RIA Weekly

This week, Coté is joined by Intuit’s Jeff Collins to talk about the Intuit Partner Platform, or IPP, a ready-to-use PaaS for building on-top of QuickBase, including with Flex.

You can download this episode directly directly and it’ll also show up in the RIA Weekly feed for iTunes and other podcatchers. Or, just use the controls below to listen to it right here:

IPP is a very interested Platform-as-a-Service (PaaS) available from Intuit used for building on the existing QuickBase ecosystem. It uses Flex as the user-interface technology and there’s also a marketplace available to sell the applications in. More so than just using Flex, I think IPP and platforms like it are a space where RIAs will really start to shine.

As you may recall, we’ve discussed IPP before in the podcast, notable in episode #45 with guest Alex Barnett, also of Intuit.

Transcript

Michael Coté: Well hello, everybody! This is Michael Coté with RedMonk, one of your ever present co-host for RIA Weekly. This is a special sponsored edition of RIA Weekly, sponsored by Intuit. What we do here is we take a pretty good deep dive over the course of about 35 minutes, if I recall, into the Intuit Partner Platform or IPP as we’ll refer to it, which I think is actually a really interesting platform that Ryan and I have talked about a couple of times, Ryan Stewart, the regular co-host, in reference to the sort of Cloud based ways of doing rich Internet applications.

So with that brief introduction, let’s get right into the special guest introducing himself. Would you like to introduce yourself?

Jeff Collins: Sure! I am Jeff Collins, I am the Group Architect of Intuit Partner Platform and I work on the technology side and making sure that all of the pieces work together well for the benefit of the developer and for Intuit’s platform.

Michael Coté: So for people who aren’t — I am sure pretty much everyone is familiar with Intuit as the people who make QuickBooks and TurboTax and all these other sorts of things like that — But for people who aren’t familiar with IPP, can you sort of give us an overview of what IPP is and kind of the development ecosystem it fits in?

Jeff Collins: Sure! Well, a lot of people will probably remember that QuickBooks has had a pretty vibrant developer community around it for a number of years, and what we thought we wanted to be able to do was move from the desktop model of QuickBooks and plug-ins on the Windows desktop to a much more web world and be able to do a Platform as a Service on the web so that application developers could go, place their SaaS application on our platform or federate their application on our platform and integrate online with data synchronization technology to QuickBooks desktop, but then also other SaaS apps that Intuit provide.

So we are translating our existing developer frameworks for the web so that developers can get access to all kinds of different business data that our customers store and have a place to run their application, build their application, manage their application, and run a business and make money, which is a key part of Intuit’s Platform as a Service effort.

Michael Coté: Along those lines, it seems like a fair amount of the companies and developers, who are building these sorts of extensions to QuickBooks, so to speak, are themselves — they are smaller companies that are also servicing smaller companies who might be using QuickBooks. So how would you characterize the types of developers who are working on things in extending QuickBooks?

Jeff Collins: Yeah, it’s a great question. I think traditionally you found a lot of smaller developer businesses on the desktop like model. In the Intuit Partner Platform version of our developer platform, I think you find two classes: you find the existing class of developers who are working as a small development shop that it’s tapping into a great Intuit channel and small businesses and really adding value to our products. The truth is that Intuit can’t build everything for everyone. So there is plenty of opportunity for small vertical solutions or great horizontal extensions to our core small business, management capabilities to just put in their application, and those will be smaller businesses that really are looking to tap into what QuickBooks just — where it stops or where our online products just stop.

What’s interesting is we are finding now that since we’ve gone to a more SaaS web model with our platform, and we’ve introduced the concept of federation where data can be shared with single sign-on with Intuit. The other opportunity that it brings to the table is established larger businesses that run SaaS applications already that just want to integrate and federate their data and create kind of a, I guess, shared workload would be one way to say it, where at some place QuickBooks stops or our Intuit products on the web, QuickBooks Online and other products like that stop, and you go into another entire workflow hosted by these federated developers.

Michael Coté: Yeah, it sounds like — and we should get into this next, but — it sounds like when you have like the sort of SaaS or Cloud based or whatever you want to call, when you have an online Cloud based SaaS version of your platform essentially. I guess you are getting into something that’s kind of intuitively make sense, but I hadn’t kind of put together that it seems like it’s a lot easier to do integration across different systems and if you had a desktop bound behind the firewall sort of system. I guess that is like an exciting sort of new thing that I’d see among SaaS things. It’s a lot easier just to kind of like switch to pass off the baton of integration if you will between a SaaS service than something that’s a complete desktop application, which is what we are seeing here.

