Annotated Bibliography & Problem Statement

Agile_by_Design_An_Implementation_Guide_to_Analyti..._----_Chapter_4_Project_Justification_and_Prioritization.pdf

Home >Business & Finance homework help >Operations Management homework help >Annotated Bibliography & Problem Statement

Chapter 4 Project Justification and Prioritization In data-intensive businesses, information needs across the organization are high. Teams supporting analytic initiatives often struggle to achieve strategic analytic goals while managing day-to-day tactical requests. A formal demand management mechanism helps balance tactical and strategic delivery. An important component of prioritizing the analytic initiative lies in understanding its value to the organization. While some companies may set aside budgets for innovation projects, most organizations need to balance innovation with other work at hand. Value- and financial-based prioritization methods help teams quantify and qualify that value.

Organizational Value of Analytics Most analytic teams provide a “free” service to their organization, making it difficult to quantify the value of the team and to effectively prioritize the analytic requests that come into the team. Analytic teams have long resisted the business chargeback models employed by IT because they believe that it gives them agility and flexibility in supporting their business partners. The upside is that this is true; the downside is that this can result in chaos—since there is no effective cost to a project or ways to identify trade-offs in completing one project over another, the team lacks any ability to prioritize their work and ensure that their resources are deployed on high value projects.

An inability to say no to business customers also results in resources performing tasks that may not be well aligned to their skillset. For example, let's say one of your data scientists completed a project for a business unit. That business partner was really happy, and now calls that data scientist for every informational request that they have. The data scientist finds herself running reports and pulling data to support the requests, but now doesn't have the time to take on any analytic project work. She gets frustrated because she's not doing the work she's trained to do, and then she quits. It happens all the time.

Another challenge in this situation is that it is difficult to institutionalize any of the analytic output. The process that each analyst uses in the creation of his or her analysis is unique. The resources become such specialists that they end up being bottlenecks: No one else on the team knows the process or system for getting work done.

In a worst-case scenario, a business customer may make the same request of multiple people on the team to see who can get the work done the fastest. The team members aren't talking to each other, so no one knows that they're all doing the same work. And yes, this happens all the time, too! Not only is this a terrible waste of resources and talent but also team members are likely to come up with different answers to the problem, requiring additional time spent justifying the results.

The good news is that you can overcome these challenges, but it does require strong leadership Alt-Simmons, Rachel. Agile by Design : An Implementation Guide to Analytic Lifecycle Management, John Wiley & Sons, Incorporated, 2015. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/harrisburg-ebooks/detail.action?docID=4041094. Created from harrisburg-ebooks on 2020-11-16 07:31:39.

C o p yr

ig h t ©

2 0 1 5 . Jo

h n W

ile y

& S

o n s,

I n co

rp o ra

te d . A

ll ri g h ts

r e se

rv e d .

(and retraining your business customers!).

Analytic Demand Management Strategy In fact, one organization exactly faced the challenges outlined above. As part of an organizational transformation, it completely overhauled its business intelligence and analytics (BI&A) delivery model.

The first change made was to shift alignment of the BI&A team. The team formerly reported to the finance organization, but had responsibility for supporting the entire division. Not surprisingly, finance-related analytic projects were always a priority. In the new model, the BI&A team reported to the divisional president; this gave them an equal seat at the table with the other functional areas.

The second change was the team's funding model. The other functional areas would pay for the team's budget. The budget included money for resources and a pool of funding for strategic projects. Recognizing the need to “keep the lights on” for tactical day-to-day requests, the leadership team agreed that 40 percent of the overall budget would be allocated to tactical requests and 60 percent to strategic projects. In addition to full-time staff, the budget supported contract and offshore delivery resources.

In turn, the business would get to participate in the project prioritization process (strategic and tactical), and receive a full accounting of how the overall budget was being allocated. To support the prioritization process, the organization established a two-tiered governance system with a triage process for requests and enhancements, as shown in Figure 4.1.

Executive Steering Committee The steering committee comprises senior-level management with representation across each functional area of the business. The steering committee is accountable for determining the overall funding of the BI&A organization, and prioritization and oversight of the strategic BI&A portfolio of work. The steering committee meets on a monthly basis.

Working Committee Mid-level managers make up the working committee, each responsible for representing their functional area's needs. The working committee delegate serves as a gatekeeper for their business unit. They are accountable for the requests from their area and submit them as appropriate to the BI&A team.

Triage Process The BI&A team then triages the request: If something can be completed within a certain time frame (i.e., less than six hours), the request is completed. If it will take longer, it goes to the working committee for prioritization. If there is a conflict over prioritization, it can be escalated to the steering committee for resolution. If the work effort exceeded a certain threshold, the request had to be prioritized by the working committee. The request owner would present a business case, and the committee would vote on where it fit in the prioritization matrix. Any request that had not yet been started could be reprioritized.

Alt-Simmons, Rachel. Agile by Design : An Implementation Guide to Analytic Lifecycle Management, John Wiley & Sons, Incorporated, 2015. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/harrisburg-ebooks/detail.action?docID=4041094. Created from harrisburg-ebooks on 2020-11-16 07:31:39.

