Palm Pre, FastFigures and powerOne

I have had a steady drip of requests since the Palm Pre launched for a webOS version of our software. I wanted to take the opportunity to talk about the platform, what I think Palm is doing right and wrong, and relate that to our own software products FastFigures and powerOne.

First, let me say that I like what Palm is doing with the Pre from a developer perspective except one major flaw. A year ago I wrote a post on the recipe for beating Apple and highlighted three things that the company needed to do. Palm has nailed two of the three:

  1. Build a beautiful, touchscreen device.
  2. Make it synchronize with web-based applications.
  3. Focus on offline use of web-based applications.

Palm is flubbing #3.

Palm has built their platform so that all applications are written in CSS (the page style), HTML (the content) and Javascript (the interactivity), the core languages of the web. But these three languages are intended to handle the user interface, or client, side of the equation.

What’s missing — and what Palm insists it doesn’t need — is the underlying technology that handles the business-side of web applications. Developers use a multitude of server-side technologies to do this, including Ruby on Rails, PHP, .net, and Python. Most mobile platforms use either C or Java to handle the business logic.

Palm insists it doesn’t need anything. And this is a major mistake.

Our software requires the business language to run the engine that performs all the calculations. Javascript won’t do primarily because of security and speed issues. In addition, insisting on using Javascript for business logic flies in the face of everything I learned about how to do web development.

Palm’s perspective is that applications that need business logic should interface with the web, such as Google’s search engine. Except an application like ours works best when the calculation is resident on the device, not because the calculations are better but because our customers don’t trust the Internet connection with their devices. There are just too many holes.

So what do our customers do when it comes to the Pre? They can either use the Classic emulator, which we don’t officially support but seems to run our software without a problem according to customers who have tried it, or use the web-based version of FastFigures at http://www.fastfigures.com/mobile. Either way, if you want our products on Pre, please drop us an email so we know.

And hopefully in the future, Palm will realize their mistake and give us a business logic language to work with. For now, though, I won’t hold my breath.

Psychology of iPhone Pricing

When I started in this business 12 years ago, there was tried-and-true consumer price points: $9.99, $19.99, $29.99, $49.99, $99.99. Only a few apps — Microsoft Office, Adobe Photo suites come to mind — were able to price above those levels. Anything under $19.99 was considered an impulse buy and anything above $49.99 was considered professional quality.

This never was in the mobile/smartphone/PDA world. In this world pricing seemed to break down as $10, $15, $20, $29.99, and $39.99. Only the rare app was able to get more. We charged $59.99 for powerOne Finance but bundled both the Palm OS and Windows Mobile version together for that price. If we had been in the old model of consumer pricing, we would have been between $100 and $200 (with slightly different presentation and functionality). Impulse buy was anything less than $15; professional apps were at $29.99 and above.

Now those prices have morphed again, at least for iPhone. For whatever reason — first products in the AppStore, lack of trials, single purchasing location, Apple’s devious plan, natural laws of commoditization, Chris Anderson (just kidding) — all pricing have depressed even further. It seems that the new scale is $0.99, $1.99, $2.99, $3.99 and $4.99. Only the rare app can charge more.

The interesting thing to me is that the mentality hasn’t changed. Now, $0.99 is an impulse buy and $4.99 denotes professional quality.

Case in point: In April, a competitor was at $5.99 then $4.99 then $3.99. Never really impacted FastFigures’ sales, but we could never catch up with him, either. For a week, he went free. When he went free our sales shot through the roof. When he came back to the paid side at $3.99 he made no headway. It didn’t impact my sales at all (I was way ahead of him in Top Paid) and the same was true when he dropped to $0.99. He sold a lot more, catapulted above me in Top Paid, but it didn’t impact my sales. Only when he went back to $3.99 did my sales start to fall off as I was clearly losing units to him and he was able to stay ahead of me in Top Paid.

One conclusion: Top Paid position has a clear impact on sales. I think most people try the 1) cheapest and 2) first found product that they were looking for.

Price, though, also impacts sales. So where’s this cut-off point that attracts the alternative, professional customer we were looking for? Is the connotation that anything under $0.99 is a throw-away app? Or is that point $1.99 or $2.99? Clearly, $3.99 has a different connotation but I’m curious if that “professional level” actually falls in somewhere lower.

Or is my entire theory flawed, and the reality is that software is diving toward $0 price points, as my Building an iPhone Business surmises? I can definitely say the market is shifting. And I don’t believe, unlike other developers, that it’s Apple’s fault. I think they are just accelerating the curve.

Brand Power

In the late 1970s, when Hewlett-Packard wrote the HP-12c, copyright laws were different. If a company or individual didn’t expressly exert their right then the code was freely available for anyone to use. To copyright software, the copyright notice had to appear visibly to the user on the device and it must be noted in the source code. HP failed to do this with their calculators.

What has transpired over the years is a plethora of HP scientific and financial calculator emulators based on the late-70s products, that actually use HP’s original source code. These emulators have been released for every platform you could possibly think of, including at least four HP-12c emulators for the iPhone.

Because there is little difference between these various 12c emulators in the App Store, the Hewlett-Packard official version released last week and the “knock-offs”, I have watched these apps with great interest. It has proven to be an amazing lesson on the power of brand.

The most expensive of these products pre-HP release was the most popular, peaking in the high-20s for Finance Top Paid Apps. The developer was charging $19.99 for it. When HP came on the scene a week or so ago, with a $14.99 price point, the knock-off apps dropped like a rock while HP sky-rocketed, topping out at #1 in Finance Top Paid Apps. (The knock-offs have since recovered somewhat with much lower prices.) No multi-featured calculator product, in the six months I’ve been watching, has been ranked higher than 11 before the official HP-12c release, and that was a $0.99 application.