Jeff Collins: Absolutely, and just to kind of elaborate on that. What we found was with the old way of doing things in the desktop — which by the way, we still have lots of people using those applications and building those applications, but — on the desktop what would happen is you have a connector and the web connector architecture would require that you have different credentials entered into them for the user services subscribed to that they want to integrate with QuickBooks. There was no real single sign-on technology in place. So it is a little bit odd that user might even be probably uncomfortable entering all those credentials into the local desktop, so it can make a room of connections.

Now what happens is you have this SaaS platform that Intuit is hosting and it has single sign-on framework built in. When you sign on to Intuit, you signed onto everything that’s federated to Intuit including the add-on services. So now not only is single sign-on part of the QuickBooks experience, it also — QuickBooks synchronizes data to the cloud so the federated partners don’t even worry about the desktop connectivity and all the integration stays together.

Michael Coté: Yeah, I mean with that, since you’ve thrown out the old single sign-on there, let’s get into, dig what exactly the IPP platform is. So like all platforms I always joke that, I am always way too food oriented, but you’ve always got a cake or a burger it seems like, and there’s kind of like those different layers of the application that you have. So what are the different sort bundles of services or layers or all the boxes on the diagram that IPP offers?

Jeff Collins: Sure, we also know that in the architecture community is a Stackitecture, which is the –

Michael Coté: There you go!

Jeff Collins: — highest level of marketing layout of your software, and it’s more educational than the actual technology.

So Intuit Partner Platform, like any Platform as a Service has a number of layers that probably the bottommost layer and something that a lot of folks already know is that we started with an existing Data as a Service platform, Database as a Service platform called QuickBase. So inside of that core underlying layer there was an application registry and application engine. There are all kinds of file storage, there is baseline security which includes identity and authorization and permissions. There is a capability for, as we just talked about, federation where data can be interchanged between partners for the creation of a workflow across enterprises, across systems.

So it’s very core, the fabric of Intuit Partner Platform includes all those basic services to run a Platform as a Service. As a developer I have already discussed the two different types of apps. There are native apps, which is when the developer will build an app and then upload it to our Platform as a Service, which we will then execute. We are the operating system for the app and platform. Then there are federated apps where we established a data interchanged connection between an existing SaaS app that’s hosted elsewhere and Intuit Partner Platform. So all those things run in the lowest level of services.

Michael Coté: So just to make sure I understand it. So essentially, I guess, this is what you would call the Platform as a Service or a PaaS, if you will, if you kind of divide the world into these three — if you divide the Cloud into the infrastructure, and platform, and SaaS. It sounds like what you are referring to is like the native way of running the app. It’s basically doing a Platform as a Service where, as a developer, you bundle up an application and upload it essentially to IPP and it all executes within IPP, versus more of a federated app where you are actually, you being the developer or the organization running it, you are responsible for running a part of the code on your own that just interacts with IPP and thus with QuickBooks and everything. I mean is that a fair enough characterization of it?

Jeff Collins: Yeah, that’s a characterization and the reason why we have both is that some businesses, it makes a lot of sense for them to build an application hosted on a platform or a great channel for them and existing SaaS apps that are already in the marketplace and accessible on the web, don’t need to rewrite themselves. So it’s just better to leave them where they lie and allow them to integrate with our platform.

Michael Coté: Yeah, that makes sense. I was just going to ask you to get into the other kind of services in this Stackitecture there, specifically like — or whatever you are thinking of next, but also there is — we keep getting at this thing that you can interact with the business data and the processes and things like that, and in addition to things like single sign-on and identity and sort of I don’t know how you guys characterize it, but the kind of like table stakes, if you will, or the basic infrastructure things that you need. It would be great to hear about the more Intuit QuickBooks specific stuff that’s in there as well.

