The Cloud business case

The thread about confusing Cloud terminology has spilled over into whether or not Cloud applications are cheaper than desktop accounting alternatives, with Internet Accountant Richard Messik stating the case for and Chatman arguing against.

As he has claimed in his blog, Richard thinks Cloud apps are more convenient and greener as well as "generally cheaper". Chatman countered by mentioning the £125 cost of his VT Accounts program (with access to a free Cashbook) that he didn't expect to upgrade for 10 years.

In the interests of broadening the debate, even the analysts at Gartner are having a hard time swallowing some of the Cloud industry's Hype. In a report dismantling five myths surrounding on demand technology, Gartner argued that the cost advantage eroded after the first two years as users continued to pay their subscriptions. The study is covered in more detail by our sister site BusinessCloud9.com, whose editor Stuart Lauchlan frequently says the difference between Cloud and on-premise software is like renting or buying your house.

If anyone can come forward with some solid figures to quantify the benefits and cost comparisons, we might be able to put forward a better informed business case for the new approach.
__________________________
John Stokdyk, Technology editor

Comments

Pages

Hidden costs are the leveller ....

Anonymous | | Permalink

A major differentiator between desktop & web apps is the fact that web costs include far more than an outright purchase

  • ongoing upgrades to the software - (look at Sage Line 50 & its 18 month upgrade cycle)
  • automatic backups (purchase of device/media, time costs to backup & restore)
  • etc ...

Now work out the additional cost of the aforementioned areas & add it to the cost of the desktop app

'.. Chatman countered by mentioning the £125 cost of his VT Accounts program ... he didn't expect to upgrade for 10 years ...'

All well and good but how often will he have to reload the app on another PC over his 10 year span & in this respect he is vulnerable to chages in the operating systems; which may render the app useless unless updated

Once again his time costs with installing etc should be added to the purchase price

It is going to be difficult to factor all these things in and especially in the case of Professionals who may use non-chargeable time (£50ph +) to trouble shoot their own problems. At £50ph it doesn't take long to wipe out any desktop savings

WhyNot

Inventing Problems

chatman | | Permalink

Simple instals are child's play; if you cannot handle them, your problems are much more serious than whether to choose on-line or desktop. 

I have been running Windows XP for years, but I understand all XP software can run on Windows 7. Never saw any need for Vista.

As mentioned previously, the much-vaunted backups retained by your on-line software company may not be in the format you expect, or one that you can use.

And if we are looking for hidden time costs, what happens when the internet connection (or the supplier's server) goes down?

daveforbes's picture

Commercially viable

daveforbes | | Permalink

VT is a very good niche product with a user base of 3000 or so accountants. If the users were paying on average only £12.50 a year for their software then the total turnover would be under £50K it would not be a commercially viable venture. Maybe chatman is only paying £12.50 a year, but others must be paying more.

VT

chatman | | Permalink

I am sure they are, but my point stands.

I understand VT sells other products too.

Out of interest, how many users do the on-line accounting software providers have?

Niche Product?

chatman | | Permalink

Incidentally Dave, why do you call it a niche product? Because of the small user base, because it is only useful for a narrow range of users, or another reason?

daveforbes's picture

Product aimed at a niche market

daveforbes | | Permalink

Software for accountants, software for dentists, software for vets - niche market product. There are say 20,000 accountancy practices in the UK. VT sells, very successfully, to this niche market. Maybe I should have said specialist rather than niche.

Mass market would be where you may have tens, hundreds of thousands, millions of users. I am not sure what the various cloud vendors claim, but probably none would admit to less than 4 zeros on the end !

I was just pointing out that I did not think your example of paying £12.50 a year for VT was a particularly good example to bring up as I don't think it is typical for a VT user. The cost to the software company is developing, distributing, supporting. Are these costs any different for a piece of software you download and install on your computer compared with something you run in a browser ?

Maybe the debate should centre on the relative pros and cons of the two technologies rather than what the vendor charges. Certainly in a market where much software is paid for annually anyway. We may well produce "cloud" versions of some of our apps, but I can't see us charging a different price than for the installed version.

re. "Hidden costs" - you can still choose to back up data automatically to somewhere on the net without having your apps run there.

There was a pull out in the Times today - now I feel quite inadequate not having a "federated cloud infrastructure". I'll put it on my xmas list.

VT's user base

vtsoftware | | Permalink

I would just like to add that that VT's user base is considerably higher than suggested above. We have sold over a thousand copies of VT Transaction+ in the last three months. That number excludes VT Final Accounts, which comes with a complimentary copy of VT Transaction+.

Philip Hodgson
VT Software
 

daveforbes's picture

No slight intended - VT

daveforbes | | Permalink

I thought the poster was talking about VT final accounts which is aimed at a niche market, rather than the VTtransaction+ which is mass market. Still an average of £12.50 per year seems low.

