by Joe LoLordo –
Managed Services covers a broad spectrum of IT services and support that can be provided by IT Consulting firms and other IT Services/Software providers (such as application service providers (ASPs), Web hosting companies, network service providers (NSPs), etc.) — all of whom may be looking to supplement their traditional offerings with management services.
An example of Managed Services is when a company transfers all or part of the day-to-day management responsibility for a specific IT function to an outside vendor that specializes in that function and can provide it at a lower cost. In some cases, this may include ownership of the hardware and software needed to provide that function. Some examples of these IT functions include:
- Application Development and Maintenance
- Quality Assurance and Testing
- Infrastructure Management: Set-up, Configuration, Maintenance, Monitoring, and Incident resolution
- Network and Communications Environment
Companies enter into managed services arrangements primarily to enable their IT organization to focus on more business-critical issues, strategic functions, and activities, which, in turn, results in a more cost-effective use of internal IT resources, improved efficiencies, and reduced costs.
Most of the examples below pertain to one type of managed services arrangement: covering the set-up, configuration, maintenance, monitoring, incident resolution and support activities for a company’s Hadoop Data Lake environment. Many of the concepts discussed here — process automation, team structure, onsite/offshore optimization, experienced core team leaders, automated metrics tracking and dashboard reporting, competitive pricing approaches — can also be applied to most managed services engagements.
The Goal: A Win-Win Scenario
The “Client” (a company seeking to contract for the managed services) and the “Vendor” (the firm that will provide the managed services) must both have a win-win mindset going into discussions for establishing an arrangement. A win-win mindset means that both sides must be willing to compromise during negotiations and the discussion of the approach so that both sides can be successful.
For example, if the client is simply looking for the lowest-cost solution or an arrangement with the highest fines for missed SLAs, then the client most likely will be disappointed in the performance of the vendor’s team; resources will be too inexperienced to perform the work, and the vendor will be unable to absorb all the fines when SLA’s are missed. The vendor must also be willing to invest in the arrangement, ensuring that the proper mix of team leaders and experienced resources are on the team. The vendor should also invest in continuous training to enhance the team’s ability to succeed. It’s important that price, experience, and automation are balanced to enable both parties to achieve a successful arrangement.
Keys to Success
Following are several key topics and examples that support and illustrate successful managed services arrangements or engagements. Some of the examples are specific to a managed services environment that provides the maintenance and support activities for a company’s Hadoop Data Lake environment but are broadly applicable to many types of Managed Services.
- Automation and Accelerators – Establishing automation, accelerators, and continuous operational process improvements are critical capabilities that a vendor must demonstrate to ensure a successful long-term managed services engagement. For example, a vendor that is contracted to provide overall maintenance and support for an infrastructure environment like a Hadoop Data Lake environment should be able to explain how they can automate Level 1 (L1) Support: monitoring, incident logging, automated alerts, and common incident resolution.Over time, the more automation that is introduced to potentially handle Level 2 (L2) and maybe even some Level 3 (L3) incident resolution will result in fewer team hours required, on average, per resolved incident. Of course, the degree of possible automation may vary based on the complexity of the client environment and the willingness of the client to accept the use of automation. The ultimate benefits include higher user satisfaction, increased system uptime, and reduced overall environment management costs. For Cloud-based environment solution providers, automation should be part of their standard environment monitoring, maintenance and support offering which they should be able to provide at a more competitive pricing structure. Using Advanced Test Automation is another example of accelerators that are used in a QA and Testing managed services engagement.
- Establish Standard Operating Procedures and Communication Processes – When a client company enters into a managed services agreement with a vendor there should be a defined Transition Plan that could span a 3- to 6-month period, based on the scope of the managed services agreement and the complexity of the environment. The vendor needs time to get up-to-speed with understanding the overall environment, and for working with existing client IT personnel to take over work responsibilities. During this transition period, it is imperative that the vendor establishes all operating procedures and communication processes, and begin to introduce automation.For example, in managing a Hadoop Data Lake environment, the operating procedures should be based on ITIL industry standards. System monitoring and ITSM incident logging and tracking software should be in place. The process for handling different severity levels (L1 thru L4) should be clearly defined, including procedures and protocols for escalations and communications with other IT organizations. Response and resolution times for SLAs are defined while metrics tracking and reporting procedures are put in place. Automation for system monitoring, incident logging, alerts, and incident resolution should be introduced and established. A run book should be created, and repeatable resolution techniques for common incidents established. For upgrade and maintenance activities, repeatable automated jobs and test scenarios should be established. These are just some examples of tasks that a client company should reasonably expect their managed services vendor to perform.
