Salesforce and Demandware laid bare

Salesforce bought Demandware this week. It's about time Salesforce got serious about having a proper eCommerce platform, and also show come commitment to B2C. Of course web/mobile is a major channel (digital digital digital), and my thinking is that this acquisition filled a gaping void in their offering.

Who is the most Digital? "I am!"

Up to now you have a few ISVs and customers who have built eCommerce solutions on the much maligned Salesforce 'Sites' technology - the native on-platform quicksand. I guess this is ok for small-scale use but it can result in customer/prospect feel like they are caught in a blustery headwind. The Salesforce platform will punch you in the face if you throw a spike in demand at Sites, so given you are willing to accept the sub-optimal performance it deals you, you need to make sure the context is one of small/moderate demand. Today there is a lot of focus on, and associated tools for optimising the checkout experience to avoid dropouts, but you will only get so far with Sites.

You also have had the option of custom building something on Heroku and leveraging the most expensive sync job in existence to port your data into core Salesforce (Heroku Connect, would you really go near it?). Or custom integrating your eCommerce platform with core Salesforce which isn't a bad option if you are careful, many ways to do that in realtime or batch, but you need to be careful of API limits.

Now though, Salesforce will, down the line, offer Demandware as their 'Commerce Cloud', and increasingly it will be brought closer to the core Salesforce platform. Expect them to use the term 'digital' a whole lot more.

Indeed, with this acquisition Salesforce will be talked about a whole lot more in the 'digital transformation' circles. The Salesforce core platform is essentially a very flexible business back-end, internally facing, albeit one that slots in nicely with digital transformation efforts given the great API-first approach. So you can integrate your customer apps, websites etc into Salesforce without too much headache - Salesforce becomes the kind of 'brain of your business'. This across Sales, Support and Marketing.  At Hive we have done a lot of that, and Salesforce provides a great number of integration options to tie processes together. The end goal is to have a single/simple view of what is going on, implement business logic centrally/consistently, rather than hop around different CRM, eCommerce, ticket system etc back-ends. In terms of back-end, we are always looking to change processes and rationalise systems to simplify our integration architecture. Where we can't, we integrate.

Now, however, Salesforce have a beast of a 'front-end' eCommerce framework, quite a highly regarded beast. You read your Gartner, Forrester analysis, it's up there with Oracle ATG, SAP Hybris etc. They have an impressive client base - Adidas, Lacoste, M&S etc. This is proper 'enterprise', SMBs and the like will probably stick with lower priced offerings. It will be interesting if the pricing model will still be based on revenue share as that is counter to how Salesforce generally operate.

Digging a bit deeper into customer reviews by G2Crowd, the consensus seems to be positive - again, being a 'beast' comes with complexity, but in terms of capability it seems to be very much an 'all-in-one', covering multiple devices across online and in-store. Expect the term 'omni-channel' to be used a lot (more) by Salesforce, more in the context of Sales as they have beaten the Service one to death. That's 'digital' and 'omni-channel' now. 'Domni-channel'?

Architecture-wise it has a lot in common with the core Salesforce platform - SaaS, multi-tenant, customisable to some extent, frequent upgrades etc. It should slot in nicely. One major consideration, as with all eCommerce platform, is the overlap between this and the other Salesforce 'CRM' offerings, in that an eCommerce platform always provides CRM functionality of sorts, whether that be sending emails, order management etc. A front-end always has a back-end, as I always say (I do actually, when you are trying to rationalise systems it can be a problem). So defining the responsibility of the bits of the Salesforce empire you were bludgeoned into buying is important. ie where will your product catalogue be mastered? Where will emails be sent from? What back-end interface will your agents use, for what purpose? If you have Demandware and Marketing Cloud, which predictive recommendations engine will you use? All in the architecture definition.

A key point to make is that Demandware seems very 'retail' oriented, especially fashion etc. Shifting stock. I am not sure how it will fare in the ever increasing proliferation of service 'wrappers' around things in the IoT world - there is a bit more to that journey than just filling your basket, taking a payment and throwing the goods over the wall. I thought Zuora may integrate with them but no, it tends to be just payment gateways. There is something to be said for developing a buying journey that is more tailored to service wrappers and subscriptions, the 'subscription economy' as Zuora like to put it. Maybe something for the future, or demonstration of my lack of Demandware knowledge.

