I'm extremely impressed with how much pre-work you do for some of these posts Neville. The data focused angle on this was great. I think adding some more examples to the post would help developers. Perhaps even template-tizing (is that a word?) some descriptions for them. Keep up the great work.

I'd be really interested to hear how your opinion stacked up against the data and/or your opinion on the data.

Also, in place of the overall top 100 in paid and free, which probably include things like Facebook, it would be cool to look at newly added apps that are doing well.

Thank you Ronit! I actually woke up at 5am to finish off the findings, so I appreciate that :)

I think the template is a great idea. I might actually go do that. A NevBox-worthy suggestion!

Hey Lucas, one of the things that surprised me was the shortness of descriptions. I thought app developers would just use all the space they could to add positive reviews and different uses, but many of them kept it short (especially in the free section where the most common length is between 100 and 200 words)!

I originally tried looking at some of the "newly added" apps, but unfortunately there's a lot of crappy apps too. I decided to just pull lessons from the big dawgs!

It looks like the app developers did a lot of homework or hired a marketer to determine the "sweet spot" for the amount of words in the description, bullet points (if any), amount of pictures, words used for SEO, etc.... I don't think these findings are just random, it looks like a pattern to me and obviously it is working for them. This idea of scrapping information can really be useful for many different things being sold on the internet. Thank you for doing the research and letting us know about it.

Hi Nev,

If possible, I'd wanna see examples of app descriptions you've written for companies and if you have any sort of data on them as well. Yes, app descriptions do help when a user is just shopping for these apps randomly (say category wise) on the Appstore. But in most cases, people usually directly go to your Appstore page and download your app after reading/ getting to know about it from somewhere or someone.

I believe the most influential factor in helping you get app downloads is word-of-mouth and your app actually being good and helpful for the user (of course).

But always good to have these data handy for when you're about to sit down to write app descriptions. Thanks yet again for such an insightful post.

Thanks Chris! I think analyzing the actual data is important, rather than just me spouting information about how I THINK a description should be written (although I might add some of that in just as a guide to developers who don't wanna deal with it).

Hey Abeniel, I totally agree.....most people will never read much of the description. I mainly made these observations just to share with other developers who I know will like the data more than an opinion of how to write a description!

Great post, Neville. Thanks for posting! Definitely got me thinking about a handful of things. First, I was reminded of Ogilvy's quote: “On the average, five times as many people read the headline as read the body copy. When you have written your headline, you have spent eighty cents out of your dollar.” It seems that this hold with apps. The title and initial "headline" or top line words in the short and long description probably matter a lot. I would love to see some type of analysis of the top line in the descriptions. What approaches and styles do they use? Is it a headline? Description? Statistic? Testimonial?

That leads me to my next observation and question: how frequently and in what form do top 100 apps use testimonials or other influence factors, like authority positioning?

Your suggestion to focus a bit more on images is consistent with my experience as a user. I always judge an app by its cover, but still dive in a little bit to explore. I usually look at the name, peak at the short description, then flip through the images. If I like what I see, I then may read the longer description and/or the reviews. A few years ago I was quicker to download an app. Now, with so many apps, it takes more convincing to get me to spend my time checking out an app.

Finally, your analysis reveals a lot about the business models this apps are using. While this is not as related to copy, it is very interesting as an entrepreneur. I work with a lot of people trying to build apps. Most of them have no idea how they want to make it a business. They think that an app is a business by definition. This is obviously not the case. From your analysis, you get a sense of how some of these companies make money (upfront purchase, in app, and--while you didn't look at it--ads) on their apps. Of course, others don't use an app as a revenue or profit generator, and may be positioned as a marketing tool, supplement to their core offering, or lead-in to other products.

Thanks again for the insights and analysis.

This was really interesting to read Nev! Kinda daunting to realize that the key to download success hinges much more on how professional and crisp you can make the download page look versus how it actually plays. But I suppose word of mouth and a good product will win out in the end.

Have you considered making your own app to test out different copy and strategies?

Data really does trump opinion. I appreciate that you went the data route here, it's refreshing.

One issue I do see with these data is that they only show one moment in time -- May 1, 2016. Rankings can fluctuate quite a bit and so your results could be significantly different on another day. It might be interesting to do this same data scrape on multiple days over time. It might also be interesting to categorize each app type and follow them over time to see which are consistently in the top 100 and which just got lucky one day. So, for example, you would follow the apps that use long copy and follow the apps that use short copy and see which is consistently in the top 100.

If you really wanted a deeper understanding of how the copy affects sales you'd want to control for confounding variables. Big brand name vs small time developer, ratings and number of reviews, color schemes and picture quality..... categorizing apps and following them could sort this out. I'm guessing you have better things to do though. Maybe if you get bored.

Another idea would be to have a focus group rate how compelled they are to buy an app based only on the copy in front of them. Take away the pictures, reviews, app name, etc, and see if the copy converts.