 

The Big Picture - Benefit & Cost & Time Savings & Good Fit

onlinebeancounter | | Permalink

In the discussion about desktop vs online or cloud computing, I believe there are certain types of businesses that will see the benefits/value of this more readily than others. For example, ecommerce type businesses where the customer can do self serve to buy products, etc. and the accounting is an automatic flow through to the back end. Namely, sales are recorded, inventory relieved and on hand stock updated as well as payments received via credit cards or A/R setup up. All with a click of the mouse by the customer. And that's just for starters.

The case

Greyoltwit | | Permalink

Can we turn the argument round a bit? How do we make the case to an SME, for whom computers are an innovation, whose MD is penny pinchingly mean and still laboriously backs up to tape daily, which tape has to be taken home by a member of staff? The Cloud means nothing to him, not would I even use the phrase, and the firm is locked in, like so many others, to dear old SAGE with its attendant quirks and inflexibilities. Yet he happily uses email and Skype and as he travels extensively, even vaguely understands the benefits of remote working, but is obsessed by the need for physical control of his resources. Thoughts?

The case

chatman | | Permalink

Tell him to use a desktop programme.  Cloud may be right for some, but why try and change him when desktop seems ideally suited to him?

I must say I can see where he is coming from, not wanting to trust his back ups to someone else, although I must admit I prefer on-line to tape, myself.

daveforbes's picture

backups as part of the business case

daveforbes | | Permalink

Should "automated backup of data" be counted as part of the business case for cloud and how do you quantify the value ?

Online backups are available at low cost or included with the internet package and usable by using desktop as well as web-based software. Looking at it this way the value is negligble.

However, particularly at smaller businesses, the backup regimes are often inadequate with potentially large costs in the event of disaster.

How do you value "enforcing sensible behaviour" ?

 

david_terrar's picture

The Business Case for SaaS and the Cloud - watch this space

david_terrar | | Permalink

Only just read this thread, and there's a lot to be said here.  I'm hoping John publishes a news item on something that will help in the next few days (he knows what I mean), but it has to be said that there aren't enough good case studies of TCO comparison between SaaS (or Cloud solutions) and the equivalent on pemise/traditional solutions.  The ones that are available, I'll find and post here.  This is a big topic - I plan to post some thoughts and links here in the next few days.  (I'm an evangelist for SaaS and chair of the Intelect SaaS Group, so you'll know where I'm coming from).  More later this week.

David Terrar

D2C and Twinfield

John Stokdyk's picture

Intellect UK guide - The business case for SaaS

John Stokdyk | | Permalink

As David has signalled, his Intellect UK committee has published a guide to the business for SaaS. I've done a short introduction for AccountingWEB, but you can also download the full, 24-page document for free from the Intellect UK website.

The main points raised cover:

  • Cost of Ownership (TCO)
  • No capital expenditure
  • Closer supplier relationship
  • Flexible access
  • Ease of use
  • Reduced need for IT support
  • Continuous improvement.

Each is worthy of a debate/thread in its own right, and there's a lot of valuable material that's presented in commendably straightforward language. There's a section on legal considerations, a small handful of short reference cases and a checklist of questions you should ask prospective SaaS suppliers.

So as not to replicate too much of the material, I edited the entire report down into just a few headings (and longer descriptions in the news summary). While doing so, I began to feel the Intellect document put forward some generalised assertions about ease of use, quality of support and closer customer relationships without evidence or examples to back them up. This is a classic flaw of technology marketing. The advantages are so self-evident to the converts that they don't understand why everyone else doesn't "get it".

From our conversation at the Business Cloud Summit meeting last week, it's very clear that there's still a lot of education to be done within the profession (NetSuite CEO Zach Nelson told me afterwards that even though he's been beating the Cloud drum for nearly a decade, he expects to continue doing so for many years to come).

If Richard Messik and other experienced Cloud accounting users come across this thread, it might be worth teaming up to create some detailed pieces about why and how firms and their clients made the migration to Cloud programs, and the benefits they have reapted. Accoutants will listen more readily to other accountants talking about their experiences than they will to software marketing people.
__________________________
John Stokdyk, Technology editor

david_terrar's picture

Generalized assertions?

david_terrar | | Permalink

Hi John,

Many thanks for doing your piece on the Business Case for SaaS.  I hope people read it and make up their own minds.  I have to say that I don't agree with you when you say:

"While doing so, I began to feel the Intellect document put forward some generalised assertions about ease of use, quality of support and closer customer relationships without evidence or examples to back them up. This is a classic flaw of technology marketing."

