Brought to you by

Best of Breed v. All In One

Posted 9 years ago in Tech by
Posted by

I’ve been on the road for the last couple of weeks. Brisbane, Sydney and Melbourne last week and Silicon Valley this week. It’s been fun and very stimulating meeting with lots of partners and having a good look at the state of the small business web space in different countries. It’s clear that it’s still very early days but the small business web is becoming more interesting as new services and vendors emerge and connect.

We’re starting to notice some implications and new market dynamics as the industry is maturing.

When we first thought about doing Xero, we’d have assumed that three years in, we’d be doing all sorts of value added services over the accounting platform. Even though we’ve shipped more than 45 major updates, we’re now very aware that accounting is a broad minimum feature set – there’s always more to do.  So we’ve stayed focussed on that.

As adoption of online services for small business has gained traction we’ve seen a number of new services arriving that link to Xero. Because these companies connect to Xero they haven’t had to write an accounting engine. They’ve been able to focus on improving specific business processes, often in a very elegant and high value fashion.

The incumbent desktop products that have been around for many years have added many features during their time in the market and therefore often provide more than just accounting. For example job costing and payroll features are common in many established desktop accounting products.

One of the reasons these broad but thin desktop applications evolved is because small businesses do not integrate with desktop software products. It’s too expensive to have a consultant come and do it. A good example of this is Customer Relation Management (CRM). You would think that most small businesses should have a CRM system to track leads and customers, but if they have CRM and an accounting system, as soon as they create an invoice you have customer data in two places. So integrated products have been the only real option, even if they provide a pretty average user experience.

The small business web is taking things in another direction. Using web services, the web vendors themselves are integrating their solutions to work well together. This means that small business owners can get ‘best of breed’ solutions that ‘just work’ and are best in class.

Xero and partner solutions

If you compare Xero to desktop products there are things we don’t do, things we do much better and new services that are unique, such as live daily bank feeds. But when linked to another specific solution, like Job Costing, Sales Order Management or Payroll, the combined solutions are much better than the previous all in one approach. We’re impressed by the new solutions we’re seeing coming through which are giving small businesses very sophisticated integrated ‘best of breed’ business systems. And these are at a very reasonable price point. They also provide real choice where small businesses can ‘pick a chocolate out of each box’ to find a collection of applications to best meet their requirements.

This has further implications. Currently there is a real gap in the small to mid market accounting software space. Once businesses grow to a few million dollars in turnover the cost jump from commodity desktop accounting software to low end ERP software is high. We’ve heard of many customers who have hung onto desktop software, even unsupported software, because the new versions or next product step-up is tens of thousands of dollars.

Xero diagram

The combined, linked, ‘best of breed’ solutions model will provide much better options for businesses as they scale up. We’re already seeing a number of ‘cloud consultants’ appearing that want to help business owners set up these linked solutions. While it’s still early days we believe we’ll see these hybrid solutions delivering value  for just a few thousand dollars per year, in contrast to the current options which cost 10’s of thousands per year. This will be very disruptive to the low end ERP space and great for growing business owners.


For accountants

Keep an eye on your vendor partner blogs to find new partners and integrations. Normally they produce short videos which give you a feel for what they do. Think about who among your customers might be interested in each new solution. They’ll be glad to hear from you.

For cloud consultants

Becoming familiar with new solutions is an exciting opportunity. Perhaps working with your local accounting firms will lead to introductions through their client base.

For software vendors

Working together with partners you can gain access to high affinity customer groups. Learning to jointly market with partners will be key.

For entrepreneurs

Seeing these fundamental changes occurring in small business technology, opens up new opportunities for vertical and specialist applications, even applications that connect other applications together.

The challenge for us in the small business web is to communicate the benefits of ‘best of breed’ to customers who have been trained over the past 10 years to think that the integrated ‘all-in-one’ approach is best. But we think it’s compelling and hope our partners and other vendors begin to educate the huge small business market that they really can have it all.

What do you think about this ‘best of breed’ model?


Ben Kepes
August 4, 2010 at 12.01 pm

Rod, great post, and it mirrors some stuff that I’ve been seeing. I don’t wonder if there isn’t another approach though, something I’ve called “quasi-suite” where a vendor cherry picks best of breed third party offerings and embed them deeply within their own app. BrightPearl doing some interesting stuff in this space…

Rod Drury
August 4, 2010 at 12.17 pm

I think that’s a valid approach to try in some circumstances but might be a support nightmare. For example if we embedded a 3rd party job costing system our customer care team might get calls regarding something they know nothing about and devalue our support experience.

