You’ve just finished the migration process to Office 365 – but that doesn’t mean that you’re done. Now that your productivity suite is in the cloud, there are several optimizations that you should consider to ensure quality of experience for your users.

YOU MIGHT ALSO LIKE: 5 Secrets to Make Microsoft Business Apps Faster

As you already know, moving services to the cloud provides enhanced productivity and management simplicity, while allowing services to be accessible anywhere. Nevertheless, user quality of experience can be substantially compromised if several network optimizations are not taken after migration.

Here are a a few tips for flawless Office 365 performance.

1. TCP Settings

With traffic flowing through the Internet, TCP window size becomes an important optimization opportunity. Default TCP application servers can only send 64KB of traffic before the client acknowledges them. This variable can be increased to a value closer to the real bandwidth delay product (BDP) in order to take advantage of the link characteristics.

BDP = (Bandwidth [bits/sec]) x (Round Trip Time (RTT) [secs])

Use the RTT from the branch office to/from the O365 servers. For example, if your RTT is 150ms and your bandwidth is 10Mb/s, then your BDP is calculated as follows:

BDP = 10Mb/s x 100ms = 125KB

You would increase the default TCP window size to 125KB, which is about twice the size of the default.

2. Perimeter Devices

Another TCP consideration would be to configure perimeter devices (e.g: firewalls and proxies) with a TCP idle timeout high enough that it does not interfere or close already established connections to the Office 365 servers. Taking care of this will prevent unexpected disconnections and unnecessary repeated login requests on outlook clients.

It is recommended to set this value to higher than 2 hours.

NOTE: For those who have proxies that are authenticating connections, O365 generates a great amount of TCP sessions that, when authentication is required, introduce considerable delay. It is recommended to set the proxies to allow connections to servers like,,,,,, and any other site that may be used in Office 365 communications.

3. DNS

As some of you may already expect, fast DNS resolution is essential for a proper cloud deployment and O365 is not an exception. Ensure that the DNS Geo location is within the user base to ensure that users are connecting to the closest server and avoid long RTT’s. Also, keep track of how long it takes to resolve domains such as

4. Exinda Solution

Beyond general TCP and connectivity improvements, Exinda users have the ability to check for Office 365 performance using our “Solution Centre” feature in the Exinda solution.


The ‘Office 365 Performance’ option activates specific Office 365 reporting for performance (Internal and External hosts) and Application Performance metrics for a more granular understanding of host experience and application performance.

The Application Performance metrics, help users understand whether the problem is in the network or in the Office 365 application (refer to Transaction Delays) and also keeps track of network metrics such as RTT, Jitter and Packet Loss. The reporting option follows the behaviour of the application and its metrics over a certain period of time in order to set a baseline that defines what the “normal” behaviour of this application feels like. After the baseline is set, new Office 365 traffic is compared against it in order to grade Office 365 with a score from 1 to 10 that measures the quality of experience. Users can then configure email alerts for when the score is lower than a specific value. Scores higher than 8 usually mean the application is running as expected. The Exinda Solution also applies colors to the metrics so the user can gain a better understanding the problem at hand.

But Wait! There’s More…

Check out the video demo below and see how you can solve your Office 365 performance woes. Or contact an Exinda Solution Expert for more info.