Follow us

Migrating Exchange to Google Apps: This is My Story

By | February 19, 2013 in Cloud

Migration Exchange

How It All Started

I would like to share with you my experience from this past summer in migrating from Exchange Server to Google Apps.

Some background: I have been an Exchange user for as long as I remember. It was kind of common knowledge that there just wasn't anything else out there, well maybe Lotus Notes - but doesn’t only IBM use Notes, and maybe also some of their "lucky" customers that got a good bargain on it?

Anyways, at SysAid (where I’ve worked for the past 5 years) we had Exchange serve us well for several years, but as we grew from a small company to a medium-sized company the fun began.....

As we grew, the IT team kept mentioning that the licensing model we had from Microsoft needed to be upgraded because we were reaching limitations on certain features, and capacity constraints were stretched to the limit. We started to outgrow our 2GB storage limitation per user that we were granted and needed to erase emails, or save our own .pst files.

Also our needs were growing. We opened two subsidiaries abroad and they were instructed to use OWA via browser for email and calendar. They complained it wasn't user-friendly and connecting Outlook via OWA was a real challenge. For some reason they couldn't manage to maintain an Outlook environment that worked for more than a few days. On top of that our hardware was getting old and needed to be replaced. We encountered a nasty downtime of 48 hours during Exchange’s attempt to repair a corrupted database, and numerous configuration errors that caused different parts of our email services not to function. Once it was outgoing emails that got stuck, then it was OWA being down for the weekend (no email on our mobile phones!!!!), and we even experienced bouncing emails on various occasions, not to mention the growing number of spam emails we started to encounter. All of this together got us thinking that we need a high-availability solution and also a Disaster Recovery Plan (DRP), and maybe even some experts to get the Exchange to do its job.

Finding a Solution

The IT team started to put together a budget for new servers, licensing, storage, backup, and so on.

Oh, and it turned out we needed to upgrade our backup software to properly support the daily backup of Exchange 2010! I was told that ever since we upgraded, there has been an error message in the backup logs stating that the backup may not be used. So that was another few thousand $$$, right?

Somewhere when the calculation passed $20K, even before we added the Exchange experts who we needed because anything more than simple maintenance was out of our IT team’s knowledge, we just said STOP. This was the trigger that made us decide to take a deeper look into migrating our email/calendar to the cloud.

Now I must state, at SysAid we are and have always been cloud-oriented: we use Salesforce, all our marketing tools are fully cloud-based, we support our customers with our SysAid Cloud, and use the same SysAid Cloud to manage our internal IT, which among other tasks even monitors the Exchange Server. So why didn't we consider moving the email to cloud earlier? For some reason we were convinced it worked just fine (minus the downtime and other hiccups we experienced). And like all other IT departments, we were overloaded with “putting out the fires” and busy with other projects. So it just felt right not to touch the Exchange—because it was working. But that was the point—it really wasn't working that great.

Now some will say that with better knowledge and more experienced staff and maybe a different approach, we could have managed to keep our Exchange and stay within reasonable budget and quality of service. But I ask why should we? I mean email/calendar is very important to us, but why do we need to master the management of the infrastructure and software?

So we asked the IT team to bring an offering to migrate to the cloud, and insisted to bring 2 offerings, not only a Microsoft cloud offering. They came back with a high level offering from Microsoft and just said that Google isn't as good as Microsoft so it isn't really a solution for us.

That is where I stepped in, rolled up my sleeves and gave Google a call. I was transferred to Ireland, which is in charge of my region and had a nice phone call with some senior sales guy that gave me all the details in a very simple and straightforward manner. Bottom line—1/2 the price compared to Microsoft! Since we needed 100+ licenses the Microsoft price per license doubled compared to its starting price (they both start from ~$5 per user per month). The Google guy made his pitch simple and to the point. He said that Microsoft and Google have two different products, some features are the same, some are different. You will occasionally run into a missing feature, but then run into 3 new unique features that don’t exist and vice versa. Because it’s cloud-based, features are being added all the time. So he asked me to embrace the idea open minded and also make sure that our users are aware that it is going to be a change.