Thinking about this more, tab browsing is a further enabler of this best of breed approach. It isn’t a big deal to have multiple apps open at once.

An example of where a ‘quasi-suite’ model might be good is when the app is explicitly a collection of widgets. Say for Internet Banking, where sub components might be subscribed to by a consumer and those widgets might come from different vendors.

Ben Kepes
August 4, 2010 at 1.27 pm

Agreed that support could become an issue. But this is where targeting distinct functionalities and embedding them such that they appear to be part of the overall app is a good approach. In BrightPearl, for example, the email campaign component (MailChimp embedded in the app) is supported as if it where an integral part of Brightpearl (which in effect it is – hence quasi-suite)

Nik Wakelin
August 4, 2010 at 1.58 pm

We’ve been able to see first-hand the benefits that integrating with Xero brings – we can build a simple, “best of breed” Time Tracking application and have it backed by a powerful, fully-featured accounting solution.

The really hard part is getting the integration seamless, and positioning it correctly. Is it better to simply provide an API, or to embed third-party applications directly into yours? Do you provide billing, a la App Store, or do you leave that to your integration partners?

Xero have done a great job of this so far, and I look forward to some of their upcoming stuff to make integration even easier for our customers.

Best part: It’s something that is nigh-on impossible for a desktop vendor to copy.

Adrienne Pierce
August 4, 2010 at 5.26 pm

Best of Breed is certainly the way to go although it’s getting pretty confusing in the market as to WHO actually is the best? We can cutomise a solution for our clients by mixing and matching. The SME owner is becoming more open to solutions that save them real time and real money as well as providing ‘understandable’ information that is applicable to them and where they and their business is at – Xero provides an excellent foundation on which to build the rest of the infrastructure required in todays business environment for the small business.

Dennis Howlett
August 4, 2010 at 5.34 pm

@rod – interesting to see your take on this. My 2¢:

Suites almost always beat BIC…that ‘may’ change with cloud but the closest I’ve seen to this is’s platform, NetSuite’s tools and SAP Business ByDesign SDK. In all cases, they are positioned as a platform and not just as a functional solution. To me that looks like a suite play.

At the VSB level you could almost argue that FreshBooks and KashFlow have done the same thing albeit they are simply exposing APIs. The big issue is where the risk and controls lay.

Nik’s question is on point. Most people want a seamless integration which is where SFdC scores really well. New stuff is just another tab.

One problem that isn’t discussed: if you end up with a bunch of devs doing a bunch of the same stuff then gr8 for initial customer choice but some of those are not going to survive. Who takes responsibility? I suspect it is the provider to which they are integrating because customers will say: “Hang on, why didn’t you do due diligence before allowing them to touch your systems?”

Adrian Pearson
August 4, 2010 at 8.38 pm

This discussion has been going on for years in the desktop world, in vertical market niches like accountancy firms for instance. The big players sell the benefits of an “integrated suite”, the smaller competitors suggest that they are cheaper and more nimble and that “Jack of all trades is master of none”.

Indeed, I can think back to older times (when IT was much, much less reliable generally) and the argument was actually about the integration of software and hardware. Back then, I used to recommend that clients get both from the same vendor, to avoid the buck passing that would otherwise inevitably ensue – “it’s not our software, it the server ..”

So, I think the discussion here is effectively the same but in the modern world we are now talking about SaaS integrations. Ultimately, it becomes a matter of who puts their hand up if things go wrong? Now, if both parties make it clear that they will work together to resolve any problems with their mutual customer that would make a massive difference.

Oh, and SaaS vendors need to get less precious about what they expose via their APIs – let potential integrators have access to everything.

Dennis Howlett
August 4, 2010 at 8.51 pm

@Adrian – don’t go there. One word: security.

Ed Molyneux
August 4, 2010 at 9.15 pm

I think for the VSB end of the market (by which I mean up to ~10 employees) it’s possible and highly desirable to integrate more tightly than typical APIs allow. Especially on the accounting side of things, there’s bound to be some piece of information that needs to flow all the way through if you’re going to avoid manual adjustments later on – you don’t want to have to rely on somebody else’s API to support this.