As an aside, at Hive we currently use an open source framework called Spree as a basis for the Hivehome shop and integrate the web checkout journey with Salesforce. We log the journeys in Salesforce. If someone drops out and calls the call centre, the agent can pick up the basket from within Salesforce and progress the sale. It has worked well, and the Salesforce team works closely with the web team to continually develop this. I bought some window/door sensors through it at the earlier in the week, smooth. Over the course of the year this will be changing and expanding quite considerably, exciting times ahead.

So, in summary, your 'opportunity' to go 'all-in' with Salesforce has just increased. If you are a retailer, I would hide under the counter as the sales guys will hunt you down, they will find you, and they will throw 'Domni-channel' at you.

Anyway, in summary it looks to me like Demandware would be most appropriate for heavy duty Retail/Enterprise needs, with more mid-market / open-source solutions appropriate for more moderate needs. But don't use Salesforce Sites. Let's see where they go with Demandware, but all in all, for the continuation of the Salesforce story, it's a logical and positive move.

Delivering CRM in a Consumer IoT Company - Intro

Wow, it's been 4 years since I last did a blog post. I guess then I got all excited by the 'cloud' thing then it wore off as that all became pretty standard. Though I still see a lot copy about cloud still as Oracle etc grapple to catch up in that respect, sad times.

Anyway, after spending a lot of time grappling with the world of consumer IoT for Hive (British Gas 'Connected Homes' business) in the last 3 years I thought, actually, there is something interesting to talk about again.

I am not afraid to say it, but over the last 3 years helping grow an IoT business within British Gas, I have probably been more out of my comfort zone than in the previous 10. On occasion I have felt similar to Roberto Soldado below (yellow one) when he was completely done by Firmino in that recent Europa League match.





Part of this is because IoT is a pretty new field when it comes to the consumer mass market - when thinking about some of the business and tech challenges, in a lot of areas there have been limited reference points. Part of it is due to being part of a 'startup' for the first time - Hive was kicked off as a 'lean startup' within British Gas (agile etc - the intention of Bimodal mode 2). On the first point, there will be subsequent blog posts drilling into the business and related tech challenges. On the startup point, I want to elaborate here.

On the majority of CRM-related projects I had been involved with before prior, a company had an old CRM system/process which they wanted to replace with something more modern. Some vision to move the established business from old world to new world. So there was a decent amount of context. In a startup, there is no 'existing' business to start with, no existing systems, so the world is your oyster. Blank canvas. In 2013, we ventured into the world of consumer IoT and not many others had gone there at that time. It's still early days now.

What I found at the start was that it was great being part of such a intentionally hand-crafted small 'Connected Homes' team. Co-located with startup veterans - entrepreneurial types who had exited their own startups, commercial and marketing teams, operations and the tech teams. As standard in this kind of 'digital' startup you have various tech teams which CRM needs to work very closely with nowadays. The 'CRM' system isn't managing all customer interactions these days, not like it used to, we all know that. It's a close cross-team effort with the web (site/shop etc) team, mobile apps team and in the case of IoT, the team that owns that IoT platform (more about that in a subsequent post).

All of these platforms touch the customer in different ways. Then you have the proliferation of tools around the edge - social marketing tools, ESPs etc - it can be complex to align to ensure a good end to end customer experience. The responsibility for the 'CRM platform' (in this case Salesforce) needs to be agreed - ie you don't want to be actually storing all device interactions in Salesforce, for instance, or surfacing high volume web pages from Salesforce etc etc. Equally you don't want the web team building a bunch of custom support tools which replicate Service Cloud. Salesforce has it's place and I think from the word go we have done a good job of blending that platform well with the other tech that exists at Hive. Architecture basically (oooh, don't say that out loud in a 'lean' environment). More on that in another post.

An interesting observation from starting to work with these teams back in 2013, and people with a startup mentality is that to some extent, you feel very 'enterprise' being the owner of 'CRM'. Or maybe a bit of a dinosaur. You know, Salesforce is great, and UI-wise may be getting there with lightening etc, but you show it to someone and compared to the funky stuff they are doing with web/mobile UX or the funky IoT tech, they will quickly switch off! Also these guys are all in 'open-source' land (a land sometimes needing a decent amount of containment, another post for that) so anything 'proprietary' is best case seen as 'boring', worst case 'evil'.

