The days of the blockbuster deal to outsource the IT department and the company\u2019s infrastructure assets are long gone.\nIT outsourcing deals are now much more targeted. Most enterprises prefer transactions that cover discrete service towers and functions on an increasingly granular basis. This is particularly prevalent in modern network infrastructure deals where enterprises increasingly acquire a complex mix of managed network services sourced from multiple providers.\u00a0\n\nWhen structuring managed network service arrangements, enterprises are selecting providers based on their capabilities and expertise in specific network technologies and functions such as legacy voice services, unified communications, WAN, wired and wireless LAN, and security. To optimize service delivery in these multi-sourced settings, enterprises vary the degree to which functions are outsourced or kept in-house. They keenly focus on retaining responsibility for important processes that are supported by strong and capable in-house resources, and then purchasing from the market the technology, expertise and resources that are either unavailable internally or sufficiently commoditized that they can be delivered cheaper and more efficiently by external suppliers.\u00a0\nThe service-provider landscape has also shifted. Carriers and large system integrators continue to feature strongly, but a range of smaller and nimbler providers specializing in individual technologies and services are growing their market share. Several network-equipment OEMs are starting to push their managed-services offerings (not least Cisco), and some SD-WAN providers are selling fully managed solutions.\nWith this increasingly complex use of managed network services, the upfront strategy, sourcing process, transaction structure and contract negotiation approach employed by enterprise buyers must evolve to meet the demands of flexible, numerous and interconnected service arrangements. Success in this new environment requires specialized and mature sourcing and contracting capabilities and processes tailored for managed network services.\u00a0\nClearly define the deal\nPrecisely defining the scope of the managed network services that you need is a crucial step. The most common cause of confusion and disputes in managed network services contracts relates to whether a particular function or activity is in or out of the provider\u2019s base scope of services. If it\u2019s in-scope, the base charges apply to the provider\u2019s performance of the function or activity, whereas if it\u2019s out-of-scope, the service provider can charge you an extra fee for the work.\nAvoiding such issues begins at the start of your sourcing process, by clearly and comprehensively defining and describing the required in-scope services in the request for proposal (RFP) document that you issue to potential providers. Spending the time to get this right will reap dividends later. The process employed to document the scope creates a shared understanding and alignment internally for the sourced services. It also helps ensure that the proposals you receive cover all the functions and scope you need and can be evaluated on a like-for-like basis to facilitate an efficient and objective down-selection.\nAs enterprises have become more familiar with integrating managed services from multiple service providers, they have also developed mature standard working practices, operating procedures and policies that they expect to apply to all of their major service providers. The practices, procedures and policies cover areas such as change management, incident resolution, asset management, integration with and use of customer tools, reporting and invoicing requirements, personal data processing safeguards, contract compliance management and minimum security controls. Imposing the same processes and procedures on major suppliers makes it easier for companies to integrate a diverse mix of managed service providers into a holistic network infrastructure operation.\u00a0\nNot all service providers are readily willing to conform their processes to the customers\u2019 policies and procedures. Instead, they prefer to apply their own approaches, tools, procedures and conventions, which may not align well with those of their customers. Where service providers have invested in training their personnel on specific processes and tools and are proposing to deliver managed network services via shared network operations center (NOC) resources, they will naturally resist customizing their processes for you.\nBut seemingly minor differences between a service provider\u2019s process and the equivalent customer process can sometimes cause unforeseen problems and friction between the parties during service delivery. Accordingly, you should provide the processes and procedures with which you expect service providers to comply at the start of the sourcing and service provider selection process, not after you start negotiating contracts. (For example, include a copy of your change-management policy and procedures in the RFP document.) That way, you can pick service providers that best fit your processes and, as a result, can integrate more seamlessly with any other managed network services providers in your environment.\nDetermining which service providers best meet your needs and priorities involves a thorough comparative analysis of each prospective provider, including the niche service providers that only bid on a subset of the managed network services covered by the RFP, such as supporting only a particular network technology or geographic scope.\u00a0 To help define the optimal service-delivery arrangements from the available proposals and meet any regional requirements, enterprise customers should fully scrutinize and evaluate each bidder proposal independently, and also assess certain combinations of managed network services from multiple suppliers on a best-of-breed basis. To do that properly, you must solicit proposals via procurement documents that structure the service scope into discrete standalone lots or modules, which can be evaluated both in isolation and in combination with other lots or modules.