We're trying to put across a balanced view without the sales hype and without it becoming War and Peace.  When we started 20 pages felt like way too long a document, but by the time we had been at it for a while it took a great professional writer to condense down all of the contrbutions, so 24 pages (which includes a couple of pages of company intros at the back) seemed about right.  We're trying to compare and contrast the SaaS approach with the On Premise/traditional software approach and highlight how and why they are different.  Every assertion is backed up with a common sense argument and I'd be happy to debate them all with you individually.  The specific ones you pick out.... why don't you start a thread on each of those three and see what the discussion brings.  We cound have filled the document with case studies and testimonial statements, but it would have got too long, unwieldy and salesy.  We need more individual customer case studies told by the customers themselves - wer're going to do that on the Intellect website.  We need more research to back up some of the key areas - trying to figure out how we fund that and make that happen.   

And I agree there is still plenty of education to be done.  That's exactly why The Intellect SaaS Group, EuroCloud and the BASDA Cloud SIG have all been formed within the last 6 months.  The UK market is behind the US in terms of SaaS adoption... could be 18 months, could be a couple of years.  The topic is just "crossing the chasm" here and getting more attention.  Right now is when we need some of Richard Messik's clarity that you have quoted in several places around here, rather than vendor jargon and hype (note that Vantis with a quote from Richard is one of the case studies in our Business Case document).   

David Terrar

D2C and Twinfield

John Stokdyk's picture

The points at issue in more detail

John Stokdyk | | Permalink

Hi David,

My comments were intended to be constructive, but since they have drawn a pained response from your direction, I might as well cite my evidence, which covers the following comments (some of which are repeated in slightly more detail later in the guide):

Ease of use
"Most SaaS is designed to be intuitive, with online help and online support services. As such it tends to be very easy to learn to use."

No software developer sets out to make their program unintuitive, but you cannot blithely claim that an online product is inherently easier than an on-premise one (except, I will concede, when it comes to dealing with the underlying operating system).

Integration
"Easy data importing is important, but so is exporting and integration with other applications and services. Professional SaaS service providers ensure that these things are all taken care of through widely accepted web standards and procedures."

This is wishful thinking - "just let us take care of it". At our recent Business Cloud Summit, NetSuite CEO Zach Nelson acknowledged some of the difficulties of integration when he commented: "The web is good at loosely coupled applications. Businesses run on tightly coupled data. Connecting the apps has got easier, but the problem is the data model."

User sensitivity
"The SaaS vendor builds up knowledge about how customers really use the software as opposed to how they say they do. This enables the vendor to target enhancement programmes based on real usage, delivering rapid innovation, improved usability and greater productivity to all users of the software."

I'm not sure how Sage or any other desktop vendor would respond to that - but I know they put a lot of money and effort into support, usability research and customer-driven product development. Cloud vendors can certainly get the innovations into users' hands quicker, but you're pushing things a bit far to boast that SaaS developers are inherenetly more sensitive to user needs - it's what you do with the information, not the fact that you have it that produces improvements.

I definitely think it's worth digging into some specific threads to explore the various advantages and user experiences. I'll run the idea by Richard Messik (now the official manager of this group) to make sure it doesn't cut across any threads he's got planned.

Also think it's definitely worth investigating ways that we could develop educational content by tapping into AccountingWEB members' experiences.
__________________________
John Stokdyk, Technology editor

david_terrar's picture

It's my baby so I'm bound to defend it - 1st ease of use

david_terrar | | Permalink

Hi John,

Expect some more responses sometime later or tomorrow on the other two, but....  

On ease of use, one of the ironic things that the "fringe" meeting threw up was the 4 accountants talking about how they have to regularly train their clients on how to use the system and they still make mistakes, and then they have to train them again.  Sage 50 just isn't as easy or intutive as a more modern SaaS based application.  Now I haven't got a usablity lab study to prove it, but it's common sense.  If there were any on premise accounting applications that had been released in the last year, I'm sure they would be designed with ease of use in mind, but there has not been anything conventional new on the market.  All the new products are SaaS and Cloud based.  They are influenced by consumer websites like Amazon - I don't have to go on a training course to order my books, and when I do get in to trouble there are FAQs and videos and online help.  If you talk to any of SaaS accounting competitors they'll explain how this is part of the design psyche and how we have to provide "self service" support options.  So it's my contention that "in general" the current SaaS products are easier to use than the current on premise products.  Hopefully we'll hear from some real customers who will chime in.

More of this kind of thing later, and I hope it will be useful. 