So, the challenge awaits, it's Feb 2013, and we need to go live as a new business in September 2013. We need a system to run this new and undefined business, to enter a new and undefined IoT market. And I need to work out my purpose in life in relation to the other teams. Salesforce and I are NOT dinosaurs, no we are not...that's the intro, more to come in the next post.

Ellison's Dodgy 'Cloud', Oracle Fusion and Gillette Fusion

Why not just add to the confusion?

Earlier this month Oracle CEO Larry Ellison took to the stage for a much anticipated announcement of Oracle's Cloud strategy. This includes the long-awaited 'Fusion Apps' and other 'cloud' services. So, as always I ran this by the NIST criteria to assess whether these are proper Cloud offerings. Well, that's easy. By virtue of the offerings being virtualised single-tenant, they aren't. It's more akin to the old ASP/Outsourcing model, with none of the efficiencies of the more modern multi-tenant architectures such as Salesforce.com, or indeed Facebook etc.


Lots has been written about this keynote so I am not going to retread too much old ground, the links below give decent overall POVs with which I mainly agree:

• Phil Wainewright summarises the Oracle motivation for continuing to milk their traditional model very well in this interview. He also pulls apart the 'security' argument against multi-tenancy that Ellison uses in this blog post.
• Frank Scavo point out that Oracle have exaggerated the availability of Oracle Cloud and pulled Ellison up on the claim that they have spent the last 7 years working on 'Cloud' apps. Project Fusion started then, but there was no mention of a Cloud model until recently.
• Ron Miller points out that their model is 'expensive rather than cheap' and 'complex rather than simple' and thinks this is 'doomed to failure'.
• The list of the 100 Cloud Apps by Constellation Research. As they quite rightly point out, Oracle are pushing it a bit by claiming that the 500,000 and 100,000 record version of Fusion Marketing Segmentation are two different apps.

Oracle have let me down

I am disappointed in Oracle. Although I have in recent years moved away from their technology to Salesforce and equivalent Cloud solutions, I thought they would catch up. Ellison is swimming against the tide, hanging on to the old world. He is 68 now, maybe with little incentive to revolutionise their approach for the long-term.

The Oracle Fusion Promise (2006)

I remember in 2006 Oracle had made many big acquisitions and started a project to bring the best of each of these together, announcing that:

"The Fusion project aims to meld technologies from PeopleSoft, J.D. Edwards and Siebel Systems"

A guy called John Wookey, Oracle's senior vice president of application development made this announcement. At the time I was excited about the potential of this. Interestingly Wookey left Oracle in 2008, the year the first Fusion apps were due to appear. Oh, and he now works for Salesforce as of 2011.

So now we are in 2012 and we find Fusion Apps being the main chunk of the 'Oracle Cloud' offering along with a Java hosted PaaS and a hosted database service. No mention of PeopleSoft etc, that idea must have been binned.

Elision mentions in his keynote that 100 Fusion apps are available NOW. Well, I tried to register for a demo and just got the following (same as Frank Scavo above):

It has been 6 years! And still nothing. Disappointing. I just want a trial of Oracle Fusion CRM, not too much to ask.


Want Fusion? Go to Gillette.

Co-incidentally Gillette also announced their 'Fusion' range in 2006. If we compare Oracle Fusion to Gillette Fusion which was also announced in 2006 we see that Gillette have been much more productive. They have released 6 razors in that time and numerous complimentary products! From the Gillette Fusion Power to the Fusion ProGlide, now that's more like it. Oracle even pay tribute to this with the colour of their 'Register Me' button above.

Oracle CRM OnDemand - Where are you?One of the blog posts quite rightly asked why Oracle CRM OnDemand is not in the list of 100 Cloud apps. I implemented this a few times and it was like a more basic version of Salesforce Sales Cloud without the extensibility/platform elements. For a long time this was mooted as a competitor to Salesforce, and it in fact sits on a proper multi-tenant architecture. However it looks to me like Fusion CRM has superseded that so I would not expect it to be supported/enhanced for too much longer. I tried to get clarity from Oracle on this, my queries were ignored:

@OracleCloudZone @Oracle Why is Oracle CRM OnDemand not included here? It is a multi-tenant cloud app. Decommisioning?


@OracleCloudZone @Oracle I will try again. Why is Oracle CRM OnDemand not in this list? Is it classed as legacy now?


@OracleCloudZone Can you please reply to my question as to why Oracle CRM OnDemand is not in the cloud apps list?