I write a bit more about my take on this at

Chris Tanner
August 4, 2010 at 9.28 pm

Good post Rod and a point very close to our hearts too! We have deliberated long and hard about which parts of the Brightpearl system are built-in and maintained by our development team vs those that we ship out.

Attempting to build, maintain and support and end-to-end solution is a big ask of any young company, but time and time again we hear people breathe a sigh of relief when they find themselves able to run everything from one login.

“new stuff is just another tab” says Dennis. This is really powerful. Small businesses are one set of people, they need one system. Big businesses are separate teams – they can deal with multiple systems (but ideally they shouldn’t have to either). Just around the corner for Brightpearl are more “managed plugins” – where we build the code to integrate with 3rd party APIs like SagePay, ConstantContact, FedEx, UPS, Shopify … the list goes on. Modern technologies like REST APIs, openID and oAuth are a great help.

With a “quasi-suite”, The user benefits from a well-built and maintained interface to their other apps without having to manage their own inter-connecting system, and the vendor benefits by reducing the amount of features they have to build from the ground up. User gets best of breed all over the show. Bonza.

chris Jangelov
August 4, 2010 at 9.41 pm

A few thoughts:

Our industry is young. Very young. It reminds me of the time when media in our homes (music, radio and TV) was delivered through one large piece of furniture that held it all, The inner workings were controlled by the manufacturer and if one piece broke the whole system needed repair.
Then came suites of amplifiers, tuners, turntables, loudspeakers etc and that problem was solved. It was also possible to buy the best of breed, or stick with one brand (because the parts played well together). To make that possible everything needed to be connectable, i.e. standards emerged.
So that’s my first point: Standards – and not proprietary APIs.

Second: First comes sales, then you get payed and not until then do you need accounting. I seriously believe that accounting software aims at a limited market and that we are wide open to new competitions from, say SalesForce and the likes. They are first buy for many businesses and sell more licenses per customer. They will have the money to compete. I don’t think our approach is future proof.

Third: Accounting software is a first generation blueprint on early 20th century manual routines. Lots of people writing and checking numbers. If anything was wrong – you go to prison. Business driven by terror. Today numbers are handled by computers. Once entered they should never have to be entered again. The Xero bank connection (and others similar) is the first modern thinking I’ve seen in years. Let’s take that thought a bit further: Who knows all about tax rules? The Inland Revenue. So it is only if they do a poor job that it is possible for a software vendor to make money on tax software. To me that is misuse of taxpayers money and will have to change. It will take some time but you can clearly see how this is forcing Intuit to more lobbying than arguing already today.,0,6498588.story
So numbers should float freely between the business, the bank, Inland Revenue and many others in the future. We can’t sell on fear anymore but we can supply overview, analysis and up to date control of your business.

Four: Earlier customers bought software packages. Today the buy interfaces and support. They need a certain level of functionality but the final choice is made from usability, ease of learning and service level.
We could do like the car industry. The platform represents the code and the brand (the “car”) represents the interface. When software is delivered through the browser – who cares about the part of the code that is run om the servers?
This is also a strong argument for standards.

So our choice would be: To willingly cooperate and compete on best of breed interface, service and platform code functions, or to die with the rest of the All in One industry. (I am talking from a pure SMB perspective.)

Charles Verrier
August 4, 2010 at 9.43 pm

“Build it, and they will come”

Surely, once your core product reaches a critical mass of users (and you have a good API on offer) then the ‘ecosystem’ of niche bolt-ons arrives as a matter of course as people spot a potential market. Xero seems to be doing well enough that it won’t need to worry about IF this stuff will happen, only about how to manage it when it does.

Microsoft do very well out of that approach with SharePoint without having to worry about managing an approval process or support issues for each add-on, and iManage Interwoven almost EXPECTS you to purchase add-on 3rd-party bits and pieces.

The trick, in both those cases, is that they are quite careful about not recommending the add-ons, They put effort into ‘Partner’ schemes and ‘Community’ building, but those partners do all the marketing and promises themselves.

Perhaps the obvious other example is Microsoft Dynamics – you can’t move for add-ons over there…

@Adrian – API issues a sore point at the moment, by any chance? 😉

@Dennis – Obviously an accounts system could be seen as needing more protection than, say, a SharePoint library (depends on what’s in it though). But after all, the whole point of an API is to give 3rd-party devs a ‘safe’ way of doing stuff (so, risky operations can be blocked or require extra authentication). FreeAgent (sorry for bringing that up on a Xero blog, but I know it better than I do Xero!) requires that the user specifically activates the API in their account settings page, which I imagine helps no end with security worries about the API being used as a hacking tool.

