Scrum and Salesforce

I have just read a blog post where someone talks about 90 days as really short implementation time in contratst with a typical 18 to 24 month implementation for a CRM system. Some implementations can be even shorter. So for users who are happy with the standard Salesforce.com B2B CRM model I have undertaken a 5 to 10 days implementation engagement including end user training.

So what has this to do with Scrum? Well firstly Salesforce.com itself uses Scrum to manage all its own product development and they have seen a 60% productivity improvement as a result.

The difference between Scrum and other project management methodologies is that it emphasises managing implemented system functionality against a fixed time period whereas an approach such as Prince 2 focusses on managing time against a fixed functionality scope.

The challenge that both approaches face is that project estimation is uncertain but in the first case you get a fixed price but a risk that not all the functionality will be implemented. In the second case you might get all the functionality but at the risk of cost overrun.

The benefit of using Scrum is that you can decide at the end of the fixed period whether the return that you would get from the missing functionality is worth the extra cost. The reason that Salesforce.com makes this possible is that unlike other types of software infrastructure (particularly on premise databases) you will have a working system. It may not do everything but it will function and you can then evaluate it against the real world requirements.

Its not just technology

A lot of the work that I do is helping people get started with Salesforce.com and we sometimes deliver this help as fixed time package of work. Now these work packages (QuickStarts) are pretty short engagements (3 to 5 days) and we ask the clients to do some pre-work so that they can get the maximum return on investment. This consists of some online training and completion of a questionnaire which outline their goals for the engagement.

You know what? A lot of the time it doesn't get done. People are so busy with their day to day work that when we turn up on site on the first day we have to spend the first half of the business process review going over stuff that they could have covered in the pre-work.

Well last week was different we turned up on site and not only had the clients staff done the training but they had started researching some third party apps from the App Exchange. "Hey this is great we can really get moving here" I thought even when they started quizzing us on some of the deeper aspects of the third party apps which quite frankly we had not used before.

Unfortunately after a while it became clear that they were complete techno junkies and had not thought very deeply about anything other than technology. The conversation degenerated into something similar to some of the patients consultations that my wife gets involved in. Occasionally she will get someone coming along to her GP surgery with a cold but they are convinced that they have the Ebola virus and will have printed out a sheaf of notes from the Internet which prove that they have all the symptons. When we probed a little we found we had a similar situation. Our potential clients had tried to use Salesforce but had loaded their data in to it in way that showed they had not understood the Salesforce.com object model. They also had no clear vision of the processes they wanted Salesforce.com to support. Even worse they could not grasp the importance of change management although they were concerned that taking staff away from front line operations for user training would negatively impact their productivity.

Sadly it all ended in tears. Their assessment of the value that we could bring to their project was based solely on our knowledge of technology specific to their situation and we could not convince them of the crucial importance of people and processes in their CRM deployment. I wish them well but my fear is this will join the statistics of failed CRM projects that didn't work because those in charge failed to adequately manage the change process.

Of course technology is important and one the key reasons we specialise on Salesforce.com is because it offers a superior technical advantages but the success of any CRM project depends on getting the people and process issues right. 5 rules that were presented by another Salesforce.com consultant (Apirio) at recent seminar that I was involved in provide a neat blueprint for CRM success:

• Lead from the top
• Get the requirements right
• Scope the business impact
• Invest in change management
• Plan for support

Gnu Force.com?

In a recent interview with the Guardian Richard Stallman GNU founder and Free Software campaigner said that Cloud Computing is a trap http://www.guardian.co.uk/technology/2008/sep/29/cloud.computing.richard.stallman

"One reason you should not use web applications to do your computing is that you lose control," he said. "It's just as bad as using a proprietary program. Do your own computing on your own computer with your copy of a freedom-respecting program. If you use a proprietary program or somebody else's web server, you're defenceless. You're putty in the hands of whoever developed that software."

At the root of this is his fear that once they have got you hooked they can make you pay through the nose but is this realistic? Even if you put your software on your software on your own web server you would be at the mercy of a proprietary chip maker and if you want to implement a high volume internet application you need to have access to an internet backbone which is again another dependency even if you could access it directly.

The fact is we are all dependent on proprietary computer infrastructure of some kind and we have to rely to some extent on the vendors not extorting us too much really out of self interest. They understand that in order to grow their business they need to sell us more and that its better to have willing customers than a blackmail victims.

That is not deny the contribution that the Free Software foundation and the Open Source movement has made, but I would argue that its main benefit has been in fostering innovation. Allowing via the LAMP stack for example a host of Internet based businesses to be created. Its difficult to believe that a proprietary vendor like Microsoft would have facilitated that development because its focus would have been on meeting the needs of its target corporate IT customers.

Currently there is no doubt in my mind that the Salesforce.com .Force platform offers the best cloud computing infrastructure and that more and more people will start to take advantage of the benefits of moving to the cloud. These are largely connected with not worrying about building infrastructure but on concentrating on the added value of the application. The best response from the Free Software adherents is not to stand on the sidelines wringing their hands or taking the occasional pot-shots about privacy but take it head on and develop an open source multi-tenant database with a Java like Data Manipulation Language (DML) that could provide a real alternative. There's your challenge Richard. Good Luck!

Salesforce in the Cloud

Salesforce.com has been a leader in cloud computing since its inception. Although launched as on demand CRM solution other SaaS applications have always been part of the vision. You only have to see the No Software logo to realise that.

Two years ago Salesforce launched the Force platform and more recently VisualForce to enable third party developers to deploy applications on the Salesforce.com cloud platform.At the recent Dreamforce conference more announcements , such as Sites, were made which demonstrate the Commitment of salesforce.com to this goal.

Other bigger players are now beginning to show their hand: Google has had its app engine around since 2007 and Amazon is pushing its Elastic Compute Cloud (EC2) and most recently Microsoft announced Azure.

Who will win? Well without doubt Salesforce.com offers the best technical solution because its multi-tenant database scales so much more easily than the other solutions but its achiles heel is its reluctance to fully embrace a VAR model. With Amazon a vendor can buy the service embed his application and resell it and of course Microsoft have well established reseller channels.

Salesforce have taken the first steps along the VAR route by offering an embedded licence but this only applies to a single app. If a vendor wants to integrate other third party apps to provide a complete solution then the customer must buy a platform licence from Salesforce.

Its early days and this is very much an evolving situation. Its easy to see that Salesforce.com does not want to risk cannabalising its revenues streams and changing from a culture of centralised control to decentralised management is a challenge but if it wants to compete with the other vendors who are looking to park their tanks on the Salesforce cloud computing lawn, it will need to unleash its own attack dogs.

Welcome Forcists

I must admit this is not the first blog I have attempted. My original idea was to blog about general marketing issues but of the three posts that I made 2 were about Salesforce.com and then I stopped for a while because I couldn't really think of a direction to take. So as I spend most of my working day involved with Salesforce.com I have decided to focus on that and create a dedicated Salesforce.com blog. Ofcourse there are lot of other Salesforce.com blogs out there but I hope my slant on the subject will be a useful contribution. I will also flag up other bloggers that have interesting things to say about Salesforce.com. One question that has occurred to me: Is there a collective name for Salesforce.com users? What about Forcists? Perhaps not. Maybe we don't need one. After all there is no collective name for iPOD users. Or is there?