@Oracle @OracleCloudZone hello? This is all a bit one way. Please answer my question. Or are these twitter accounts just 'one-way'?


@Oracle @OracleCRM @OracleComms @OracleCloudZone Why is Oracle CRM OnDemand not in this list? Surely you will respond?


@OracleComms Does the 'Comms' bit stand for 'communication'? If so, is that just one-way? You are not responding to me? @Oracle@OracleCRM


@OracleComms I am asking why oracle crm ondemand is not listed in the list of 100 cloud apps announced by larry Ellison recently.


@Oracle Hi, this is my daily repeat of the question 'Why is Oracle CRM OnDemand not in your list of the 100 Cloud apps?' Still no answer


Still waiting for an answer.



So I can't see this ending well...

Based on the general understanding that proper 'Cloud' multi-tenant services are more cost effective and allow more focussed/rapid business innovation, I expect Oracle to fail slowly in the Cloud race until they buy a proper Cloud platform. But I guess we will know more when Oracle make the Fusion apps and Cloud services generally available. 


Until then you can always get a bit of Fusion Hydra-Smooth Aftershave Balm to keep you going.

Salesforce Summer 12 and Disco Mirrorballs

My friend's house is never finished. An extension, new kitchen, patio on the garden, wires hanging down etc, absolutely lethal. Then knock it all down and start again. He loves it.

This got me thinking about the Salesforce upgrade path over the last few years. One of the reasons I moved to work with the Salesforce platform was because of the constant upgrades, making the platform effectively future-proof. However, after an initial good start, adding useful features such as APEX and better reporting etc, I am losing the faith. The focus on adding new 'look at me' functionality as opposed to developing the core platform is frustrating.

Going back to my friends house, he has recently redone the kitchen and redecorated the bedrooms. Good. Rooms that are often used and every day. He can walk into those rooms and appreciate the work that has been done. I can imagine though that if he had got Mark Benioff free reign to improve his house he would be sitting with a bill for a back garden water amusement park, a carousel in the living room and a lap dancing bar to replace the bedrooms. All very nice, but what about the missing wallpaper in the living room?

There are many ways in which the existing platform could be improved in areas such as build/deployment (too manual+resource intensive), or workflow tools (pales in comparison to Siebel). Since the introduction of Chatter, I think Salesforce have lost the plot a bit, and are focussing too much on whizz-bang new features to grab the attention of CIOs who are moon-walking to the tune of the social enterprise.

Anyway, enough metaphor. Let me get some quantitative proof by rifling through the Summer 12 upgrade release notes. Here is a summary of what is to be delivered, categorised as 'Replace the broken fridge' (useful) or 'Install kitchen disco mirrorball' (less useful). Ok, maybe not quite enough metaphor...

Chatter Messenger: Kitchen Mirrorball
Just get Google Apps - as a collaborative platform it trumps Salesforce.

Additional Chatter Enhancements: Kitchen Mirrorball
Chatter, chatter, chatter. Too much focus on this nice but certainly non-essential activity feed feature. Get me some decent way of creating a fully automated release build of an environment without any manual steps instead please.

Sales Cloud Enhancements - New Fridge
The bread and butter. No qualms here.


Case Feed Enhancements - Kitchen Mirrorball with added smoke machine
Unsure about Case Feed. Again, branching into new territory when the standard Service Cloud could really do with some better routing/queue functionality to bring it closer to Oracle's RightNow in that respect

Service Cloud Enhancements - New Fridge
A bit more bread and butter, but minimal

Salesforce Knowledge Enhancements - Kitchen Mirrorball
Who knows of a client that has the Knowledge module? Why this isn't free is beyond me.

Live Agent Enhancements - Kitchen Mirrorball
Nice but niche

Analytics Enhancements - New Fridge
Now it is going to tell you when it times out which is good. But seriously, some improvements to cross filters etc - I really like what they are doing here and all credit for going back on the initial charging model


Data.com - Kitchen Mirrorball
Lots and lots about this in the release notes. But the use of this is still fairly niche.

Schema builder - Kitchen Mirrorball
Noddy.

Site.com enhancements - New Fridge
I think this has potential, but ultimately any enterprise that wants to build out web facing apps are going to use Heroku. If they really beefed up this offering and made it scalable it could be very useful (abstraction layer on top of Heroku for instance?)