David Terrar

D2C and Twinfield

dahowlett's picture

Here's the answers

dahowlett | | Permalink

@john - I'll answer your points:

UI - Once you have developed a UI it is very difficult to change when they're tightly integrated to the solution. That's the old way to build software and how on-prem is developed. It is very different in the new world because of the design architecture. Example - SAP built new UI for ByDesign in <90 days. It took years to get WebDynPro. SaaS vendors work directly with customers to solve UX issues as a large group. Very hard for on-prem people to do the same. We're also seeing more saas on mobile - try doing that without a SOA architecture. This is both choice and usability. They are going hand in hand.

Integration - Open Web APIs solve some of the problems. Check Kashflow's ecosystem. Check App Exchange. That's not wishful thinking but reality. Netsuite has specific issues that are not necessarily seen elsewhere. Data export is high on accountants' minds. It's been/being addressed. Any data going into an app will present an issue regardless of the application/service.

User sensibility - Sage can say what it wants but it can never compete with the vendor who is providing a multi-tenant service because they don't have visibility into what customers are using, how they're using it etc. That's why services like Mint are so important. They can see the patterns of usage very easily and provide additional services. This is one of the BIG wins for saas. It also explains wjhy Intuit acquired Mint.

Apologies for the technical answers but that's required on this occasion.

david_terrar's picture

On User Sensitivity

david_terrar | | Permalink

The key thing here is that the on premise provider is told how the software is proposed to be used in the sales cycle and the demos and in the initial implementation.  The SaaS provider LIVES with the way the product is actually used day in day out by their customers.  It's a huge difference in terms of collecting real usability data - living with it and being able to see and analyse what is happening on your own servers, compared to doing a survey to look for it.  When you couple that with the fact that the SaaS vendor makes regular software updates - on a weekly, monthly or quarterly basis (depending on how they are set up) rather than the usual 12-18 month release cycle of the traditional vendors, this adds up to a lot more really useful function flowing by customer demand in to the typical SaaS product.   Again, I hope we get some of the others or customers chiming in.

David Terrar

D2C and Twinfield  

david_terrar's picture

On Integration

david_terrar | | Permalink

I'm glad Dennis has chimed in too....

In practical terms, take a look at the various SaaS providers.  Most of us have a published API to our web service with guides on how to integrate.  Look at Twinfield, Kashflow, Xero, e-conomic, Salesforce, etc.  Almost by definition you are going to want to feed your accounting system from some operational application.  If you are Salesforce, then you are front ending all manner of systems.  We all have to make it easy, and the natural way is to have a proper (and here we get some techy developer jargon)  SOAP based web service using XML.  But ignore that... whatever technology the particular SaaS provider uses, then for the programmers that understand that stuff, the providers have sensible documentation, or a guide in a wiki somewhere.  I'm sure liquid Accounts, or Aqilla or Liberty and the other guys will say roughly the same thing. 

Now go and do a comparison of half a dozen on premise accounting products.  They'll have integration available in some fashion, but maybe it's via a CSV file, or maybe you tend to have to get the help of the author.  Then the documentation isn't as easy compared to the newer products.   On average, those products are more difficult to integrate to other systems than the more modern SaaS products.

With NetSuite it is a slightly different problem as their product covers a lot of the business process, a more complete ERP.  Whatever issues they might have, and based on my very limited knowledge of the product, I'm prepared to take a bet that integrating a given external app (say a particular UK payroll) to NetSuite will end up cheaper than doing the equivalent thing to Oracle Applications.   

David Terrar

D2C and Twinfield

various ...

JC | | Permalink

John

Would have to pick you up on '..Connecting the apps has got easier, but the problem is the data model...' and ask for this statement to be expanded; what do mean the data model is the problem??

Dennis
Suspect NetSuites issues revolve around their metadata structure; good way of allowing everything to be infinately customisable but raises problems further down the line when you need to 'open up' the system with WebServices etc; far more work to be done because everything is dynamic. Anyway most web-based apps don't use metatdata to the same extent as NetSuite so as you say they are a special case.

David
'..products are more difficult to integrate to other systems than the more modern SaaS products..'; not strictly the case. It is perfectly possible for desktop system to have underlying dll's which are exposed to their user base. Have done this many times and in fact the same dll can be available in Excel/Word etc. Probably the crux of the matter is that desktop solutions are lazy and simply cannot be bothered to go that extra mile to help their user base. Whereas web-based are new boys and need to make the extra effort to gain market share.

With either desktop or web-based (provided the data structure/object model is sound) it is simply not a big deal to expose parts of the system in api/web services
 

david_terrar's picture

"not strictly the case"

david_terrar | | Permalink

Hi JC,

You've got it spot on!  Part of the problem with our Business Case document is we are trying to talk about a whole collection of SaaS products generically to highlight the trends and differences to the typical on premise equivalent products.  You are absolutley right to say "not strictly the case" as an on premise developer could develop a really easy integratioin/API if they wanted to, but I agere with all you go on to say.  In general, the SaaS boys have integration as part of their mindset from the start, and the typical architecture of a web app makes it more easy to do. 