Further research aside, I think with any copy the length depends on your audience and how much you'll need to convince them to buy. Saying the average word count of all 100 apps is x is not very useful. If I am selling an app that stores your businesses data and keeps track of inventory, my copy is going to be longer than if I am selling an app that makes fart noises. Both apps could be in the top 100 on any given day, it's not useful to average the word count on very different apps like that and say "therefore, 250 words is best."

Neville,

Thanks for sharing this… I read one of the comments about a template (nice idea)…. I put together an Excel template with a word count for the different Kopy Blocks (check your NevMed email). Please feel free to share it with the Kopy Kommunity. Hope it helps.

Thanks again.

So if you really want to nerd out... Could there be a case of correlation vs causality? You do kind of address it, but with Google or EA having most of the top 100, they could write 10 words or 1000 words and skew the data. Maybe you need to take a random sample of the top 1000 apps and see if they correlate to ranking. Of course, because that wouldn't take a whole of time...

You must have poured in a lot of work on the data collection, Nev!

Another dimension that can be useful: Is there a correlation between the number of reviews an app has and its ranking in the app store?

If so, app developers can think about how to actively encourage reviews from its users.

Good article - one I will be referring to for ideas.

I like the raw data about using bullets - interesting fact!

Neville, good stuff! I don't know if I've seen this type of analysis before, so way to go!

I haven't developed an app but I'm guessing developing is the easy part, the hard part is getting people to download so this has to be really important for new apps. Also, I'm sure the same type of information is relevant for Amazon and eBay, what are the top sellers doing in their titles and descriptions.

Speaking of titles, that might be something worth looking at. Many of the free app had a few word description as part of the title.

Few things I noticed glancing though the data is many of the top 100 are the well known apps (twitter, uber, google, EA etc) so I tend to question the value of those app descriptions. Not to discount them but it seems like a no name developer would be more interested in what the successful "little guys" are doing vs the big names that people may be downloading regardless of the description.

So now that you've done all this work :), I wonder if data apps ranked 50-150 or 100-200 would be more relevant for the newer unknown apps? Maybe I'll do that.

It appears there were few apps with less than 3 stars with the exception of well known companies, like Starbucks. I'd assume reviews play a big part as well.

How much of the content includes a "changelog"? I suspect quite a few description will have at least some part of the changelog. I'm also assuming that the 0-100 word count apps are the typically the big apps which already have a significant amount of traction and whose audience sustains them in the top-100 list

Do you also have stats on how many of them had a promo video?

> But in most cases, people usually directly go to your Appstore page and download your app after reading/ getting to know about it from somewhere or someone.

Definitely agree in this.

Having crappy descriptions/bad store page will cause a bit of friction - a "WTF". To re-convert the user, that user needs to discover MORE social proof (one that should have been done with a good store page). A tiny moment of hesitation can mean the difference between getting the download or not.

So why leave it to chance?

A couple things I think might be interesting to delve in to.

1. I mainly looked at the paid apps spreadsheet because most of the free ones I already heard of and would have downloaded them no matter what the description said (I would download Instagram if the description said that this is the worst photo app in the world, but I know it from outside of the app store world so I will download it anyways)

Maybe an analysis of free apps with no associated online services would be more beneficial.

2. Most of the descriptions have testimonials. I wonder if there is a sweet spot for the number of testimonials or if they even matter.

Incredible content!!! Your data-driven approach is very impressive and clearly superior to just stating your opinion.

Great job Nev!

Hi Nev,

Thanks for doing all of this work and analysis! I have a couple questions:

1. Do you think it is more effective to use your first 50 words for third-party endorsements and reviews? Or for your core product description?

2. What role should image captions play?

3. Do you believe the description as any impact on ASO (despite Apple's insistence that it doesn't)?

4. Is there any way to scrape for keywords?

Look forward to reading your newsletters!

Best,

Alex

Giz, this will be very helpful for too many. *Sharing Now.

- ALSO..

Here's my comment about the data, and about other experiences I've been involved with launching apps.

- REVIEWS ARE TOO IMPORTANT FOR APPLE, I MEAN, REAL REVIEWS

- If you are Electronic Arts, chances your other games will be downloaded if you have (as they have) TOO MANY GAMES (some are cool)

- The description you put in your app, does work with the advertising you use (keywords and developer site), as in ALLINSEO strategy. Because, facebook and google now have targeting for apps instals exclusively.

Advice for games:

- give youtubers free apps (and of course, get them to use the apps.. that's the challenge)

- hire some chic to do the commercials :P (

)

That's all (for nahow)

.. with the EA games, they get promoted because of the same developer and appear below the description like recommendations for the user

Totally true Rocky! I probably didn't convey that so well. Will probably put a template for people to follow in the post for some additional guidance.

What I usually suggest for app store descriptions is BRAGGING by showing how many ratings you have, or great testimonials from users.

Thanks Greg!

Well I think an AMAZING APP will gain momentum if enough people like it. However, having some good marketing surrounding it definitely helps. You're right, good word of mouth and an amazing product will win in the long run.

I haven't considered making my own app just yet....as that may be a several-months-long experiment, and I've got a lot of other stuff to write about first!