Force.com: Flow / VisualForce / Apex - New Fridge
Unfortunately improvements to the core platform in these areas tend to be quite minimal, but having said that they are certainly welcome. Flow has the potential to be useful if they bolster it a bit to be more of a behind the scenes workflow engine rather than a scripting tool. VisualForce and Apex are incredibly core to the platform, and the increases in limits and improvement developer tools in Summer 12 are welcome.

So I guess I would sum up the Summer 12 release as follows:

False Cloud, Cloudwashing and Dishwashing

I used to be a dish-washer for a restaurant when I was a lot younger. The value of my service was easily quantified - either the plates and pans are clean and the chef always has a good supply, or not and you get torrents of abuse and ladles thrown at you. Ahh, those were the days.

Unfortunately when it comes to Cloud, and specifically cloud-washing, the lack of value in a service is not as obvious. A definition (there are a few) is as follows:

"Cloud washing... is the purposeful and sometimes deceptive attempt by a vendor to rebrand an old product or service by associating the buzzword "cloud" with it"

It also equates to Mark Benioff's 'False Cloud' line, and is important when it comes to judging whether a service does offer the benefits of the cloud model (which have been well documented so not covered here).

The motivation for writing this post was a recent article I read which seems to me either to be a blatant attempt at cloud-washing, or just a demonstration of misunderstanding. This frustrates me because 'cloud' is not a new concept any more and NIST have provided a widely accepted standard definition. Please, have a quick read:

Article charged with Cloud-washing: Case Study: Making Cloud ROI a Reality
Cloud-washing indicators: "Outsourcing made the most sense." "install new off-site servers", "leases physical servers"

What they really mean: Outsourcing and virtualization


It is pretty easy to take the essential characteristics of cloud outlined in the NIST summary and make a judgement against any offering as to whether it it 'true cloud'. So in this instance does the article fit the cloud definition at all? Here goes:

On-demand self-service. A consumer can unilaterally provision computing capabilities, such as server time and network storage, as needed automatically without requiring human interaction with each service provider.

No: No indication of putting provisioning in the hands of the consumer


Broad network access. Capabilities are available over the network and accessed through standard mechanisms that promote use by heterogeneous thin or thick client platforms (e.g., mobile phones, tablets, laptops, and workstations).

Yes: The capabilities are available over a network


Resource pooling. The provider’s computing resources are pooled to serve multiple consumers using a multi-tenant model, with different physical and virtual resources dynamically assigned and reassigned according to consumer demand.

No: This is normally the easiest was of judging whether a service is 'Cloud' or not - simply outsourcing does not give you the economies of scale associated with a multi tenant architecture


Rapid elasticity. Capabilities can be elastically provisioned and released, in some cases automatically, to scale rapidly outward and inward commensurate with demand.

No: No indication that the consumer can scale up and down rapidly. The fact 'leasing physical servers' is mentioned suggests this would not be as easy to scale as a proper IaaS platform such as AWS.


Measured service. Cloud systems automatically control and optimize resource use by leveraging a metering capability1 at some level of abstraction appropriate to the type of service (e.g., storage, processing, bandwidth, and active user accounts).

Yes: The virtualization element here and moving from 50+ servers to 4 suggests this is true


So, I come to the conclusion that although the term 'cloud' is mentioned 17 times in the article, what they really mean is outsourcing and virtualization. Not Cloud. So please, anyone who starts an article on 'cloud', have a quick look at the NIST definition so we are all on the same page. Otherwise I will throw a ladle at you.


- Posted using BlogPress from my iPad

The Multi-Multi-tenant Upgrade Jigsaw

One of the most commonly cited advantages of an app sitting on multi-tenant architecture is the fact it is frequently upgraded so each customer is always on the 'latest version'. Historically the upgrade path for on-premise enterprise software had been a complete pain. Taking the 'one-click' upgrade path was about as effective as throwing a bag of spanners at the thing, as it had been butchered beyond all recognition. Not 'out of the box' anymore, the box had been burned and the ashes whacked repeatedly with a baseball bat. So the standard upgrade routine would basically have a look and then run away laughing.

With a multi-tenant app, the abstraction layer prevents customers from butchering their app to such an extent. This results in a fairly smooth upgrade process, but then a new issue rears its head. The upgrade itself and the timing are forced. It is not possible on a true multi-tenant platform to choose when the upgrade occurs. All customers are upgraded around the same time. So, let's take a couple of multi-tenant examples from my experience:

Salesforce.com : Upgrade occurs to the Production instance every 4 months.