Just to amplify the point, CODA's a realy great product (used to work for them) with good integration tools.  However the follow on SaaS product that the guys have developed and has now become FinancialForce.com does it better. 

David Terrar

D2C and Twinfield

daveforbes's picture

online vs desktop

daveforbes | | Permalink

Web based software is new. It therefore has a modern interface. Software that has been around for many years with a large installed userbase is constrained to change its user interface incrementally. There has to be a balance between keeping existing users happy and attracting new ones. I don't think this is an inherent advantage of being online.

 

DuaneJAckson's picture

Not just architecture and desire for market share

DuaneJAckson | | Permalink

It's not just architecture or desire to win market share that leads to SaaS apps being more integrate-able.

It's the very fact that they're web-based APIs

So the "endpoint" for the API is in the same place all of the time for everyone, and accessible from every node of the internet.

Whereas with a desktop app you have to make sure the other app is installed, that it's the correct version,plus ensure the appropriate DLL's are registered, set DCOM permissions and so on.

With SaaS, it's just there and works.

daveforbes's picture

hmmm ...

daveforbes | | Permalink

Just because it has a web based UI does not mean everything is available in an API. For example we want to extract the TB from one particular cloud app to feed into our final accounts but this is not available in the API. (We can get and set customer address etc - but can't get the info we want !)With desktops apps we always have the option of extracting data directly from the underlying database / file.

 

dahowlett's picture

Well...sort of...

dahowlett | | Permalink

@david - the UX issue is massively important. On-prem vendors cannot do much more than tinker, services based software has the inherent ability to make deep changes. That's a huge deal because it is no longer about switching something on or off but about the ability to do many things we couldn't do before.

DuaneJAckson's picture

Always?

DuaneJAckson | | Permalink

@dave "With desktops apps we always have the option of extracting data directly from the underlying database / file."

Always? in my experience, there's as much variance with what's accessible on desktop apps as there is with web-based apps. You can't always go straight to the underlying data.

The difference with web-based apps is you can ask the vendor to add a new function and it can be there tomorrow (we've done that on many occasions). Whereas with the dekstop apps you might get it next year in the next release if you're lucky

daveforbes's picture

Online user interfaces

daveforbes | | Permalink

@dennis

If I have a UI based on html running in a desktop app and the same UI running in a browser connected to a server, whats the difference?.

If a cloud vendor decided to produce an "on premise install" that installed a webserver and their backend on your PC, still accessed through a web browser, how would the UI change ?

I am not disputing that some cloud apps havie nice UIs - I really like some of them, I am just saying that I don't think this anything inherent in cloud apps.

daveforbes's picture

always

daveforbes | | Permalink

@duane

Maybe I should have said "always have the option of attempting to ..."   but if you have SQLserver, access and dbase covered,

then that gives you the underlying data of a fair percentage of packages.

david_terrar's picture

Technically possible vs Practical reality

david_terrar | | Permalink

Hi Dave,

Technically, a desktop app can be created wth a lovely UI / UX.  We're talking about the practical reality of what is currently available, not anything theoretical. Compare some current cloud products like FreeAgent Central, or Kashflow or some of the others mentioned above with the typical on premise equivalent.  When you add that to the responsiveness of the current SaaS providers as Duane just mentioned, then you have some very practical advantages.  

You've also got to bear in mind that the average customer reading this is probably using Sage 50 or QuickBooks.  

David Terrar

D2C and Twinfield 

daveforbes's picture

pretty desktop apps only theoretical ?

daveforbes | | Permalink

Excel is quite nice - and its not cloud.

dahowlett's picture

Not just UI but UX

dahowlett | | Permalink

@john - I think you're missing the point. UI is part of the UX topic which is central to app design. SaaS vendors distinguish themselves by keeping close to the understanding of what works and what doesn't for the benefit of users. The design allows them to do things faster, cheaper, better than the on prem guys. That's a fundamental difference in approach and one that the architecture allows, by design.

You either believe that or not but I've seen it and discussed in and around countless saas players systems. It's a core differentiator. Good example - check the original SAP Business ByDesign UX. Terrible. We all hated it. It's been iterated several times in the last 2 years. Only yesterday, they showed me yet another iteration. Always improving and now fundamentally different from the first version. Now compare that to the SAP GUI which has been around for years. UX/UI is a central part of analyst presentations today. It's that important.

The main issue for me.

gsgordon | | Permalink

This thread seems to have lost the thread - where are the "solid figures" that John asked for.

Nevertheless, as a cloud accounting user, I have found the discussion to be interesting so I'll carry on.

So far, I am very impressed by the performance and usability of the cloud application we are using. The supplier also seems to have made integration with other applications a priority.