\nKey solution and service-provider differentiators\nWith the proliferation of managed-network-services providers, the differences between them and their solutions has grown exponentially. The relative importance of these differences can vary dramatically depending on the deal context, the state of the customer\u2019s existing network environment and the customer\u2019s technology roadmap. For each potential deal, identifying, understanding and assessing the relevance of these differences is an essential part of service-delivery design and vendor selection. The following are examples of some key considerations:\n\nOff-shoring versus on-shoring \u2013 Most managed network services providers have built off-shore organizations to cost-effectively deliver managed services. Off-shoring typically underpins the outsourcing business case, but customers often prefer certain key functions to be staffed on-shore to facilitate better communication, collaboration and levels of support. Understanding the mix of on-shoring\/off-shoring proposed by potential service providers is, therefore, important, including what resources will be local to your regional headquarters around the world. \u00a0\nNOC model \u2013 Service providers have varying approaches to providing NOC support \u2013 from a shared NOC model where the NOC resources support multiple customers, to a completely dedicated model where a specific complement of NOC staff is dedicated to an individual customer, along with various hybrid models. There are pros and cons for each approach, and each approach will work better or worse for different customers. What model is being proposed by potential service providers, and how it will meet your needs in practice (e.g. how resources in a shared NOC model will have familiarity with your network infrastructure; what additional resources a dedicated NOC model can draw on in times of heightened activity or for particularly complex issues) should be closely examined.\nResource model \u2013 When providers respond to an RFP for managed network services, they develop a resource model to deliver the scope of services that you have defined. The number and mix of the resources in that model (including the on-shoring\/off-shoring mix) has a fundamental effect on the cost of the proposals, and on the degree to which a proposal will satisfy your requirements. The service providers will inevitably interpret and understand the required scope of work set out in your RFP differently, and thus potentially develop quite distinct resource models. Reviewing and understanding the proposed resource models and helping potential service providers make sure that their resource models are not under- or over-sized, is a necessary step to make sure that all the proposals you are considering can be fairly compared. Selecting the proposal that is cheapest but has an undersized resource model will result in severe problems later.\nTooling \u2013 The tools that each potential service provider will deploy to support your environment can be a key differentiator, as is the level of access to the tools you\u2019ll get from the supplier. Understanding how the supplier tools will interface and integrate with the enterprise\u2019s tools should also not be overlooked during the proposal review period. E-bonding between service provider and customer tools is common, but some enterprises find it complex and time-consuming to implement. Manual \u201cswivel chair\u201d updates between customer and supplier tools (e.g. for incident management) aren\u2019t that uncommon but are costly and rarely real-time. And some customers would prefer that the service-provider resources work directly in the customer\u2019s tools to avoid any need for e-bonding or swivel chair, but some service providers will resist that, particularly where a shared NOC model is being proposed.\nLanguage support \u2013 Depending on the customer\u2019s geographic footprint and the scope of services required, local language support can be a determining. For instance, where the managed service provider will receive and potentially update tickets in the local language, or where the provider will engage on your behalf with other suppliers around the world (e.g. raising and managing incidents with local in-country network-transport providers that may not operate in English).\u00a0 Multi-lingual support can be difficult for service providers to provide, and language translation services are not always satisfactory, particularly in critical incident-resolution situations.\u00a0\nService Levels \u2013 Service levels are crucial for managed network services because they help define the level of required support. If, for example, the customer requires service requests to be fulfilled more or less quickly, that will affect the number of resources a service provider will have to deploy to meet the required service-request fulfillment time. You should therefore present your service-level requirements in the RFP. The service levels also directly measure the performance and quality of the resources delivering the managed services. \u00a0A service provider that offers real consequences for service-level misses (e.g., meaningful service credit amounts that increase with the severity of the miss, executive escalations and rights to terminate for cause when the misses are severe or chronic) will be far more likely to take appropriate steps to address performance issues should they arise (e.g., by replacing poor-performing resources or simply adding additional resources).\u00a0\n\nThe range of issues and considerations to account for and contemplate when sourcing and selecting managed network services and their providers (the examples above are just a few) makes the evaluation and selection process a significant undertaking. You should give yourself the time to work collaboratively with potential service providers to make sure their proposals are fully understood, and then help them improve the alignment of their proposals with your requirements. An arm\u2019s length review of proposals, with limited interaction with the bidding service providers, can work for commodity services but does not work when sourcing managed network services.\nManaged network services will continue to evolve, and customers will continue to become increasingly sophisticated in their selection and use of managed services and managed services providers. Sourcing strategies, RFP processes and contracts must adapt to keep up. In our companion article, \u201cMaking network-services deals: Structuring the transaction,\u201d we explore the key commercial terms and contract practices that are crucial for negotiating flexible and resilient managed network services arrangements, and navigating succesfully the outsourcing industry challenges currently facing enteprises.\nBen Fox is a managing director at TC2, a global IT and networking consultancy. He can be reached at email@example.com.\nMarc Lindsey is a partner at Levine, Blaszak, Block & Boothby, LLP (\u201cLB3\u201d), an IT and telecommunications law firm dedicated to advising enterprises clients on technology transactions and telecom regulations. He can be reached at firstname.lastname@example.org.