The fun: One month prior to the production instance upgrade, certain sandboxes are upgraded by the Salesforce upgrade fairy. Now, in an enterprise environment you will likely have at least 2 sandboxes - Dev and Test. You may also have a separate UAT sandbox, maybe a Staging sandbox also. You might have a parallel development stream with another Dev and Test. Anyway, the point is that to avoid a sandbox being upgraded early by that naughty but ultra-keen fairy, the only way is to refresh it in a certain window of time, differing depending on what server instance it sits on. This kills the current sandbox and creates a new instance copied from Production. So this needs to be planned well in advanced into the dev schedule to avoid for instance your dev sandbox being on a different version than your test one, and to make sure the release path sandboxes are on the same version as production. And of course if you are midway through your dev schedule when the Production upgrade occurs, and your sandboxes are upgraded at the same time, you need to regression test the in-development functionality. It is also a good idea to have one sandbox upgraded early to regression test current Production functionality against the forthcoming release. To get the refresh schedule right I try to get my head around it all by drawing a couple of timelines. Firstly for the Summer 12 Salesforce release I drew myself this:

And generically as a reminder for the future I summarised the above paragraph into this:

Now lets look at Zuora:

Zuora: Upgrade occurs to the Production instance every month.

The fun: One week prior to the Production upgrade, all Zuora sandboxes are upgraded by the Zuora upgrade pixie. This necessitates a regression test of the current dev cycle, and a regression test against the current production functionality to pre-empt any Production issues. However, outside of this schedule is an absolute avalanche of patch releases by the mini-pixies, which with the best intentions, it is impossible to keep up with. Usually there will be about 1 days notice before the sandboxes are patched, then a few days until production. If you are lucky you get some release notes, but usually not until after the sandboxes are patched.


NOW integrate Salesforce with Zuora, and the solution as a whole has 2 multi-tenant platforms, with sandbox and Production environments upgraded on different schedules. Bloody fairies and pixies running around all over the place. The joy of forced upgrades. Then maybe add on Eloqua, or BigMachines - the more the merrier!

But we choose to enter the multi-tenant world so can't complain. Three points to make:

-When planning your dev/release schedule, review the upgrade schedules and ensure the impact of forced upgrades on resources and release dates is taken into account
-Automate your regression testing to the max to gain confidence in the upgrades quickly, because you will have to! Review release notes and prioritise areas that have changed.
-When cloud platforms prove themselves to be more robust in the future, the need for regression testing may reduce. Maybe.

Upgrades are more likely to have an impact on the more complex implementations which are found in enterprise implementations involving lots of integration, customisation etc, and spanning more functional breadth than smaller implementations. You probably don't need to worry too much about your little standard Sales Cloud thing you stood up for the old lady next door. But who wants to work on a small implementation?

- Posted using BlogPress from my iPad

The Cloud Lone Ranger and A Chinese perspective on Salesforce.com

There is apparently only one
man representing Salesforce
in Beijing, the 'Lone Ranger'.

I popped across to Beijing recently. What a huge place with tons of people. Takes a bit of getting used to, not quite the same etiquette as the UK when it comes to queuing or letting people get off the Metro train. Lots of 'scuffling'. Culturally it is certainly a bit different and if a day went by without me getting zapped by a static electric shock it was a surprise. Anyway, besides 'holidaying' and all that, I was curious where they are in terms of the current shift to Cloud Computing. The move to the various cloud computing models is quite well advanced in the US and UK, how would China compare?

I thought a good barometer would be Salesforce.com and so I decided one day to meet up with a consultancy there that specialise in this area, Meginfo.  I wanted to see if Salesforce has any foothold in China and also learn a bit about their offshore delivery approach in general.

Meginfo specialise in Salesforce for Chinese and overseas clients. they have done Salesforce work for some very large multinational enterprises and top tier system integrators. Their owner Aaron Lau has worked with Salesforce since 2005, firstly for a top 10 Salesforce end user then as a freelancer. This culminated in him starting Meginfo in 2008. The company now has 30 employees and has been consistently profitable.

I did also want to visit Salesforce in Beijing but Aaron said 'he' is out of town. One person to cover Beijing!? A mysterious 'Lone Ranger'? Digging deeper into this I found the perception that Salesforce could really improve their efforts to cover China. Anyway, here is my discussion with Meginfo:

NP - How did Meginfo start given that back in 2008 Salesforce was still pretty unknown, especially in China?