Stuart McLeod
August 5, 2010 at 2.18 am

Best of breed or all in one? In my mind, and my thoughts on this have only solidified over the last 18-24 months, there is no argument. There are 3 reasons why this is the case.

1. The consumer has choice. In nearly every category of software you can choose between an integrated set of products from multiple vendors or an all in one solution from one vendor.

Medium to large businesses might choose salesforce for crm and timberline for manufacturing or SAP to do all of the work. Small business might choose Xero and Paycycle for their GL and payroll respectively or saasu to do everything.

Both are fine options, and your decision rests entirely on the needs of the business in question. No one (and certainly not us) will criticize that choice.

The issues arise when your do-everything software becomes a compromise in one particular area. What if the CRM component is not powerful enough to drive the sales results that you thought it could? What if the point of sale component no longer meets your expanding needs? Do you change just that component? You now need some level of custom integration. Which brings me to the second point.

2. Who’s better at doing integration? The customer or the vendor? Or more pertinent perhaps is who can afford to conclude a comprehensive integration project? And maintain that integration through various points of future upgrade. It is in the interests of best in class vendors to provide that comprehensive integration.

Even in this day and age I hear ridiculous arguments like “but we’ll need 2 logins.” Please. I bet your home network has more than that. Irrespective, as integration becomes deeper and more sophisticated, the rudimentary friction points will be overcome by vendors who are committed to providing a first class offering.

3. Best in class vendors are hungrier than the all in one solution. Back in 1999, when I worked in sales for Oracle, we always knew that if we lost to SAP on the ERP sale we’d still win the database. That’s because our database was the only one in town.

So the development effort, at that stage, didn’t really focus on the parts we couldn’t win. Times have changed, but the reality is, I know that we can offer the greatest payroll interface in Australia. Because that is our job. And we won’t rest until that becomes a reality. We don’t need to worry about GL or CRM or anything and everything else. Just payroll. Boring as batshit to most people, but something that we are deeply passionate about. I know Xero feel the same way about their GL. You can feel it.

Do you see that level of passion in the MYOB UI or the UI of other large vendors?

Ben Kepes
August 5, 2010 at 7.25 am

APIs are totally different from the quasi-suite approach. Both Ed and Chris get this and, to a greater or lesser extent, are bringing this approach into their products. APIs are great but still rely on some degree of manual intervention by the user, and also force them to become accustomed to different UIs.

In the quasi-suite approach, the integration is at ground level and automatic, and there is no discordant UI to get used to as the third party functionality is embedded directly within the application.

Yes choice is good, but for the mass market VSB guys, if that choice results in them being unable to actually achieve their desired result, then maybe a little less choice, but a much better experience, is preferable.

Mike Deam
August 10, 2010 at 10.14 am

I’m kind of with Stuart on this one.

Take payroll. When setting up integrated payroll it’s a nightmare. The first question asked by the client is – So Mary can have access to payroll but not Mick and Pat. The consequences of this are that you have to stop Mick and Pat getting at the data through the back door as well as the front door. So you have to restrict the employee cards, access from the Bank (so Mick and Pat can’t do the bank rec), watch how super is paid etc. It is difficult to shut all the doors in an integrated world.

There are so many issues associated with payroll and so many rule changes its hard to keep up. You just need to to consider payroll tax, superannuation, awards, personal leave. What is a contractor versus an employee, workcover and I could go on. Its a full time job on its own.

So after a number of years I’ve quickly come to the conclusion that Payroll is a specialist area and deserves its own dedicated team to support it and its own application to run it. Which is why I’m also a member of The Association of Payroll Specialists. I find it’s the only way I can keep up.

As such best of breed is often a moving target. Recently I’ve switched my favorite payroll vendor from one supplier to another. Because Xero has not committed to one vendor migration of my client base was easy.

What is really interesting is that I work quite closely with another Xero partner. Sure we compete for business but we must talk 2-3 times a week sharing what we’ve found works, what’s not working, what can be integrated here. We often go in the same direction and often see things differently. Payroll is one that we differ on.

My point what maybe best of breed for you – may not be best of breed for all! One of the great things about Xero is it allows you to build your best of breed system. I don’t want Xero to pick what it thinks is best of breed because somewhere it won’t get it right for me or my client.