- Team Structure and Experience – It is important that a client work with their vendor to ensure they provide the right level of team experience and team leaders, both onsite and offshore. During the initial transition phase, the team leaders and more experienced resources should be onsite to work with the client personnel in establishing the standard operating procedures, communication protocols, SLAs, and automation techniques. This requires the offshore leads to come onsite during the transition period. It is also effective to have certain offshore resources rotate onsite for selective 3-month periods throughout the duration of the managed services agreement. This will help enhance the knowledge of the offshore leaders, and also serve to improve overall communications between onsite and offshore.The offshore resources that come onsite should be the leaders and most experienced members of the offshore team. They are responsible for mentoring the remainder of the offshore team and assisting with any issues that arise. The client company must guard against the vendor overstaffing the offshore team with too many junior and inexperienced resources in order to reduce the overall price; otherwise, the client will most certainly encounter team performance issues at some point. Team leaders rotating onsite, automating processes and providing continuous training, will enable the overall team to successfully leverage more junior resources, helping to keep the price low for the client.
- SLA Driven Performance – SLAs (Service Level Agreements) are sets of metrics that are used to measure team performance and maintain high quality in the provision of services. During the Transition Phase, it is important to define mutually agreed upon SLAs between the client and vendor. These SLAs should be reasonable and based on existing achievable environment realities and customer expectations. A number of common metrics are typically used for an Infrastructure Management environment, such as a number of incidents by severity level, response time to log and respond to an incident, and resolution time based on the severity level of an incident.SLA values may need to be adjusted over the duration of the managed services engagement. But, over time, it is reasonable to expect that these metrics will improve based on automation and the increasing experience level of the team. Automated metrics tracking and reporting should be part of the standard weekly status reporting. Typically, a managed service agreement will include vendor penalties or fees that are exacted if the team fails to perform to expectations.
- Multiple Vendors Providing Managed Services – A client company should use different vendors for different types of managed services. For example, if a company has a managed services arrangement for Applications Development and Maintenance, and another for Infrastructure Management, they should use different vendors or suppliers for each. As the old saying warns, “You shouldn’t put all of your eggs in one basket” because it is too risky. A Client company may believe that they will get a better price or endure less complication by dealing with only one vendor, but the more likely result is that the client’s leverage and ability to act will be reduced if performance issues arise. A client should foster a competitive pricing environment between at least 2 or 3 managed services vendors. Ideally, you should be able to compare price, performance, experience, and encourage vendors to introduce new ideas and technologies. You want the vendors constantly competing for your business.Also, once you enter into a 3 to 5-year managed services agreement with a vendor that vendor begins to better understand your business and operations, and it becomes increasingly difficult to transition them out if they are not performing well. An alternative vendor that’s available and has some knowledge of your business — though perhaps in a different IT functional area — is easier to transition into a new managed services area if you are having a serious problem with a vendor’s performance.
- Investments in Training and Resource Planning – The vendors must be willing to invest in developing and enhancing training material on an ongoing basis, encouraging continuous training for the team, and planning for team resource attrition. Enhanced training enables better team performance, provides opportunities for advancement on the team and minimizes the negative impact on the team when a team member leaves the company.
Foundational Components for Managed Services Success
Above, I’ve highlighted certain key success criteria for managed services engagements. These specific topics and examples may vary based on the type of managed services offering, and the above is not meant to be an all-inclusive list of success criteria.
However, certain success criteria are essential to virtually all managed service agreements. Automation, defined standard operating processes, mutually agreed upon SLAs that are reasonable, and a win-win mindset will all provide a strong foundation for establishing a successful, long-term managed services agreement.