C o p yr

ig h t ©

2 0 1 5 . Jo

h n W

ile y

& S

o n s,

I n co

rp o ra

te d . A

ll ri g h ts

r e se

rv e d .

Figure 4.1 Business as Partners

As you can imagine, an engagement model like this requires a lot of top-down support and discipline to be successful. One of the agreements made by the leadership team is that all analytic resources would be concentrated within the BI&A team. This strategy was intended to keep the business units from creating their own shadow analytic teams. However, as organizations grow in size, a centralized analytic team may become a bottleneck. In this case, it might make more sense to decentralize the resources, but coordinate them through a “center of excellence” type model where standards are defined and governed, but execution happens locally (see Figure 4.2).

C o p yr

ig h t ©

2 0 1 5 . Jo

h n W

ile y

& S

o n s,

I n co

rp o ra

te d . A

ll ri g h ts

r e se

rv e d .

Figure 4.2 BI&A Demand Management Model

Results Because the BI&A team had executive support, the new governance model worked extremely well. The functional teams were able to get work done in a democratic way. The prioritization process helped the entire group take a more organizational-centric view, by aligning their requests with enterprise and divisional business strategies. The team's capacity to work on larger strategic projects increased dramatically. In fact, the earliest strategic projects focused on creating self-service capabilities for the functional areas, so they could begin to answer their own business questions, freeing up even more capacity.

The hardest part about implementing a model like this is the business change management. The BI&A team needed to create a governance and request intake model and retrain their own staff as well as their business customers. Much of the resistance came from the BI&A staff, who prided themselves on being the “go-to” people for certain business areas.

In addition to their new governance and demand management processes, the team created more formal processes for the delivery of their analyses that allowed them to be more effective business partners (see Figure 4.3). The important lesson learned is that any new process discipline requires training, education, and the ability to adjust as business conditions change.

C o p yr

ig h t ©

2 0 1 5 . Jo

h n W

ile y

& S

o n s,

I n co

rp o ra

te d . A

ll ri g h ts

r e se

rv e d .

Figure 4.3 BI&A Implement New Processes

If the analytic team wants to define boundaries around the scope of their support within the organization, consider establishing an analytic portfolio. The objective of the portfolio is to support discrete business priorities across the company, avoiding siloed solutions. It consolidates different long-term business objectives and actions needing a common set of data or analysis techniques, and helps prioritize investments the team needs to make to achieve specific goals.

Regardless of the size of the analytic team or organization, a reporting mechanism to internal customers assists everyone in understanding and communicating the work that's being done and who it's being done for and determining the work effort's importance to the organization. This will assist the team in finding commonality across requests (which can lead to more strategic self-service information delivery projects) or in uncovering unnecessary information or analytic requests that don't align to business goals.

Project Prioritization Criteria Organizations use a number of factors in determining whether to move forward with a project, especially if the project requires a hard-dollar investment. Our earlier example shows us that many organizations value their soft-dollar investments as well (but just remember that there's no such thing as a free resource!). Organizations calculate the financial benefits of a project before moving forward, but financial benefits are not the only mechanism for determining whether a project is worth doing. What's different about taking an agile-based approach to prioritization is that you don't have to know everything up front. Agile analytic projects are full

C o p yr

ig h t ©

2 0 1 5 . Jo

h n W

ile y

& S

o n s,

I n co

rp o ra

te d . A

ll ri g h ts

r e se

rv e d .

of unknowns. Figure 4.4 highlights many of the considerations that organizations make when determining whether to move forward with a project. What makes analytic projects unique is that you often don't know the value of the output upfront: You might find that the results aren't relevant or usable, but you also just might find something that changes your business!

Figure 4.4 Project Justification Considerations

Value-Based Prioritization Value-based prioritization is one mechanism used to help companies prioritize and select projects by looking at the ways that the project will bring value to the organization or customers. Frequently, a project prioritization matrix is completed to assist the organization with project selection. Selection criteria are created and weighted based on importance.

Value-based prioritization is often seen as part of standard project portfolio management practices. After the project has been classified, project team members complete a high-level assessment of the effort and cost required to implement the project. If the organization doesn't have experience with a particular type of project, they may seek estimates from third party vendors or consultants or just give it their best guess.

Analytic teams benefit a lot from value-based prioritization since with an analysis project, you can never be exactly sure what the output of your project will be. You may not uncover anything useful, but on the other hand, you might find something groundbreaking.

More and more organizations are carving out budgets for innovation projects. Many companies are building out “innovation labs” to jumpstart breakthrough ideas. Analytics are absolutely a part of innovation capabilities, and many companies fund analytic ideas with no more than a leap of faith. Still, you want to ensure that there is a mechanism for validating and prioritizing innovation ideas. Innovation must be aligned to strategic business objectives. Most innovation

C o p yr

ig h t ©

2 0 1 5 . Jo

h n W

ile y

& S

o n s,

I n co

rp o ra

te d . A

ll ri g h ts

r e se

rv e d .

teams don't take an idea past a prototype; there still needs to be a formal mechanism for taking that prototype and making it operational.