My main concern is loss of control and I suspect that's the case for many users.

For example, if cloud performance gets worse with time, there won't be much you can do about it. With on-premise software, there may be things you can do.

Data security is also a "loss of control" issue. I know user backups are often lacking, but at least it's your own fault if something goes wrong.

I know many (or all?) cloud suppliers claim to address those issues, but (in my opinion) they need to do more to positively convince us.

We also need more than just reassurance on Data Protection issues.

Perhaps a vote (restricted to 1 vote per issue per member) on a number of such issues would be useful, and of interest to suppliers.

dahowlett's picture

Numbers?

dahowlett | | Permalink

@gsgordon - asking for numbers is not really the right question which in part explains why it is not being answered. I'll give you ONE stat that should give you something to consider. Amazon can offer storage at 15 cents/GB/month. Tier 1 data storage is going to be $1-3/GB/month. That's a x8-20 cost advantage before you factor the additional costs you need to spend in managing the envronment. SaaS vendors can readily use Amazon to keep their costs down. You keep the benefits because it's all wrapped up in their offering.

To your point about control etc and what you've said here: 'I know many (or all?) cloud suppliers claim to address those issues, but (in my opinion) they need to do more to positively convince us.'

Does this work?  There isn't a SINGLE recorded catastrophic loss of business critical data from any saas vendor. Period.

How about this: ADP has been providing payroll services globally for 40 years. They're still growing. I'd argue that payroll is business critical but I've never heard a customer talk about data loss or downtime. If I had I'd suspect ADP would be out of business in very short order.

How about this: Google goes down (ie is not available 3 times a year) and it's a big deal in the media. Did you know that most business systems are down at least once a week and often for a day or two at a time? Why? Because we hold saas vendors to a higher standard than the on-premise vendors.

Or try this: if you can't enter something to the general ledger for an hour/2 hours/a half day then to what extent would that impact your business? I'll guarantee it won't make an iota of difference, even in an FMCG business - unless you're Waal-Mart. Contrast. If you can't get bills out the door because they're tied to your on prem system that's not working then how long before you're dead in the water? We've seen case after case of this happening. But not in SaaS systems.

Final data point: You'd think that security issues would be top of mind in larger businesses. It is and is a much more complex problem than applies to SMBs. Even so, I never hear this type of discussion. Because the risks are understood to be much less than existing environments and in any event, other benefits far outweigh the risks.

david_terrar's picture

Pretty?

david_terrar | | Permalink

Hi Dave,

I realize accountants love Excel.... even the UI.  But take a look at the Cloud equivalents like Zoho, or Google Docs, or EditGrid or SocialCalc and they have just as nice a UI / UX as Excel, but they don't quite have the same functionality yet.  Perfectly good for some  applications, which is why major companies like Jaguar Land Rover have made the switch from Microsoft to Google Apps.  However, the emphasis is on the "yet".  Wait to see what Microsoft does with Azure and the next release, and what happens elsewhere in the collaboratiuon space. 

David Terrar

D2C

Numbers - why not?

gsgordon | | Permalink

I appreciate the examples given, which certainly support the case for SaaS. However, my point about "loss of control" was about perception rather than facts. I don't suppose a high proportion of potential converts to SaaS are reading this thread, so the vendors really have to make a clearer case.

On the question of numbers, I'm sure that a comparative analysis of the costs, security and performance issues would go a long way to convince the sceptics and less well-informed users.

dahowlett's picture

Data points and perception

dahowlett | | Permalink

@gsgordon. Perception. OK - so the questions being raised here are exactly the same as those in some of the 2006 threaded links in John's swipe at me. We have more data points, better stories, answering the same questions and still people don't want to know? In which case I'm done. I'd rather put my time into helping those that do get the story than banging my head against a brick wall of indifference. One thing's for sure - it confirms what I've known for 30+ years. Buyers are rarely logical.

However, I agree there are better ways to communicate.

david_terrar's picture

Vendors need to do more, but that's happening

david_terrar | | Permalink

Hi gsgordon,

Unlike Dennis I'm not done, and I hope I can talk him in to carrying on spreading more of the word here.  However, I understand precisely what you mean about perception and it's an important barrier we SaaS providers have to overcome.  You say:

"For example, if cloud performance gets worse with time, there won't be much you can do about it. With on-premise software, there may be things you can do."