Jeff Collins: Yeah, so one level above the basic infrastructure, so you can think of the infrastructure as almost the application housing or the application engine that hosts applications and/or connections to federated apps. One level above that is kind of more of the basic infrastructure. When you think about putting an application on a Platform as a Service, what Intuit wants to provide is not just core infrastructure, we aren’t trying to be in a long-term just a place to put bits and execute. We are trying to build a channel and we are trying to build a place for end-users to discover the apps because they use other Intuit apps. So most importantly, we are trying to help developer to make money. One of the key aspects of being able to do that is you need to be able to sell your app, and we want you to sell your app in our channel. The next level above of services has in it billing capabilities as one of the core components.

So when you build an app, whether it’s federated or whether it’s native, you are able to go in and configure a set of plans and a set of prices for each plan, and each plan might represent some capabilities that your app provides. Then when you offer your application, you are able to check to see what you sold the current customer and then enable or disable features accordingly. Some entitlements are system entitlements, like for example, you can sell a number of users per tier or you can sell some disk space per tier. So the business engine that behind everything in our Platform as a Service will actually evaluate all of the configurations that you, the developer, has done and you can change prices and re-promote it anytime. So that’s another level of capability up at the application capabilities level.

Michael Coté: Oh yeah, I mean it sounds similair to what everyone probably experiences with like their cell-phone company or cable company where there is — you have a pricing and billing model that’s kind of — and the engine behind it that’s complex enough that you can come up with all sorts of things to charge for beyond like a flat rate and then adding on additional services and metering things and stuff like that, which to your point of — I always like it when people, in a sentence, try to get like “developer” and “money” as close as people, because that kind of gets to the heart of the matter. But to your point, the more ways, the more plans you can have and the ways that you can have to collect, the more marketing segmentation essentially you can do. I think that’s interesting; you don’t hear a lot about that for many infrastructure sort of Cloud based platforms. I mean there’s all this other stuff on top that’s kind of like the boring actually collecting the money part, which is interesting to hear about.

Jeff Collins: Yeah, exactly. So what’s really key for our developers is that we are trying to take care of them and make sure that it’s not just infrastructure, because that’s not the whole decision that developer is trying to make. A developer has got an idea, they’ve got a business, they’ve got an expertise, and we are trying to give them an avenue to tap into a lot of end-users that work with our products today. Small businesses, like I said before, use our products and have great success with what would be almost considered the fact of standard small business management tool, but we can’t build everything. So we are trying to have developers come, help us. So it’s not just a question of where your bits are running; it’s how you perform with your application in the market with developers in our channel.

So that’s an example of another capability that’s one level higher, it’s at application level. Another one that I think I referenced before though is that this only works if you can get really good integration with our products, and then make your business, whatever that is as a developer, seamlessly any way for the benefit of the end-user, whether you are in adjacent space to finance management or you’ve got a product that does great marketing to the set of customers that are tracked by the small business, this all kind of breaks down, if you can’t talk to the list of customers or the list of financial transactions that are available from the business.

So built-in into its Platform as a Service is a synchronization with the desktop version of QuickBooks, and what that means is QuickBooks has sync setup and is able to — with built-in support, automatically transmit data down to the Cloud, which becomes an API that you can call as a developer on into a partner platform, and then changes that you write into the web services will be synchronized back to QuickBooks and take forth that on all of the different side affects that changing an entity like a customer or a financial transaction would have in the desktop.

Michael Coté: I think this is a good point to kind of clarify, what would you call, a kind of the end-user, the different end-users, the experiences that people might have. So is it solely the desktop applications that your end-users will be coming from or are there other sorts of places the end-users will be interacting with the system?

Jeff Collins: Well, so we have desktop integration with QuickBooks, because it was deemed to be such a great opportunity for developers, because there are so many customers of QuickBooks, 4 million at out current count, and that’s what amounts to something like 25 million employees. So that’s an amazing channel, if you can get that data available.

But that’s not the only API we have. In fact there are several things in the works right now that already have API and I think you can kind of divine the answer of which products that it might be that aren’t desktop at all. And we want our developers to be able to integrate to those and we just haven’t released that integration yet, but those are the things that are in the works today and we are pretty excited about the next way of opportunity to jump into the channel that’s represented by some of our non-desktop products, and that would include some of the resent acquisitions that we have done like the acquisition of PayCycle.

So we are looking at those opportunities now and more news on that will be coming out soon.

Michael Coté: Oh Yeah! I always like using Mint as well, which you guys recently acquired, if I remember it correctly, and that’s a service that would be fantastic if people could extend or started extending it. It would be nice.