AL - I knew many foreigners and went to different countries such as Singapore, Australia, and also the Salesforce headquarters and salesforce online community. Many people had requirements, so i connected with them  and put myself forward to do the work. Eventually there was too much work for just myself so I started building the company. We basically know everything about Salesforce and focus completely on it - we have a team that is able to resolve most difficulties without help.

NP - Are the companies you work for mainly overseas companies?

AL - Most of them are US companies, most do not have any people in China. We have 2 type of customers - Partner (SF consulting companies who have their own customers) and other are End Users. For our partners we are generally the offshore development team for them. For other customers we need to do a lot more work because we need to work with them to gather the requirements directly. US companies prefer not to have lots of documents, unlike companies from Japan, they prefer to have very detailed specs that the developer can read. For US companies we need to spend a lot more effort gathering requirements, do the spec then provide the quotes and options before we do the actual work.

NP - Quite often some kind of documentation can be handy as output to get the right level of thinking/architecture done before embarking on build but quite often a client only knows if the application is 'right' once they start using it.

Aaron Lau, Owner of Meginfo

AL - Yes that is right.

NP - Do you buffer in some time for reviews and changes to the built application along the way to account for the fact not everything can be anticipated/captured by documentation?

AL - That depends on the size of the project. For many projects they are really small - 80 hrs, we use a very basic development cycle - we did the documentation then the development work then testing and deployment. But for some bigger projects, the onces we have been doing for years, we really need to have a process where customers can get involved and we are having updates every day. Also we have integrated Chatter into that so customers can have realtime interaction with our developers, testers and project managers. They review the system and whenever they have comments they can post them into chatter - our people are immediately aware.

NP - How do you communicate with clients - web conf etc?

AL - Skype, GotoMeeting, we recently switched to Webex as it has much better performance. For developers they can read and write english but not speak so much, so mostly they use email and Chatter.

NP - In terms of the projects that you have done, what do you think are most important factors

AL - Communication with customers, the right people to to the work, and the dedication of the team. There are so many companies providing offshore services but working on multiple projects at the same time. When the developer is working, they need to be to be focussed on the project, not distracted by some other work. Some years ago I worked for a company who provided very similar services but the success rate was not very high and a major reason was that people at the time were working on multiple projects.

NP - How important do you think the dedication and ability of a client is? From my point of view I may have a very good team, but if the client has not allocated a good project manager for instance or the client does not allocate time for the right people to be involved, then it can be difficult.

AL - When we are working on a project we need to let them know the importance of being dedicated to a project, we push them to reply to emails etc and sometimes client does not like that. But if we deliver the project successfully that makes everyone happy. In process of doing the work we need clients very involved in project. It is hard and sometimes clients do not like being pushed.

NP - In terms of the companies you are in competition with in China, what is the main differentiator for Meginfo?

AL - I think right now we are the biggest Salesforce consulting and development company in the Greater China region. There are many big companies in India. India is famous and popular for outsourcing and they speak English well. In China we do not have a lot of competition, not because others are not strong, rather that Salesforce is not used too much in China, there is not much awareness. Those companies focus more on MS CRM and SAP, some other technologies. Those other companies are really good at marketing and sales, dealing with people, they are very good at selling to clients and we want to get better at that.

NP - Are there any cultural differences between working for companies from different regions? A Japanese company vs an American companies etc?

AL - I thing that the culture difference is very big. Eastern culture is dramatically difference from Western culture and is one of the difficulties that our developers are having right now. American people tend to be very dedicated and very straightforward, and whenever there are mistakes that they make or we make, they point that out. In China, I am not saying it is bad for the Chinese culture, but it is just the culture, that when we make mistakes we do not say anything and try to fix it, we try to not show we made mistakes. We are not very good at being open. It is almost like an opposite of cultures, a conflict. When I brought my people to US to Dreamforce when, when we got out the airport everything was different, it is something that people living in China do not know. I send lots of articles and blogs that I read and I send them to our internal Chatter to help them understand the culture and the difference.

NP - Salesforce as a company is very big, bold and open and very much has this 'Social Enterprise' vision. This is about being more open and sharing things internally and sharing things externally with customers over Twitter and Facebook. Salesforce isn't big in China now, but I get the impression that this kind of vision would never be something that Chinese companies would aspire to.

