I would like to get your opinion about a way that I recently used for UX proposal. I still do not know if that was the right way but the budget was so tight.

What did i do?

The business was about pregnancy context. I searched both good and bad apps in different stores. I had a general check about their functionality. I copied all the comments and searched for some patterns.

I analyzed these positive and negative comments both from app stores. It was strange that I feel like I did a user interview after spending one day. Especially 4-3 stared comments.

Client liked the output in the end. It gave some insights but I wonder if there is anyone who used a similar way to validate some ideas? If yes, can you please share your experience?


Do you think that the results of this analysis is valid? If yes, how can we defend that? What kind of constraints do you think that this method has?

What i learned:

  • If you feel there is a lie about comment, just take it out
  • Increase the number of comments and take different examples for comparing
  • I disregard any man input since the idea is about pregnancy...


  • I like this topic, but can you reword it a bit so it's not just a 'share your experiences' type of question? Alas, those types of questions don't tend to stay open on StackExchange.
    – DA01
    Commented Jan 10, 2014 at 18:22
  • Hi, I tried to edit this question according to your feedback. I am aware of the fact that, share your experience is some how abroad but it is also a way that we can collect more information ;).
    – Abektes
    Commented Jan 10, 2014 at 22:34

2 Answers 2


What you did was a competitive analysis based on user reviews. You're asking if anyone has used a similar technique to validate ideas, and I have.

Online reviews are fantastic sources of qualitative data, whether you are using Yelp reviews, Google+ endorsements, or product reviews on the app store. But they need to be taken with a grain of salt, as you found when you excluded any reviews that struck you as dishonest. If there are hundreds of reviews, you can evaluate and plot the sentiment of each review and see where they fall on the continuum from very negative to very positive. If most are somewhat positive, and there are a couple of terrible reviews and a handful of glowingly positive reviews, you can throw out the outliers and focus on the reviews that are consistent with the majority sentiment. Since you only want to analyze reviews from appropriate users, you excluded male reviews from your pregnancy app review.

I would add that you can generally throw out any reviews that are written in ALL CAPS with lots of punctuation as well ;-)

From there, you can create a matrix of features from the reviews, noting which features are hits and which are misses.

This sort of competitive analysis can be very insightful, and it is no wonder that the client was pleased. It just needs to be packaged as exactly what it is -- an analysis of user reviews of Products X and Y from Sources A and B. To go further with it, when you have more time, download the apps and use them to accomplish specific tasks over time. You will learn even more. As a next step, you can do a SWOT Analysis, armed with what you know about the problem domain, gleaned from using these apps.

  • Furthermore you can analyse how mature the business or app is. Early Adopters will ask for or miss features. Early Majority will just moan about some bugs and say that this wasn't expected from a senior company.
    – FrankL
    Commented Jan 11, 2014 at 12:58

Reviews from other competitors are great for insight, but don't always take them at face value. One thing I do find very useful from reviews on competing products is finding the pain points that the users will constantly complain about and filling in those gaps which can put your product at a competitive advantage. Aside from a competitive analysis, there are other approaches you can take at this point.

Geared with the information you've already come across during your research I would quickly build out an MVP with a prototype. If it's a mobile app, Flinto (resource link below) works wonders and you can easily get something going to validate very early on. If it's a desktop app, check out Invisionapp.



You can create personas and scenarios for the client prior to this, however you'll find that you'll get alot more insight and validation when you have a working MVP in a users hands early and often. Create test scenarios with each iteration to validate your design decisions and iterate constantly based on that. The feedback you get from the testing is a great deliverable to bring to the client. I've been using usertesting.com for my tests lately and it works wonders. If you're under budget constraints you can always "get out of the building" and find people to run the prototypes by. I would stress though, that if there is a certain target market you're going after (women I assume?), aim towards finding them to test.

Again, having a working viable product early on, even if it's just a prototype or sketch, and putting into users hands early, will allow you to gain the most insight and help direct you and your client on where you need to take it.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.