Ian Landsman is Starting From Scratch, July 12, 2006:
Hosted Feedburner
If you're in the market for a powerful and user friendly Help Desk solution, please take a look at my company's flagship product HelpSpot.A nice idea by scrivs. Everyone is so caught up with SAAS (software as a service) that they constantly overlook the fact that most IT shops (and many normal people) don't like leaving their data in other peoples hands.
I think many web pro's gravitate to SAAS because they're lazy. They think it's easier to deal with because everything is on their servers and it's quick to make changes and you don't have to convince people to upgrade, etc etc. I think most of those statements are actual false, but even if they are true none of those reasons have anything to do with your customers. Those aren't your customers needs their your own. Of course there are some apps that work great for SAAS and some customers will always want the hosted option and that's fine, but it's not a panacea. There are many many companies and people that want to maintain control of their information beyond simply being able to download it.
Discussion
I totally agree about the SAAS problem. I want to collaborate on a new idea with someone - I DON'T want that idea on a publicly hosted wiki service.
However, in my target industry, I have been offering a real estate application as SAAS and it has been a great selling point for most agencies, however, I do feel that some customers would be happier if it were hosted in-house.
The big problem is cost, setup, and support. What do we charge because updates are much more complicated, they have to purchase a database solution, server, backup solution.
Supporting is tough because we have to login to their network using VNC or some similar solution, creating network issues, etc, etc.
Where do we STOP our support? If it's a problem with the server, how deep do we dwell into the problem before dissappointing the customer with "it must be the server"?
I don't know if tagging it as "lazy" is correct, because we are certainly not lazy. We are more concerned about upsetting the business plan that is working well right now. Also, there could be many more costs involved that we are not aware of - like protecting the IP and time to install.
Questions we ask ourselves:
Is taking another risk while the company is steadily growing a good idea?
When is a good time to take another risk?
Created by Ben Mc on 07.12.2006 11:07 am
Definitely good points Ben and certainly not all SAAS models are because of lazy developers 
I'm mostly referring to the folks who think any software that isn't SAAS will be obsolete in a few years.
That said it all depends on the market. I expect that small to medium real estate firms have pretty much no IT department. In that market a hosted option makes tons of sense. On the other hand if your product also has appeal to larger agencies you're probably leaving lots of money on the table. They'll want to host in house as they're already spending lots of money for IT people, backup systems, etc.
Usually offering a self hosted version will allow you to charge many times more for the software than a SAAS version. Since you already have the SAAS version though it may be hard to justify the higher price without some other benefits since potential customers can see your current pricing that already comes with hosting.
The support issues are what they are and you'll probably have to help with some system setup. In my experience it actually isn't that bad.
Created by Ian on 07.12.2006 11:07 am
Ian, it is very interesting watching your success and all the hard work that goes into UserScape - thanks for including us. I have to agree with you that most of the assumptions that developers may be making about SAAS are false: hosted applications are not, in fact, easier to deal with. Providers of hosted services are on the clock 24 hours a day to ensure the reliability of the system and safety of their users' data.
I also agree with you that non-SAAS software will continue to be the "norm" for the future.
That said - I am an advocate of SAAS because of the ability for users - especially small offices as mentioned by Ben - to begin using products immediately. Most small offices do not have IT staff or even IT literate, or IT-interested, people. When given a choice to work through purchase or download, installation, and configuration steps on their PC or start using a product over the web, many small business owners would rather skip straight to using the product via SAAS. SAAS also provides a method for small offices to share information in locations where a LAN may not be present.
Starting out, small software companies needs to focus on small firms because the sales cycle for larger firms is too long and there is too much risk in spending time and money selling to larger firms. Many small software start-ups need to target firms in which a purchasing decision can be made by one person (usually the end-user of the system). Like most things in life - the risk and reward ratio is in play - selling to large firms can have, as you state, large rewards; yet can also leave you having spent large amounts of time with no sale.
As a web shop, I gravitate toward SAAS because of the opportunity to target the market I just described.
The way to break from SAAS to self-hosted for larger agencies would be to ensure the SAAS versions have size limitations (e.g. max size of customer database if CRM). Offer a self-hosted version that appeals to larger agencies because it does not have the size limitations.
Created by Scott on 07.12.2006 11:07 am
I think a major SAAS-obstacle that a small-ISV must deal with is the reliability of their web host. Because unless you can have good countrol over the uptime, your customers will blame you when they can't access their data.
So unless a small-ISV is willing to invest in the network (e.g. servers, a good connection, colo, etc.) SAAS doesn't seem feasible.
Created by Jake on 07.12.2006 11:07 am
Just to add a point to the FeedBurner part, it is quite obvious that there are those who would prefer a self-hosted solution rather than go with the free FeedBurner service to avoid lock-in and the various problems hosted services have[1]. So while I do not agree with the wording of the post that SaaS is pretty bad and mostly due to laziness, I do agree that there is a marked for a self-hosted version as well, either by FeedBurner or someone else to complement the service offering.
For a primitive feed statistics self-hosted program you can look at http://baruch.ev-en.org/proj/feedstater/ I created it for my own use but the complexity of making it a product seems to be a bit too much for me for now. A problem in the case of feed statistics is the need to update the software about new bots to be ignored.
[1] They actually had lately quite a few problems and a few blogs I'm usually reading had their feeds unavailable or mangled beyond recognition.
-----
Created by Baruch on 07.12.2006 11:07 am