that “journaling” is a category that grows. Combine that growth with the purchase trends seen in planner apps, and you have a (theoretical) winner app idea.
And the other ideas? I killed them… Let’s move on to the next step!
Should you look for ideas that no one has invented yet? It depends. In most cases, I’d rather build an app with fierce competition than with no competitors at all. Your competitors help you create a market, and there’s often enough to share. If there aren’t any competitors, you might have to create the market on your own…
Market Research Insights with Keyword Planner
Yes, the next market research step involves another Google tool: Google Keyword Planner.
With this tool, you can find fine-grained volume data for search phrases. How many people Googled a specific search query? It’s typically used for search engine advertising and Search Engine Optimization (SEO), but as a research tool it’s just as valuable.
I use Google Keyword Planner for two things:
- Find out what people search for
- Find out how many of them search for that
A great thing about Keyword Planner is that you don’t have to come up with ideas for search phrases yourself. It can make suggestions for search phrases based on some input you give it.
Before you start your research with Keyword Planner, you have to ask yourself what it is you want to find out. Before I built Crest, this is what I wanted to know:
- The people who search for information related to journaling and to-do list apps, what specific search phrases do they use?
- When they search for those things, what do they find?
- How many of them search for these phrases?
When you’ve done your research with Keyword Planner, you end up with a list of search phrases and their average search volume. You can use this for App Store Optimization, but also as a list of topics for website articles you can write.
Remember how we wanted to look for research data that already exists? The search phrases you find with Keyword Planner are the words your customers use to (potentially) search for and find your app.
Let’s break that down. Say you’re a potential customer for the Crest app. You have a problem: you’re having trouble to achieve your goal, starting a business, because you don’t have a plan. Your days aren’t structured, so you struggle to get where you want to go.
This is a problem. It’s the problem the Crest app solves. What would people with that problem search for on Google?
- “How do I achieve my goals?”
- “Best day planner app”
- “To-do list app for goals”
I put those search phrases into Google Keyword Planner and this is what came out:
Here’s what you can learn from my results:
- The “seed” search phrases I put in are OK. One of them has zero hits, so I’ll have to refine that. Always keep in mind what the intention is of someone when they type a search into Google. Are they looking to make a purchase? Discovering options? Looking for alternatives? Or just browsing?
- The keywords Google came up with are pure gold. These are the words you’re looking for. Phrases like “goal setting”, “the power of habit” and “weekly planner” are all words I couldn’t come up with on my own, but now I know people search for them!
- Unfortunately, Google only gives you ranges to work with. This isn’t absolute search volume. Instead, Google is just giving you ballpark figures for you to work with. Fortunately, there’s a trick where you can get to actual search volume.
Next, I want to get more accurate search volume data. These ranges are OK, but I want to know exactly how many people search for these terms. This also helps you to compare search phrase volume for different app ideas.
- First, in the keyword ideas overview as shown in the image above, click the blue arrow for a few keywords. The arrow is in the last column that says Add to Plan. This will add these keywords to your keyword plan.
- Then, click Review Plan. This will take you to a new page. It’ll show a graph and spreadsheet. Make sure to go to the Keyword tab. Also, make sure to choose a period of a month in the top-right Monthly forecasts based on ….
- Then, choose Match Type -> All -> Exact Match from the menu.
- Finally, at the top of the page, set $1 or €1 for Enter a bid and press Enter.
Quick Note: You can also set the maximum bid here, i.e. choose a bid based on the line graph. This means you’re only seeing impression data for the first search results page.
The spreadsheet should now update, showing you more accurate clicks and impressions data compared to the ranges you got before.
Your version of Google Keyword Planner could look a bit different than in this tutorial, and that’s OK. The principles are the same!
Make no mistake – the dollar values you see are advertisement costs based on those search phrases. The impression data is real though! This is the number of times Google thinks your ad (or search result) will be seen in the given period (hence “Impressions”).
What do you do with this data? You can use it to rank app ideas based on popularity. A popular search phrase doesn’t necessarily mean it’s a good app idea, but it gives you a rough estimate of the demand for your app based on the number of people that search for a given phrase. Awesome!
Let’s move on to the most important research step!
Finding Stories, Problems and Questions
Don’t scope out the market in numbers from your Ivory Tower. It’s time to now move away from the cold research tool, and get into stories, problems and questions.
Empathy is the most valuable market research tool.
Thanks to Google Trends and Keyword Planner you now know where to look. That doesn’t mean you have a winner app idea yet!
You’ve eliminated some ideas, but the real litmus test comes when you present your app to someone that has the problem that your app solves. You can then validate your app idea, and see if there’s a real demand for it.
The next thing I did when researching the Crest app is this. I took all my researched keywords and phrases, and started using them:
- I Googled the phrases, which gave me a ton of research on competitors, blog posts, alternative products, etc.
- I used these phrases on social networks and discussion boards like Quora, Twitter, Facebook, Reddit, Hacker News, LinkedIn, Product Hunt, etc.
- I used these phrases in the App Store and on Google Play, as well as searching for alternative physical products online
What that got me was this:
- The stories people tell about themselves in relation to the problem they have and the solution they’re looking for
- The problems they have, like not achieving a goal and feeling disorganized
- The questions they ask in relation to the problem, like “How do I set goals best?”
By trying to put myself in the shoes of the people I seek to serve with my app idea, I learned a lot – more than I ever could from staring at numbers.
That’s all there is to it! Here’s what we discussed:
- Why it’s important to kill your darlings early
- How you can compare the potential for different app ideas
- How to find stuff people are Googling for, and how that affects app ideas
- What you can do with that data: finding stories, problems and questions
In a next post in this series, I’ll show you how I did competitor research and how I got my app idea in front of potential customers. It’s going to be interesting!
Want to learn more? Check out these resources:
Code Swift right in your browser!
Go to the Swift Sandbox