AL - I think they would. There are a few Chinese companies that are providing very popular social networks and there are hundreds of millions of people using the micro blog services. We have customers who want to integrate their system with social networks. Chatter is going to be a big one - its going to change the way that people interact with other people within a company. We use Chatter to talk about tech stuff and daily lives - people review and post comments. In my company I have a few opponents - they point out my mistakes and it is totally fine using email, chatter, any way - as long as their voice can be heard i am totally fine with that. We are trying to build a company that is more like a US company.

NP - Is a typical Chinese company very hierarchical?

AL - Yes, that is right. Lots of bureaucracy and people at different levels telling people below what to do and the end result is people down below they prefer just to listen, they cannot express their ideas, and that is something I believe is pretty bad and we do not want that to happen in our team.

NP - In terms of Salesforce what do you think the major strengths and major weaknesses are?

AL - A major advantage was the free edition but hey have stopped that now which is a pity, we did set up a few free editions before it was shut down. I think the free edition has lots of potential in China. It was a way that companies could start feeling the benefit of the salesforce platform easily.

There are so many companies providing cloud computing platforms that are essentially 'fake clouds'. In china there are so many companies saying their services are cloud computing services but they are really not. So the free edition was a good way to promote Salesforce advantages and get as many companies as possible to start using Salesforce. I think Salesforce need to do some more marketing activity in China because at this time I don't believe salesforce put any major effort into marketing campaigns in china.

NP - They do everywhere else...

AL - Yes, but china is not a market they care about. The companies using Salesforce in China are generally US companies and for local companies they still do not know salesforce, they do not know what Salesforce can do.

NP - They buy their own servers, have their own data centres etc?

Al - Yes. Though the is a Chinese company which provides very similar services as Salesforce does.

NP - 800apps?

AL - Yes, I think so. They even have the same user interface.

NP - I talked about that in another blog. That must have some Salesforce code in there, the similarities are too striking.

AL - Yes, I also tried it myself. It is not a company that can provide the same level of services as Salesforce can do however.

NP - No.

AL - So I believe the chinese market has lots of potential for Salesforce but again Salesforce need to care about that, they need more people promoting their advantages and they need to support Chinese customers.

NP - Is Salesforce perfectly accessible within China, no problem with access given the Great Firewall?

AL - No problems. There were a few cases in the past when the earthquake happened in Taiwan when the internet between China and US was broken at the time but it did not last too long, a few days. Also Salesforce moved their data centre from Singapore to Japan, so the connection between China and new data centre has been great since the move. It is almost the same as US people get.

NP - Where do you see Salesforce and Cloud Computing in the next 5 years?

AL - I think Salesforce will finally become a very popular system not just in the US but also in China. It is a platform that is so flexible and easy to use and very powerful so a company can use Salesforce to manage almost everything. We are using Salesforce to do project management, to do financial work, manage expenses, invoices, salaries, everything is managed within Salesforce.

NP - China is maybe how the UK was about 5 years ago in terms of the adoption of Salesforce, 5 or 6 years ago most companies would say 'no way am i using that', not secure etc.

AL - I think another reason is 5 or 6 years ago Salesforce was not really as powerful as it is today. Now it is a fully fledged cloud computing platform that is really really flexible

NP - I keep hearing about China building massive cloud computing data centres as big as many football pitches etc. What is going on there?

AL - I think most of these projects are sponsored by big companies like IBM, but they are not being used by companies. I am not a person who can say they are failures but there are really not people using them. They are not as popular as those cloud computing providers in the US. But you are right, as far as I know there have been so many projects in China that claim to be cloud computing but again they do not have users.

NP - Where do you see Meginfo in 5 years time?

AL - Our goal is to become the best Salesforce consulting and development company in Asia. We have succeeded with every project. So in 5 years we are going to be an even bigger team and will be able to work with a greater number of customers.


So, looks like Salesforce either are not too bothered about the Chinese market right now, or are biding their time. Rest assured I did not spend all of my Beijing holiday debating this. No, I also spent a good amount of time doing my Service Cloud and Sales Cloud certifications as well. That's the life.

Just as an addition - see this video of the Meginfo office - silent. Big difference between this and a typical UK development team (which can be fairly boistorous) , and emphasises the difference in culture that Aaron talked about.