When Something Stands For Nothing

So, while listening to the new Headstones song (BINTHISWAYFORYEARS…check it out!) I was also looking through twitter, when yet another #SFDCrevealed tweet came across my stream.

Now, those who know me understand that I am not a big fan of the “attack ad”, or using any method of advertisement that boosts your service/product through the belittlement or trashing of a competitor’s service or product.  Think of Volkswagen – they have made some pretty good (and a few crappy) cars, but their ads are ever so impressive.  They have an aura to them – funny, sophisticated, different.  And because of that, they stand out.  Everyone remembers The Force ad from last year’s Super Bowl, or the No Bathroom Stops TDI commercial currently playing every ten minutes on my TV (and making me chuckle each time).  What’s missing?  Not one comment about any other car….ever.  And I truly wish other companies would start to follow suit.  However….not the reason for this post.

In their tweet, @MSDynamicsCRM makes mention of the SLA issue, and the fact that salesforce.com has none.  So, I thought I would go searching and look for the MS CRM SLA.  I found this site, which outlined the ‘financially backed’ SLA.  And kudos to Microsoft for doing so – it’s a great marketing campaign to put money behind your uptime.  But this led me to reading down further, to the exclusions.  Section D lays these out, and they are the following:

D. Exclusions

1.       Downtime does not include:

a.       The period of time when the Service is not available as a result of Scheduled Downtime; or

b.      The following performance or availability issues that may affect the Service:

i.         Due to factors outside Microsoft’s reasonable control;

ii.         Related to add-on features for the Service, including, but not limited to Internet Marketing or Reporting Services;

iii.         That resulted from Customer’s or third party hardware, software or services;

iv.         That resulted from actions or inactions of Customer or third parties;

v.         That resulted from actions or inactions by Customer or Customer’s employees, agents, contractors, or vendors, or anyone gaining access to Microsoft’s network by means of Customer’s passwords or equipment.

vi.         That were caused by Customer’s use of the Service after Microsoft advised Customer to modify its use of the Service, if Customer did not modify its use as advised;

vii.         Intermittent periods of Downtime that are ten minutes or less in duration; or

viii.         Through Customer’s use of beta, trial offers, early access programs and/or demos (as determined by Microsoft).

Let’s look at these in detail.  First point is a given – if it’s scheduled, it is not really downtime.  But when is the downtime scheduled? I have yet to find a site where Microsoft keeps track of the current status of their online servers for MSCRM or their scheduled downtime.  Looking at the next eight points though, I am hard pressed to think of a scenario where an outage would not fall under one of these eight processes?

• Did weather or such cause the outage? Not covered (which is expected and understood, but also raises the question of how many data centres does Microsoft have for their MSCRM application – Salesforce has five I know of, all highly secure and redundant for disaster recovery).
• Did an add-on feature cause the issue?  When looking at their add-ons, to me it is similar to salesforce.com saying “The outage was caused by reporting/marketing within salesforce natively.  Sorry”.  So, unless it can be unequivocally proven that your system has no add-ons that could have caused the issue, the downtime is excluded.
• The third point makes me laugh – hardware.  Tee hee…
• Points 4 through 6 to me, mean that if there is something you are required to do and you missed it, you are SOL when it comes to a financial recovery attempts.
• Downtime that is less than ten minutes in duration doesn’t count.  So, if an incident is eight minutes, the server comes back up and then crashes again, is it a separate downtime count?  Let’s also remember that in a month of 31 days, 99.9% uptime includes an allowance of 40 minutes, and that doesn’t include any ‘scheduled maintenance’ time.
• The final point I’ll just leave be.

When looking at salesforce.com’s history, which you can all view on the Salesforce Trust site, you can view the history of uptime for each and every server.  Taking a look at the incidents, the one outage for one server was 7 minutes, and the others were intermittent login issues.  These intermittent issues (all quickly solved by the way, and not long in duration) would have been covered under the exclusions list offered by Microsoft.  In fact, the last major outage I remember with salesforce.com (knock on wood) was January and March of 2010, and these were of durations of just over 60 minutes and 30 minutes.  Depending on the amount of scheduled downtime, the January 2010 may have still fit under the 99.9% uptime rule, and the March outage definitely would have still given 99.9% uptime.

Yes, Microsoft has done a wonderful thing for their clients with their financially backed SLA.  Is it all encompassing or give them an advantage over salesforce.com? Well, as we have just walked through, I don’t believe so – it makes for wonderful marketing and a great paper for their customers if they have a large outage that falls through the cracks of the exclusion, but it is not a major point of contention in the difference between the two applications.  All this, and it is time to answer the question @MSDynamicsCRM asked – is a trust dashboard enough?  To me, yes.  Yes, it is.