He set me up with a trusted local partner that has a few hundred hours experience in migration and implementation projects, and we launched the process!

I got the offer that included the migration, training, and the licenses. I won't give you the costs, but let me just say that when we ran a 3-5 year cost simulation, it seemed to break even compared to managing it in-house, but that is only direct costs without surprises and other hidden costs.

We launched the process and set a goal to fully migrate within 3 weeks, which included an initial migration and test, training, and a final migration. We were told that the migration of data may take a few days to run depending on our bandwidth…we did after all have more than 200GB of data to migrate.

We installed the migration tools, defined a group of 20 users, which included everyone in upper management plus some additional key users. We wanted to make sure to get all departments onboard in the testing phase so they can later assist with the go-live process.

Once we migrated the data we asked each of the 20 users to log in and check their emails, calendars, and contacts.

We then set up a 3-hour training session for around 15 people and were ready to go live. One more migration run with the delta and we were ready to set the go-live date.

We ended up having to wait longer than we estimated for the migration to run, but we decided to go live anyway and let the migration keep running in the background.

Going live was exciting. We “pulled the plug” from our Exchange and directed all incoming and outgoing emails to Google. Then watched how the emails started to arrive via Google. Then we tested all our email accounts including our special CRM integration accounts, our HR job accounts, etc. We had some issues, but all of them were easy to understand and fix together with the assistance we got; most of them were even dealt with by our own team and yours truly.

Most of the 100 users managed with the training and by helping each other. We set up a password sync between Google and our LDAP with a tool called GAPS (Google Apps Password Sync). We set up a special forum in Yammer (our company’s internal social network) to communicate the steps of the migration and also for support for those who ran into trouble and needed help logging in. I was able to deal with most of the support issues related to the migration, and most of the issues were related to how-to questions. The cool thing for me at this stage was the social aspect that occurred. The support and assistance moved from traditional IT support to social IT support—everyone was helping everyone—tips and tweaks were being posted daily and soon enough we discovered some new non-IT department employees fast becoming Google experts in various topics: mobile apps, customizations, lab add-ons, and more. They were all sharing their knowledge and helping everyone else out.

How The Story Ends

Believe me, it wasn't all that smooth. We had some rejections, encountered some time zone- related bugs that moved some of our meetings an hour ahead, ran into technical issues with shared calendars, etc., but we entered the project open-minded and overcame the obstacles. We were smart enough to decide to drop trying to struggle with migrating the problematic calendars and instead set up all our recurring meetings with new blank calendars created in Google Apps. It was much easier and not too much work, and it gave us a good opportunity to update the days/hours of some meetings that were historically stuck on Monday mornings J.

Now, six months after we migrated to Google Apps, I can tell you that we are very satisfied with the decision. There was a very short learning curve and we got a lot more value out of the project than we initially planned. We started using Google Docs and Drive, we communicate with Google Chat, and we keep on discovering more features and benefits. And I almost forgot to mention the 25 GB of storage each user gets!

My message is—companies can run without Exchange because there is an alternative. You just need to check your company’s individual needs and decide. But please don't forget that support from your management team is crucial. Take it from me - if you don't have support across the board you shouldn't enter such a project until you get EVERYONE on board!

If you want to know more about Google Apps, check out this clip:

Please share your thoughts in the comments or on Twitter, Google+, or Facebook where we are always listening.

Oded Moshe

About Oded Moshe

Oded is VP Products at SysAid, with over 15 years of experience in various product and IT management positions. Proud father of two young (iPhone/iPad-addicted) girls and one baby boy (that they're trying to keep the gadgets out of his reach). Fond of new technologies, and enjoys good conspiracy books and movies.

Leave a Reply

Your email address will not be published.