Jeff Collins: We have a lot of plans for some of the products that we have recently acquired and we are getting together some messaging now on what we are doing with that.

Michael Coté: Definitely. So, I feel like we are kind of at the top of the middle of the stack essentially. So as we get towards like the top level of a stack, which is inevitably the user interface or different interfaces towards things, like what’s like that last layer of the Stackitecture?

Jeff Collins: Yeah, yeah. So at the highest point you still got to go in and out and especially if you are a native developer, you really need again, to have the platform be very easy to use, so you can get out there quickly and make money.

So what we have provided — and this is our basis and Flex technology, is that we have actually provided a pretty complete toolkit just to get started and really producing application very, very quickly. What’s interesting about Flex toolkit is that Flex has a technology that’s great and it makes really easy to use applications.

But our platform also supports all of the capabilities we just talked about, everything from user management to metering and provisioning of applications, doing the subscriptions, checking out entitlements, and what’s really kind of nice for developers is when you use our toolkit, all of those things are already prebuilt as UI elements and APIs that may take an advantage of those features really almost out of the box. So becoming a citizen of the ecosystem with your app is as easy as taking advantage of the Flex framework that we have provided, and you’ll get a user management UI by default out of the box, a bunch of advance integration into our framework and just generally be part of ecosystem quickly.

Michael Coté: Do you guys have a fair amount of like prebuilt Flex widgets and things like that that work with the various models that IPP has?

Jeff Collins: Yeah, absolutely, and we have got a bunch of Flex UI controls, so to speak, that are pre-integrated to data sources that come right out of QuickBooks data or data that you define in your natives schema, and the binding is set up to work so that you just start using the UI widget or control and data will be read and written to the appropriate table or to QuickBooks on its own automatically.

Michael Coté: Alright, I guess the other advantage of that being sort of using the out of the box UI controls is that the end-users, that developers are targeting here, are already familiar with the same sort of, what would you call, it’s kind of motives and UI sorts of metaphors that users are used to in their existing QuickBooks stuff.

So you are not sort of — you are not foisting the Winamp problem on them where you have got like a multitude of weird different interfaces that they have to figure out how to use.

Jeff Collins: Yeah, and part of that is the style guide and having technical standards, we have a card-check process, we want to make sure, everything agrees at that level. In the way that you described, we want to make sure that things are pretty consistent.

But there’s another force that work and that is that, we want developers to be successful, and I think what developers are finding and will find is that the forces in our market work best, because we have ratings and reduce. Those applications that fit well within ecosystem and seem consistent and help the user the most, because of their consistency, are going to find the best kind of reviews out of the box from end-users that are picking up these applications. I think we are trying to help out with what the standards are; I think pleasing the end-users is obviously to the advantage of the developer and both things work together.

Michael Coté: That’s right. I think happy end-user plus developer equals money. To narrow down to that fun of getting those two words as close together as possible, so that definitely seems to be the case. I think — you are getting one of the other things that I think is exciting in this sort of “post iPhone era,” essentially and that’s having a market place, we actually sell these things.

So can you tell us a little bit about like — so once we have through this Stackitecture kind of built our application, what’s the actual Intuit market place that people can sell those stuff in?

Jeff Collins: Yeah, that’s a great point. So, this is all headed for something like, you build all the applications and you have completed your code, you’ve got it through review, who is going to find it and how? And absolutely we have a whole marketplace and the channel word I keep using is a place to really sell your application and where Intuit’s own users are going to be presented with the options to buy your app.

We call that the App Center, and that store is a listing of the developers applications that we have worked with. Some of those applications are natives, some of them are federated, and what’s also interesting is that a lot of the applications are built by Intuit as well. We want there to be in very little difference, between how applications are presented in the App Center, whether they are built by Intuit or not, because we want to show that developers have good footing with the Intuit users that come in and want to have somebody help them out with the business problem.

So when you go into our channel you get to choose the price, you get to choose how your content appears, you get to upload screenshot or working on being able to upload even more content for applications and really sell yourself. Then when the end-user feels comfortable, they just have a couple of clicks and they can acquire your application, it sits in there “My Apps” page, which is one of the ways that the user can find apps is they log in and see, which ones they’ve already bought. That store can continue with new updates, as often as you want to put them up there, and you can market directly to our App Center with your Goggle search words, if you want to, and we can actually sit down with you and help you for how to figure out how to market your application and get it down. So, we provide all that out of the box with our App Center.

