By Kirill Bensonoff

Amazon Web Services (AWS) offers a wide array of features, allowing customers to customize the service to fit their needs. The downside of this flexibility is that the bill can quickly expand faster than initially intended. When this happens, customers frequently look for ways to reduce their AWS cost by reducing unnecessary options that typically go unnoticed. As an MSP, this gives you an opportunity to help them reduce costs.

There are three pricing options with AWS you need to be aware of to make the most of their service:

– On demand: In AWS’ most popular model, the customer is charged by actual usage of the service. On demand sometimes seems like a great deal initially, but then gets out of control as the customer’s business grows.
– Reserved instances (RI): In this model, the customer can optimize their cost of cloud by reserving instances. RI can sometimes offer price reductions of 30 to 50 percent.
– Spot instances: Using spot instances, the customer can negotiate their price for short-term usage. This is typically used for non-mission-critical applications.

All of these plans require proper planning to make the most of your customer’s budget.

Here are 4 best practices for reducing your customer’s AWS cost (and becoming their hero in the process):

1. Clean Up Orphaned Snapshots

Once an EC2 instance has been terminated, the EBS volumes that are attached to the instance are automatically deleted. However, most people forget to clean up their orphaned snapshots. These snapshots no longer need to be backed up, take up valuable resources if not deleted, and add unnecessary costs.

2. Right-size Their Instances

Most customers have an average CPU utilization of 30 percent for their AWS instances. To right-size your customer’s instances, you need to:

3. Reduce Their Data Transfer Cost

While transferring data into EC2 may be free, transferring data out is not. Transferring the data out of EC2 can quickly become expensive and be a large portion of a customer’s AWS bill.

In some cases, they may have to pay to transfer out to additional AWS services or to move their data to various AWS regions.

For re-hosted applications that are not configured to use AWS features, they may need to be re-architected (another opportunity for an MSP with the right skills) to make sure that data transfer uses the least expensive route.

Hybrid cloud can also be a concern because it communicates with the on-premise systems and the data charges can quickly add up.

4. Utilize Reserved Instances

If the customer knows they’re going to be using AWS servers for 1 or 3 years, talk with them about using reserved instances (RI). With RI, you can find discounts of 30 to 60 percent if the customer is willing to make an upfront commitment.

RI provides a reduced per hour usage cost which can help cut expenses if they know they’re going to be committed to one or three years.

These instances can be identified with tools that monitor capacity to determine if RI is the right choice. You can also repurpose reserved instances in AWS without having a penalty if they’re not using them as part of an initial instance configuration. If it turns out that they have RIs which are not being used it’s also possible to sell them using the AWS marketplace.

Are you ready to save your customer’s money with AWS in 2017?

For additional tips – like how channel partners can increase revenue with IaaS clouds, download Kirill Bensonoff’s new ebook or join him for an exclusive online seminar on January 25 – register here to reserve your seat. 

About the Author:

Kirill Bensonoff is a seasoned entrepreneur and the founder of Unigma, a unified cloud management platform. Unigma has been featured in a number of publications, and Kirill blogs regularly about cloud, tech and growing your managed services business. He can be reached at [email protected].