As Dennis said, look at the record of the providers.  The accounting provider I represent for the UK has been running their solution since 5 October 2000 (and Salesforce has been going even longer).  No significant outages in all that time, and they've kept capacity in line to give good performance over a 9 year period, which is one of the reasons their customers include the likes of Deloitte and BDO.  It's actually much easier for a Cloud provider to add capacity because of the "elastic" nature of cloud based resources.  It means I can rent additional servers to cover a peak (with a credit card transaction) and have that capacity available within seconds, compared to ordering hardware and waiting for it to be delivered and commissioned.  With on premise, you are in charge of monitoring that capacity problem and the asociated IT management headache.  With a SaaS provider, they are contractually bound to provide you with a service level.  And by the way, it's more cost effective as well as less hassle.  With on premise I need to buy and install the capacity for the peaks.  With the Cloud I can ramp the capacity up and down.  I was listening to a Cambridge examination board telling me about the money they've saved moving their systems to the Cloud over exactly that issue (they have two big peeks each year, but don't need anywhere near that capacity in between).

But the SaaS providers need to do more.  We need more hard evidence and case studies and better "jargon reduced" messaging.  It is no coincidence that in the last 9 months the Intellect SaaS Group, the BASDA Cloud SIG and EuroCloud UK have all been formed.  We're all trying to adress that market demand for providing better information on SaaS and the Cloud to business.  The chairs of all 3 groups (I'm one of them) met with Richard Anning of the ICAEW's IT Faculty yesterday (an excellent initiative suggested by Dennis Howlett).  The aim is for the 3 vendor groups to work together with the Institute representing their membership (the buy side) to address topics like security (on various levels) to see if we can agree best practice or some form of quality mark.  We have a number of initiatives that we are now working on, as well as making sure each group uses their resources on complimentary things.  Expect some specific announcments soon.

In the meantime, I'll be looking to post some more hard evidence and numbers/comparisons here.  

David Terrar

Chair Intellect SaaS Group

D2C and Twinfield 

Covering the same ground ...

JC | | Permalink

David

I'm with Dennis on this one (unusual!). What is the point in banging ones head against a brick wall.

All the evidence supports a very strong case in favour of web-based and yet there is a constant flow of '.. please convince me..' eminating from the profession. Whilst a naturally conservative body life is really too short to keep covering the same ground over & over again; and to what purpose? Is it really going to have an impact on those who don't really want to know (horses to water etc..) except take up a huge amount of time and effort by the champions of the web-based cause, trying to convince them, when time could be better spent on a receptive audience.

Quite frankly the profession will have to adopt web-based in the future if their clients go down this route; which seems highly probable. So it seems eminently sensible for them to be leaders in the field rather than laggards driven my the more elightened clients; being taught by their clients really will not do their credibility any good at all

Perhaps not diplomatic and definately not PC but as once said by a partner in one of the big 5(4) - clients will expect their accountant to adopt the old accounting maxim of FIFO (fall in or ** off)

daveforbes's picture

I would love to be convinced

daveforbes | | Permalink

All the evidence supports a very strong case in favour of web-based.

Unfortunately I have not seen this evidence yet. I am not a luddite, I synchronize my data to an amazon S3 bucket, but I still run my apps on windows. I might occasionally use google docs, but only when caught without my laptop.

I am sorry if this has been covered elsewhere - could you point me at the thread and I will read up on it there.

I would love to be convinced too. Why would anyone with any sen

chatman | | Permalink

But there is no thread providing all the evidence. All the SaaS crowd do is say "it is cheaper" "it is better" "users will drive you to it", without providing any support for these conclusions.

dahowlett's picture

Bingo!

dahowlett | | Permalink

@chatman - I've used this thread to provide ample examples, evidence, links and pointers. Try this Google Search: cloud economics

BTW - one of the reasons you'll struggle to find all that you need here is that AW can't be Googled and unless search has been completely revisited, it was a nightmare.

daveforbes's picture

googling

daveforbes | | Permalink

You can google accounting web e.g. - http://www.google.co.uk/search?hl=en&source=hp&q=cloud+site%3Awww.accountingweb.co.uk&meta=&rlz=1W1ADBF_en-GB&aq=f&oq=

 Thanks for the links dennis. The first one I hit on was http://community.citrix.com/display/ocb/2008/09/27/Cloud+Economics+101+-+Part+1

Which unless I am missing something, makes the cloud option look expensive ( $35K vs $7K per annum) for their example scenario.

Choosing not to read ....

JC | | Permalink

David

But the comments need to be read as well!

  • no account taken of staff costs, software etc ....
  • scaleable capacity not sensibly available on premises at a reasonable cost. So one has to have redundant over capacity to accomodate any data peaks

and this probably highlights the communication bottleneck. No amount of argument in favour of web-based is going to convince people if they are selective in their reading/understanding

I suppose some may spend 24 hours a day in the cloud!

gsgordon | | Permalink

... and the $35K is for 24/7 access to the Amazon EC2.

Their more reasonable case of a 40 hour week comes in at $8.3K, not much more than the $7.8K (not $7K!) for on-premise hardware - ignoring software and possible real-estate costs.