Michael Coté: Oh! That’s interesting. So, that’s kind of the equivalent of SEO assistance, Search Engine Optimization assistance, for the App Center that you can offer people, the kind of picture they get.

Jeff Collins: We try to help. Yeah, so we are learning about, what developers are looking for and what we do is we just want to cure, how are you thinking about trying to get your app discovered and we will try to figure it out with you, how to set that up fast, so that they are found on our App Center.

Michael Coté: Right, right. And while we are on the topic of the App Center, can you give us a sense of how many apps are in there at the moment, and kind of how that marketplace is doing?

Jeff Collins: Yeah. So, right now, we got about thirty five applications in the App Center and there are more than that and some of them are hidden by choice of some of the developers. For whatever reason there is a few that they want a different channel port, but there’s about 35 and there’s a lot more on the way.

So far, we are tuning up our matrix and understanding what people like and don’t like about those, we actually put in a hold, there are three major touch points in the QuickBooks products itself, QuickBooks 2010. One is that customer and manager. If you see that on QuickBooks 2010 and buy it, you are going to be led into the context of IPP and App Center, so other apps are discovered there. And then, document management in QuickBooks 2010 is also another IPP application, it’s native app, and when users use document management, they get marketed and sold apps on App Center as well. And then, there’s just simply a button in QuickBooks that is App Center. So it leads users into the choking experience itself.

So it’s a really nice channel. We are in earlier days of calculating metrics and we talk a little about that with the developers. But so far it seems to be going in pretty well, getting a lot of traffic there.

Michael Coté: Yeah, I mean the tracking and metrics and stuff, I find that stuff fascinating, just because it’s sort of a new science, if you will, when it comes to this, I don’t know, to use the buzz word phrase to kind of Cloud based applications or applications that aren’t necessarily desktop or Softwares as a Service. There is a whole green field of opportunity for what the best metrics are to track on going to kind of figure out how to tune your application to a customer, whereas in the past, behind the firewall or a desktop bound application, it seems like at best you can have the trash reports that get generated, and then you can sort of keep track of the number of support calls you have coming in.

But when you actually have theoretically, real-time access to what your customers are doing and being able to collect all of that, it’s like it opens up a lot more interesting opportunities for making your software a lot easier to use, to kind of cut to the chase of it right. In theory, you can cut down that window that it takes between a user being frustrated with the software and then figuring out someway to fix that frustration.

Jeff Collins: Yeah, absolutely. We built in some support for tracking your application and its purchase performance and some usage, you get a free integration to Google Analytics and that’s how we help the developers when you build an application on our platform. So there’s a feature where we embed a bunch of APIs for you to call. And yes, absolutely, without being able — everybody always says, when they finally started seeing their metrics, like I can’t believe I was flying blind for so long.

Michael Coté: Yeah, it’s kind of like, I always have this — I am not lucky enough to develop software anymore with the analyst role that I have. But I have this, in the future when we we have all this SaaS stuff, this fantasy meeting where you sit down and you can like very quantitatively say, no one ever uses this feature, therefore, we should get rid of it. That would be almost like one of the top five developer dreams that you could say, this feature is used or it’s not used and all this effort we are spending on it, we should either amplify it or not amplify it and that would be fantastic.

Jeff Collins: Yeah, I was just having that conversation, with an internal developer this morning, who was creating histograms of data, based on usage to show, prove a point about — they had a document and an invitation object and they were trying to prove a point that the invitations were useful feature and they wanted to continue with that feature.

Michael Coté: There you see, and there’s the positive angle on it, like you need motivation to keep something up that for whatever reason people are I guess no pun intended in correctly intuiting that like no one is using it, but you can prove that it actually is useful, which is equally valuable, if not more actually.

So like I kind of the forced the cart before the horse a little bit here. So maybe we should — I just want to get back to the horse before we wrap up, and that’s you were kind of — you were talking a little bit about obviously using Flex and this built into the tool and all of that. But I am also curious to hear what — so if I decide I want to us IPP essentially, like what’s the tool chain that I get and what’s kind of the integration to existing tool chains, like once I get my fingers on my mouse and my keyboard what’s the two experience going to be like?