What has struck me is the power of the brand. Again, there is no discernible difference between HP’s version of the 12c and the knock-offs yet HP was able to run them off. Second, I’m amazed at the power of this ancient brand. It has been able to propel it to the top of the Finance category when far more interesting products have never gotten as close.

And the buzz for the HP-12c has been fantastic. Before HP released the 12c, most of the calculator conversation on the web had been about the built-in calculator’s ability to be turned into a scientific calculator by turning the device sideways. (Or, more worrying, the propensity of iPhone users trying to dial with their calculator.)

While I’m not convinced that brand building happens in the App Store, I am convinced that a powerful brand makes all the difference. The question I ask myself, of course, is how do I build such a powerful brand with FastFigures? Working on this and other questions as you read.

 

[Note 1: I convinced HP to develop the 12c emulator for iPhone. I was not involved with the development and never asked for nor received compensation for my consulting time with HP. If, however, HP wants to throw some money my way, I’d be happy to take it.]

[Note 2: In case you are wondering, HP’s release of the 12c has not had a negative impact on my sales. FastFigures is not an emulator but instead takes a fresh and truely smartphone look at calculating on the go. If anything our Top Paid position has actually improved over this time period.]

Spreadsheet Redux

I’ve always taken the quote “Those who cannot remember the past are condemned to repeat it” (George Santayana) phrase to heart.

I’ve been doing a lot of reading on the early history of the spreadsheet. I believe that the process and capabilities apply to Infinity Softworks’ work with FastFigures. And I think, like the creator’s and publisher’s of VisiCalc, it’s something new and different that is defying standard definitions.

The following includes selected thoughts on this complicated story: Dan Bricklin’s thinking around the design of VisiCalc, how they marketed the product, the relationship with their software publisher, and how Lotus 1-2-3 replaced VisiCalc as the de facto standard in spreadsheets.

  • The real power of the spreadsheet was not it’s calculating capabilities but the fact that an end-user, with no programming skills, could write a program for the first time.
  • Everyone struggled to describe the spreadsheet in the early going. “Though hard to describe in words,” starts the musings of a Morgan Stanley analyst, “VisiCalc comes alive visually. In minutes, people who have never used a computer are writing and using programs.”
  • We think of the spreadsheet as a calculating tool. But many people use it for data collection and charting as well. It really is a broad and powerful toolset.
  • What’s amazing about the spreadsheet is that fundamentally, even today, it hasn’t strayed from Dan Bricklin’s design. He conceived a mouse-driven scratch pad. You type stuff into cells and then create formulas by pointing to other cells. He conceived of features like split screen (implemented in version 1.0), graphing, and basic data collection.
  • You have to be there to play. Lotus 1-2-3 picked the winning platform for its time — MS-DOS — and then improved the speed and performance. It also integrated features customers wanted but never implemented into VisiCalc: basic databases and charting/graphing capabilities (with calculation making up the 1, 2 and 3 in the name).
  • Lotus lost out to Excel not for a feature set but because they didn’t move fast enough to Windows. Lotus picked the wrong horse, staying loyal to IBM and OS/2. By the time 1-2-3 was ported, it was too late.

And finally, a few conclusions:

  • It strikes me that every change in standard platform has resulted in a change in standard-bearer for calculating tool. Before PCs, it was the calculator. With the rise of the Apple II, it was VisiCalc. With the rise of DOS, it was Lotus 1-2-3. And then windows interfaces — Mac and Windows — precipitated the switch to Excel.
  • Platform evolution seems to take on the following software curve: small applications, custom development, specialized applications, software platforms. Right now, small applications and custom development are dominating mobile, with specialized apps starting to rise in popularity.
  • I’m not convinced the spreadsheets make sense on mobile devices. Do I want to see my spreadsheets? Sure. But it’s not the way I think most people will use calculation on a mobile device. Of course, I’m betting everything that I’m right.

FastFigures v2: From Good to Great

Version 1.0 is always so hard. When we wrote version 1.0 of FastFigures Finance Calculator and released it in February, we really didn’t know what we were doing. We were trying to figure out everything from what Objective-C is all about to how to make the user interface do the things we want it to do. Even worse, we hadn’t used the iPhone enough to have a clear understanding of how we should interact with it.

So we create and change and throw out and create some more and over the course of a few months — from November to February — we refine it enough to have something decent. Decent, mind you. Not great, but good.

Should we continue on for another month or two and keep creating and changing and throwing out until we approach great? Of course. Would it have been better if we could roll out a beta program to a few hundred, get their input, and keep refining until we are closer to great than good? Yes.

But after four or five months of staring at the same code, good feels good enough. So we shipped. And the response was solid. In the first few months, we sold a few thousand units, tried to manage the unmanageable (Top Paid position and search), and learned a ton from talking with and listening to customers who were using it every day in the field.

And with a break, we were ready for our next try. We took all those suggestions and all that feedback, took our new-found experience with iPhone development, boiled it down to a couple of things that we could do in a few months, and got to work. What did we add to FastFigures version 2? The following, all focused on speed, accuracy and ease-of-use:

  • Pop-up editors for improved data entry in the templates
  • Start-up options to get you where you want to go quicker
  • Faster navigation to the calculator
  • Improved button sizes and layouts for easier data entry in the calculator and templates
  • Memory locations and constants
  • Integrated help

In general, the application feels more like an iPhone application, it acts the way I’d expect, and I’ve already heard from a number of customers who dropped their other products to solely use FastFigures. We didn’t ship ‘good’ this time. We waited for great.

And now, with the learning curve behind us and the core in place, we can really get to work. The plans are exciting: template creation, add-ons, saving data, syncing to the web site, report generation… I can’t wait to get started.