Just Another iPhone Blog posted an article about the application store reviews, and I was a little surprised that some of the commentors didn’t see the problem. How reviews are handled is very important to both developers and consumers, and this isn’t just an Apple issue. With new application stores popping up everywhere these are issues that can impact every store out there.

The Review Problem
It was pointed out that reviews in the iTunes App Store are sometimes inaccurate, occasionally inappropriate, and completely incontestable. This means that someone could post a review like:

This application doesn’t make coffee!

…when in fact the application does make coffee and has made coffee since the application was first released. Or even worse – it’s software – of course it doesn’t make coffee – what sort of review is that? This sort of review is useless, misleading, and undermines the validity of the entire review system.

People Don’t Take Reviews Seriously Anyhow
This is one of the arguments I heard, but people take these reviews very seriously (at the moment anyhow). For a customer who’s never used the application before, they have no reason not to believe a review. Yes, if the customer buys the application they’ll find out this review was wrong, but they might never get to that point.

It’s Someone’s Opinion So It’s Okay
This is another argument that was voiced. Opinions are one thing. Falsehoods and misinformation are another. Using the coffee example, let’s look at a real review:

It makes coffee but I had a really hard time finding the functionality.

Excellent review! It doesn’t lie but makes clear a deficiency in the product. Most likely, the “doesn’t make coffee” review I used as an example was actually THIS review but written by someone who didn’t get help (see my note later about letting developers respond).

Developers Are Afraid of Bad Reviews
This is something else a commentor mentioned. When developers complain about reviews, they aren’t complaining about people who don’t like their application. Even reviews like:

This app sucks! 1 Star!!

…seldom get complaints from developers. It’s the false or wildly inaccurate reviews that developers care about.

This System Works Everywhere Else
Although this is true, that doesn’t mean it works for software. For music, books, and other types of media a loose system is great. For more complex products like cars, computers, or software, a simplistic system just doesn’t cut it. The products are just too complex with too many features. You need a more robust review system that better handles the complexity of these products.

Why Should You Care?
Bad review systems do the following:

1. Mislead consumers about how popular or unpopular a product is.
2. Give consumers false information about a product.
3. Make consumers eventually reject the entire review process as pointless.

When this happens consumers skip products that might be excellent solutions for their needs, they make decisions based on false information, and eventually they lose a valuable tool (a review system) for choosing which application is right for them.

The Solution
Unless they want something akin to the comments section on YouTube, here is what the retailers have to do it.

1. Review the Reviews
Block poorly written or inaccurate reviews. Is this censorship? In an anarchistic society possibly, but if reviews are to fulfill an actual purpose you’ve got to have a system for vetting them.

2. Let Developers Respond
Let us help these people! When retailers do this, we’ve had many cases where a person who initially posted a bad review pulled it and posted a positive review after we helped them!

3. Let Reviewers Change Their Reviews
Assuming you let developers respond, give the reviewer a chance to alter their review. That way they can come back and say “You know, I was wrong.”

4. Solicit Both Positive AND Negative Reviews
If the retailer only asks for reviews when the user uninstalls, you get a “Rate It If You Hate It” system. It only encourages everyone who hates the app to review it. This skews the reviews making it hard to tell which apps really are good.

Wrapping It Up…
This stuff affects consumers as much as it does developers. If implemented badly, a review system is an exercise in futility. If implemented well, a review system can be one of the most important buying tools available to a consumer. So email your favorite retailer today, and tell them to clean up their review system!

There was a lot of talk recently regarding Tweetie for iPhone and their latest upgrade. I’m a little late to the game, but rather than letting sleeping dogs lie I thought I’d throw in a developer POV. I’m not sure if Tweetie chose the best solution to their problem, but I do support their decision.

So read on after the jump for a developer perspective on this debate…

The Case for Upgrade Charges

Here is the bottom line – adding new features is expensive. Sometimes it’s even more expensive than writing the original application. This money needs to come from somewhere. There are four options:

• A constant influx of new customers
• Charity
• Someone else pays for it
• Upgrade charges

New Customers
One way that you can pay the development costs associated with improving a product is with income generated by new sales. If enough people buy the product, beyond the number needed to cover your initial investment and ongoing operations, you have money to pay for improvements.

This is wonderful – as long as you have a constant flow of new customers. The reality is that at some point after initial release you drop into something more akin to sustenance levels. In a worst case scenario, everyone who is going to purchase the product now owns it and the flow of new customers is so slow that it hardly makes sense financially to sustain the product at all.

When this happens, when the influx of new customers fails to do more than sustain the product, there will be no more upgrades. Period. There is no emotion behind this decision – it’s simple accounting. This is what we risk with the Apple store where every upgrade is free.

Right now, people have been happily using our products for more than 12 years. Over that time we’ve continually improved the products, adding features, supporting new operating systems, and generally maintaining them. We can do that because the upgrade charge model supports this. Take this away and … well, you get the picture.

Charity
You see this approach with the hobbyist developers who aren’t in this to make a profit. These guys will continue to add new features even if there is no financial incentive to do so – until they lose interest or get a paying job at least. Software developers who do this for a living can’t afford the charity model. At some point the money runs out.

Someone Else Pays
And don’t mistake “charity” with “someone else is paying.” Ad supported products, products that convince you to buy something else made by the developer, and similarly supported projects might look like charity on the surface but in the end someone is paying for that ongoing development. For instance, the money to develop our free applications comes out of our marketing budget. This works for a couple apps but wouldn’t work on a large scale.

Upgrade Charges
When 1, 2, and 3 won’t cut it, it leaves us with one alternative – we have to charge for upgrades. As I said, the money to pay for the development has to come from somewhere and traditionally this is how it has been done.

Under most software sales models the reseller offers a way for us to charge a reduced amount to existing customers for the upgrade. I mean, nobody wants to make you buy the whole application twice. This is one of the real problems with the iTunes App Store.  Not only does Apple not give us a way to charge for upgrades, they don’t give us a way to charge less to our loyal customers.

“In-App Purchases” I hear you say? Sorry. That isn’t the intent of that program, and a developer who tries to use this model to sell upgrades risks App Store rejection.

So what is a developer to do?
Let’s take a look at what Tweetie did. They start out by charging a price that is essentially an “upgrade price” from day one. Tweetie is easily a $4.99 application, but they charged $2.99 from the beginning. When they added a bunch of new features, they again charged the “upgrade price” for the application – $2.99. It’s a hacky, ugly way to do it but it works. Users who want the features can purchase the upgrade – users who don’t want them can keep using the old version.

It’s the same reason we charge $9.99 for eWallet.  eWallet for iPhone is a $19.95 application, but since Apple doesn’t give us a way to charge our existing users a discounted upgrade price (our upgrade price is $10 normally), we’re stuck giving EVERYONE the discount.

The problem is not Tweetie – the problem is that Apple hasn’t offered a long-term sustainable development environment and until Apple improves this, folks like Tweetie are going to pay the price.

Oh, and Microsoft – I hope your Marketplace guys are listening!

Marc Tassin is the Senior Product Manager for Ilium Software, a mobile software developer in Ann Arbor, Michigan. Marc has worked with Ilium Software for nine of its 12 year history, and in that time he has watched the rise and fall of many mobile devices, scores of products, and entire corporations. He joined the JAMM team to provide a developer perspective on the mobile marketplace and share some of his experience in the industry.