Jeff Collins: There are two things to say there: one is that we do webinars regularly to talk to developers, and so one of the things we have talked about is, what the role of Flex is in our platform especially, if you want to be a native developer. We have come out and said that Flex is an awesome part of the platform. It really is a productive environment for developers. But we have also, said, that we are working on something called the Server Business Logic which is going to be also a different way of building apps, where you can actually host your own kind of UI technology and you could use HTML.

So we are working on having a broad spectrum of tools, but in all cases, what that boils down to is there is a toolkit, which is an extension and today it’s an extension of the clips. So if you — where the rubber meets the road, when you go down and when you get your first downloaded of IPP developer tools, it’s going to plug into Eclipse and it’s going to be a nice easy to use deployment age for when you have build your standard Flex application, and then run a few process increments on it, and you can upload it to our Cloud. Then as we announce some of these more, our newer options, it’s the same kind of thing. We make an easy to use plug-in and development environment of choice, and then allow you to publish to the Cloud.

So the idea is that we always want to have just a minimal footprint or a minimal impact and then use standard technologies and standard built processes, the developers choose, just our last step is how to get that thing published. There’s the other angle as well, which is that if you are a federated developer and you want to just get going, and you want to get live.

For federated developers that actually just turns out, just use the web to configuring and there is no real equipped ties-in at all, and you can, within a couple of clicks, get at least a test instance going, because it’s really just a single sign-on integration. Then you look at some integration technology at the server side for back-end SSO integration, which is a little different than UI development in an Eclipse project, and it has its own sets of challenges, but it does not involve tools.

Michael Coté: Right. I mean there are just more frameworks and libraries that you might use that make it easy to integrate on the server side with the federate essentially.

Jeff Collins: Yes.

Michael Coté: Yeah, well that’s pretty interesting then. I am glad we got back to the horse a little bit. So unless you have anything else you think we need to tell people for sure. I think that was actually a pretty good overview, and it was fun talking with you about it.

Jeff Collins: Yeah, and I think for developers that want to get to the next level of the story, you can just go to developer.intuit.com and get your first download and see how you like it.

Michael Coté: And like you said, you have — is it monthly webinars that you have, if I remember, they are pretty frequent actually.

Jeff Collins: We have been doing them as — it’s usually quarterly, so that we have time to actually stuff, so we don’t have to just keep talking and talking and actually do it.

Michael Coté: That’s right. You need some time to actually code for the coders, right?

Jeff Collins: Yeah, that’s right.

Michael Coté: That’s right, that’s right. Well, that sounds good. Actually there was one last thing I wanted to ask, and we got to this a little bit, with the new things you had acquired. What can we expect in the next, in the future, like what’s in the roadmap that you guys are excited about this happening with IPP?

Jeff Collins: Well, we have a couple of things going on, not all of them I can go into great detail on, but we have been out there before speaking about our server initiatives and being able to do different kinds of development on the platform as a native developer, so that’s exciting and we’ll be able to give modern use Internet.

We got some changes in the way we are going to do identity. So we are working on integrating OAuth and that protocol is very helpful, because it allows — because it’s a much easier way to secure mobile devices. So we are starting to give into — closing some of the authentication issue with mobile, in general, and making things safe for development there. Then we have some pretty interesting stuff that we are working on with allowing other Platform as a Service vendors to have integration toolkits, so that their customers can build apps quickly and they are only federated into IPP. So we are working on that now.

Michael Coté: Oh! That makes sense. Yeah, that’s interesting what you said about OAuth being a good mobile authorized. I mean that’s certainly, for anyone who has used like Twitter on Android or the iPhone or whatever it’s — from the end-user prospective, it’s actually a very easy way to do that sort of thing, which has always been a — I think in the sort of science, if you will, of identify management, doing delegated authorization has always been pretty terrible.

I mean, back on the developer end. I am not sure if it’s necessarily any better, but at least on the user end, it’s extremely easy to use OAuth which I think is what matters most.

Jeff Collins: Absolutely, absolutely.

Michael Coté: Well, that sounds great. Well, I appreciate you taking all this time to go over with us today. So thanks.

Jeff Collins: And thanks for the opportunity Michael.

Disclosure: Intuit is a client, and sponsored this podcast.

Share This