The responses on the Citrix thread are well worth reading.

daveforbes's picture

Not exactly "selective reading"

daveforbes | | Permalink

I started at the top and read the first page, read until the figures stopped and it started getting all wordy !

I agree it did not include staff costs, electricity, insurance, backup costs, office costs etc 

I assumed, possibly incorrectly that the implication was that the marginal costs on these items were deemed to be neglible.

Choosing not to read

chatman | | Permalink

Just because someone does not accept that claims have been justified by evidence/arguments, it does not mean they have not read the posts. It may be that you have actually failed to support your claims. It might be a good idea to recognise this possibility.

As I asked before: why would anyone (apart from someone selling desktop apps) choose not to inform themselves about something that could make their life easier?

daveforbes's picture

funny thing

daveforbes | | Permalink

My involment in this thread started with me saying that I thought chatmans estimate for on premises software was too low !

For the particular scenario on that dennis' link took me to, it appears cloud is more expensive.

There may be many other scenarios where it is much cheaper.  I agree that scenario is not typical - most accountants aren't going to be installing and running 5 servers with 24/7 availability ! There must be some other case studies for more typical set ups.

david_terrar's picture

Some TCO comparisons and other documents - your starter for 10

david_terrar | | Permalink

Hi chatman and daveforbes,

I realize we have some serious scepticism here.  As Dennis said, we've presented quite a number of stories and common sense arguments in this thread - it's just a shame it isn't being read by more people.  So here goes.

Here is a good 2006 document from RayWang/Forrester on the topic: Comparing The ROI Of SaaS Versus On-Premise Using Forrester’s TEI™ Approach

Here's another circa 2006 document from SIIA: Software-as-a-Service;A Comprehensive Look at the Total Cost of Ownership of Software Applications

Here's some more recent independent research by Nucleus on nGenera's use of Intacct: nGenera Achieves 589% ROI with Intacct Financial Management Software

Here's an independent report from Hurwitz & Associates: The TCO of Cloud Computing in the SMB and Mid-Market Enterprises: A total cost of ownership comparison of cloud and on-premise business applications (this one's a NetSuite thing, so you'll need to enter your details to get it)

This one is a PPT of a specific comparison for a health management company: TCO Comparison Analysis Salesforce.com vs. Siebel On-premise vs. Siebel On-demand

This is THINKstrategies: CIO Guide to SaaS

On a lighter note to finish off, here's Ramon Chen imagining software as Toasters!: Enterprise Software Toaster vs. SaaS/Cloud Toaster

As I've said a few times in this discussion, we need more good case studies and hard evidence, and that's the sort of thing that will be coming from the SaaS and Cloud vendor groups in the coming months.  It's notable that some people are pouncng on the one negative, and ignoring the 10 postives.  People coming in to this topic need to do some research, listen to all the viewpoints and make up their own minds.  I'll do some more digging in the next few days.  This thread could turn in to quite some resource laying out all of the arguments, and pulling together some of the available documents and studies.  We'll see what the nay sayers say next.....

David Terrar

D2C and Twinfield 

 

jimwmackenzie's picture

All we have to fear, is fear of getting caught

jimwmackenzie | | Permalink

As I asked before: why would anyone (apart from someone selling desktop apps) choose not to inform themselves about something that could make their life easier?

At the risk of being a touch glib, one of the main reasons for non-adoption of new technologies is fear, whether real or imagined.  The classic example is the perceived loss of control over the data by using web applications.  As pretty much any Cloud vendor worth their salt will tell you, the steps they take to protect their customers' data are far more vigorous than those typically found in web applications.

To elaborate further, I'm not aware of many small businesses that have their servers in seismically neutral locations, with 24/7 physical security, advanced fire prevention systems, plus regular backups, redundant servers ready to hotswap in in the event of hardware failure, with a current backup of data et cetera; the cost is prohibitive.  With web applications, this is economically viable, even for the chap on the Clapham Omnibus, through the economies of scale you can leverage with web architecture.

This particular example is a common one; despite the massively more in-depth protection available (in the main) to web apps, the perception that needs to be broken is the loss of control "I can't see my server, so I don't control it".  Looking at it objectively, the benefits are clear, but just about every great business looks at things subjectively.  That's why cloud vendors need to do a better job of highlighting the real issue - how can I get at my data if I want to?  The rest is just details; nice to know, but don't address the core fear of "if this goes wrong, how would we recover?"

A similar approach is needed for the other questions potential customers will ask; I look forward to seeing if the Intellect group can do so.

Pages

Add comment
Log in or register to post comments
Group: Cloud Computing for Accountants discussion group
A place for accountants to share their thoughts about web-based systems