The ability of having two applications General Ledger and Reporting (Xero) and Payroll (Paycycle, WebPayroll, ePayroll & ADP) each focusing on what they do best. It means I get two development teams. I get enhancements more quickly. If Xero integrated Payroll how much of their focus would be split to payroll rather than other things. If there was only one vendor would the price rise? Would the vendor slack off and not provide new features for a year?

All MYOB have delivered for the last two years are basically the tax tables. Makes the price of MYOB Cover very high. I’ve even recommended to my MYOB Clients to consider an alternative to MYOB Cover for payroll based on the price and function provided.

A payrun (wages, super, tax, leave entitlements) is not the same as a purchase order for Phone expenses split into Mobile, Fixed line and Internet or the sale of your product. It has privacy considerations, calculation considerations, entitlement management, timing issues on when super is to be paid, what super is collected and when that is to be paid, employment basis (full time versus casual), allowances, time and attendance recording, compliance reporting and the surface is not even scratched. A payslip is not the same as a remittance slip.

In terms of User Interface is the Asset Manager the same as Accounts Receivable? Yes in some cases but No in lots of others, which is why full access is restricted to the Financial Adviser role. Payroll is similar, part is day to day operation (the payrun) and part is configuration (Super does not apply to overtime) and all of it is a privacy issue.

No the API’s allow Xero to build you the best of breed. There are some costs to the user of this approach (User Interface, training, managing multiple suppliers etc) but they outweigh the benefits (dedicated support, multiple development, price competition etc). So somethings should not be integrated and Payroll for me would head that list.

Jonathan Phillips
August 12, 2010 at 8.41 pm

The business rules have never changed. Focus, Focus, Focus…

Many brilliant smaller organisations have become a victim of their own success and expanded away from their core competence and within a small space of time become a poor relation to a more corporate behemoth competitor.

The art is staying niche and dominating your market – and at the same time making clever technology that integrates with other organisation’s best of breed systems.

Laid off Dad » Blog Archive » Best of Breed v. All In One « Online accounting software news from Xero
August 13, 2010 at 4.02 am

[…] Best of Breed v. All In One « Online accounting software news from Xero […]

Jason M Blumer
August 14, 2010 at 4.05 am


I agree that the cloud is opening up huge opportunities for accountants and cloud consultants. Truly, those in the cloud can focus on their “best of breed” offerings and let other apps focus on their specialty.

The greatest opportunity I see is in the accountant’s space. Our clients (mainly, those in the US) don’t understand all of the cloud offerings available to them just now. So our firm has made a commitment to study the solutions and put people in the model they need. We call it CoreCloud Services, and we typically review 10 basic potential business processes that could be moved to the cloud.

And our clients love it (often depending on their age – ha ha).

But, as I teach in an online webinar on Paperless and Efficiency in the Business Model, integration is the magic (using APIs) to make all of this fun. I can find integration among most, but who will keep building out that integration here in the US?

Someone asked if that was the responsibility of the vendor or customer. Customers can’t do it. They don’t have time and they don’t care (like the rest of the geeks commenting on this post). We care – our clients just want solutions. But I’m bumping into Integration problems here in the US.

I know more integration is coming (among banks, payroll, CRM, etc.) but we sure do need it NOW. Thanks for all you guys are doing! Excited about the future.

Jason M. Blumer, CPA

Building Best of Breed. « Thoughts from a Bookkeeping Consultant
August 17, 2010 at 11.39 am

[…] There is an interesting thread in the Xero Blog about All in One Applications versus Best of Breed. […]

Jenny Roberts
September 13, 2010 at 12.42 pm

This ‘simmer’ is just a couple of degrees from boiling.

Most businesses I have been chatting to are now saying;

“what’s the alternative now? MYOB? I don’t think so! – to be best of breed you cannot offer everything to everyone – we need experts in these fields who have passion & know their product – you get credibility with that!”


Brett Beaubouef
May 2, 2011 at 8.44 am

We have all heard the proverb “A chain is only as strong as its weakest link.” Applying this concept to business software, we would conclude that a business solution is only as strong as its weakest integration. Usually overlooked and underestimated, integration is one of the most important factors to consider as part of a best of breed vs. integrated ERP solution. The benefit of richer functionality is limited by partial integration. In the next sections, we will discuss all the factors to consider as part of making an informed decision regarding best of breed vs. integrated ERP.

Leave a reply

Your email address will not be published. Required fields are marked *