Financial-Based Prioritization If your analytic project is coupled to a larger strategic initiative and requires investments (resources, data, technology), you may want to take a more traditional financial-based approach to prioritization.

Since there are direct and indirect costs associated with implementing projects, an additional step in the prioritization process is to determine the financial impact. The process of making long-term financial investment decisions is called capital budgeting. Since there may be a significant financial investment through the duration of the project, the project needs to be financially justified. Many organizations have a formal cost-benefit analysis (CBA) process that considers the time value of money. The CBA includes several key metrics that assist the organization in project evaluation and selection. Typical financial metrics include the following:

Payback Period The payback period measures the length of time required to recoup the amount of the initial investment. It is calculated by dividing the initial investment by the estimated annual cash inflows through increased revenues or cost savings/avoidance. Often, during project selection, cash inflows are not known and may need to be estimated. The rule of thumb is that projects with a shorter payback period are preferable. Payback period is typically expressed in years.

Present Value and Net Present Value Present value (PV) calculates the current worth of a future sum of money or cash flow stream given a specified rate of return. Future cash flows are discounted at the discount rate (or minimum required rate of return for an organization). Net present value (NPV) represents the sum of the present values of individual cash flows over a period of time minus the initial investment cost. Each cash inflow or outflow is discounted back to its present value. If NPV is positive, then the organization should consider the investment.

The PV and NPV calculations provide ways to estimate the time value of money. The premise behind the time value of money is that money today has a different value today than in the future. The difference exists due to inflation and the opportunity to earn interest.

Internal Rate of ReturnThe internal rate of return (IRR) represents the rate of growth that the project is anticipated to generate. IRR measures the yield of an investment and is used to compare the expected profitability of a project. IRR is helpful when calculating the project profitability relative to other investment opportunities.

Return on InvestmentROI metrics are used to help:

Investors understand/evaluate the value of their investment and compare it to other investments;

Define the overall profit of investment as a percentage of the amount invested; and Alt-Simmons, Rachel. Agile by Design : An Implementation Guide to Analytic Lifecycle Management, John Wiley & Sons, Incorporated, 2015. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/harrisburg-ebooks/detail.action?docID=4041094. Created from harrisburg-ebooks on 2020-11-16 07:31:39.

C o p yr

ig h t ©

2 0 1 5 . Jo

h n W

ile y

& S

o n s,

I n co

rp o ra

te d . A

ll ri g h ts

r e se

rv e d .

Understand if the benefits of project are worth the cost.

The larger the project and the more dependent workstreams there are, the more formal the CBA process generally becomes. For example, if the expected benefits of an analytic project are necessary to fund a large project, the assumptions will be carefully documented and used as inputs into the CBA.

Knowledge Acquisition Spikes The challenge with new analytic projects is that there are many unknowns, making it difficult to estimate the duration or effort required to complete a project. The biggest unknown for many organizations centers on the data: If it's not an area that's been tackled before, the data will be in likely disparate systems and of dubious quality.

ABP's analytic executive steering committee only met formally once a month; they came together for a special session to discuss the retention initiative. The data sciences team outlined their current work-in-progress and recommended some trade-offs and de- prioritization of existing work in order to meet the needs of the new project. The steering committee gives Rebecca permission to lead a two-week data spike.

In an agile project, sometimes you need to allocate time toward knowledge acquisition. In this instance, the steering committee wants to better understand the amount of data work that is involved to determine this project's potential overall impact on other projects. While they acknowledge that the retention project is important, there are several other high-impact analytic projects underway that could be impacted by the reallocation of resources. The data spike is a timeboxed event allowing the team to explore the availability and quality of data needed to answer the business question.

Rebecca goes back to Sherry with the good news. “We've got approval to do a timeboxed assessment of your data. This step is important in helping the data sciences team better understand your data, where the gaps are, and the amount of effort required to pull it all together for the analysis. Once we complete the assessment, we'll have the information we need to move forward.”

Summary Nothing worthwhile is ever free—and your analytic team shouldn't be, either. Even if you don't have a formal chargeback mechanism in your organization, find a way to value the team's effort so that you can make effective trade-offs in valuing and prioritizing analytic work. Demand management helps teams remove bottlenecks and improve the flow of work, but governance still remains an important mechanism for determining which work should be done. In

C o p yr

ig h t ©

2 0 1 5 . Jo

h n W

ile y

& S

o n s,

I n co

rp o ra

te d . A

ll ri g h ts

r e se

rv e d .

prioritizing work, organizations use a number of methods, including value-based and innovation projects—where the financial value of the outcome cannot be quantified, but the organization believes the project is the right thing to do; and more formal financial-based prioritization methods, such as a CBA. Since there are so many unknowns in analytic projects, consider a knowledge acquisition spike to uncover more information before moving forward.

C o p yr

ig h t ©

2 0 1 5 . Jo

h n W

ile y

& S

o n s,

I n co

rp o ra

te d . A

ll ri g h ts

r e se

rv e d .