《IIA Research:2023年分析和数据科学团队搭建指引-Part 2(英文版)(15页).pdf》由会员分享,可在线阅读,更多相关《IIA Research:2023年分析和数据科学团队搭建指引-Part 2(英文版)(15页).pdf(15页珍藏版)》请在三个皮匠报告上搜索。
1、Copyright 2023 IIA All Rights ReservedOrganizing Analytics and Data Science Organizations Pt.23.6V1Copyright 2023 IIA All Rights ReservedWhats in this eBookThis eBook builds upon the first Organizing Analytics and Data Science Organizations,providing further guidance in some areas and updating previ
2、ous guidance in others.In the following slides,we reconcile the following changes in the analytics field since we first examined org models:1.Big data has gone from novel to norm2.Data science talent is in higher demand than ever3.More visual and analytical tools are on peoples desktops4.Artificial
3、intelligence has taken the world by storm5.Analytics of all types are now often processed in the cloud6.Software applications employing machine learning and artificial intelligence continue to multiply7.Real-time analytics and automated decisions embedded in business processes are ubiquitousIn Part
4、2 of this series,well describe and offer guidance on:1.Various coordination mechanisms that will enhance organizational2.effectiveness3.Design variables to account for while constructing the organizations4.structure5.The need to support the incubation of new corporate capabilities6.Participation in
5、cross-functional special teams or COEs pursuing large,strategic projects7.The various component teams within the analytics and data science organization8.How specialization occurs as the organization grows2Copyright 2023 IIA All Rights ReservedCoordination MechanismsWhatever the model is in place,th
6、ere will be a need to coordinate across both analytics teams and business units.Even when all analytics and data science resources work in one centralized corporate unit,the customers for their services are spread across the enterprise,and you need coordination mechanisms to manage and meet the inte
7、rnal customers demand for analytics.There are several common coordination mechanisms that can supplement a formal reporting structure by enabling groups to plan and work together,and by developing an enterprise view of priorities and resources.The ones we will cover are:Analytics Governance CouncilA
8、nalytics TranslatorsAnalytics Community of PracticeMatrix ReportingStaff Rotation3Copyright 2023 IIA All Rights ReservedCoordination MechanismsCont.Analytics Governance CouncilSet the direction for analytics,priorities for major projects,oversee progress in building analytical capabilities,and drive
9、 commitment and collaboration among business units.The stronger a councils mandate is and the more formal authority it has,the better.In short,its purpose is to:provide oversight,drive coordination,and set prioritiesAnalytics TranslatorsExperienced,versatile,and respected staff members who work with
10、 business unit leaders to advise and educate them on opportunities to utilize analytics and data science and how to make their units effective consumers of data and analytics services.Theirpurpose is to bridge the gap between the business and analytics teams.Analytics Community of PracticeCOPs are g
11、ood for engagement and retention,as well as knowledge and idea sharing.COPs may host events,internal sharing sessions,and training.They serve as an informal channel for staff development and organizational alignment.Its purpose is to foster learning,enable sharing,and encourage engagement.4Copyright
12、 2023 IIA All Rights ReservedCoordination MechanismsCont.Matrix ReportingDistributed analytics and data science teams often report both to theirassociated business units and to a centralized analytics unit,with one line solid and the other dotted.There is no universal right answer as to which line i
13、s dotted and which is straight.Corporate culture and precedent are often the biggest determinants.Its purposeis to ensure accountability and to formalize the commitment to collaboration.Staff RotationStaff rotation is a long-term play.Any organizational structure benefits from the movement and cross
14、-pollination of people and expertise.A rotation program can help staff gain experience with multiple areas of the business while bringing each unit different ways of thinking from other parts of the organization.A formal staff rotation program will have defined mechanisms that allow staff to rotate
15、between business units,the central analytics team,and centers of excellence(COEs).Its purpose is to help disseminate knowledge and drive the convergence of standards.5Copyright 2023 IIA All Rights ReservedDesign VariablesAny organizational design for the data science and analytics function must work
16、 in the context of how the business already operates.To evaluate,design,implement,and refine organizational structures,it is necessary to also consider some additional variables that must work in harmony for any organizational model to succeed.Well cover:Reporting StructurePhysical LocationFunding S
17、ourcesInfrastructure6Copyright 2023 IIA All Rights ReservedDesign Variables Cont.Reporting StructureReporting structure is comprised of the formal lines of authority and administration.As previously discussed,reporting lines are often matrixed to reflect the necessity of balancing the needs of multi
18、ple stakeholders.One of the most common questions relates to where the centralized portion of the analytics and data science organization should report.Physical LocationPhysical location means less than it used to.To be effective,though,it is important for analytical teams to find ways to get togeth
19、er to build and sustain strong relationships.It is equally important to build and sustain relationships with teams business partners.This will require a dedicated focus on creating and prioritizing opportunities for teams to get together for planning exercises,team building,and other important activ
20、ities.7Copyright 2023 IIA All Rights ReservedDesign Variables Cont.Funding SourcesFunding sources are not often considered in the context of organizational design,even though paralysis is guaranteed if organizational structure and funding sources are at odds.One successful model we have seen is a co
21、st reallocation agreement with the CFO.Essentially,the BI and report automation team automates reports that were done in a manual fashion,and thus removes burden(and staff needs)from the operational and ad hoc teams.InfrastructureIt is critical to try and standardize technology and policies across t
22、he company so that there are minimal barriers that the analytics and data science organization must deal with as it works with different parts of the company.One way to drive analytics usage and maturity across the enterprise is an up-front corporate investment in data and technology to create a pla
23、tform significantly superior to what anyone else has in place.At a minimum,effort must be made to enable localized business unit systems to communicate so that the analytics team can access and combine data from those disparate systems.8Copyright 2023 IIA All Rights ReservedCapability IncubationAs n
24、ew analytics technologies,methods,and business applications emerge,enterprises commonly form dedicated teams to consolidate expertise,experiment,build early prototypes,and then spread the capabilities by providing support services and education to the broader organization.Innovation initiatives may
25、be executed by special teams or centers of excellence,but often include a team that is broader than just analytics team members.These teams may be owned and led by another organization within the company.In other words,it is often necessary to support and participate with broader centers of excellen
26、ce where analytics and data science talent is loaned for a period of timeto a COE outside its own organization for a broader corporate benefit.Well cover the difference between the old approach(big data)and the new approach(AI and Cloud).9Copyright 2023 IIA All Rights ReservedCapability Incubation C
27、ont.Then:Big Data Often Led the WayWhen the big data explosion happened a decade ago and a broad range of new structured and unstructured data emerged,many enterprises formed special,cross-functional big data teams to experiment with big dataAs big data and its technologies went mainstream,these cap
28、abilities were eventually dispersed across the analytics,IT,and business units they served and are now just a core part of the organizations capabilities rather than a special team.Now:AI and Cloud are the FocusIIA sees a similar pattern today as enterprises form special teams or centers of excellen
29、ce to usher machine learning and artificial intelligence methods into analytics applications and to enable the corporate infrastructure to support them.A core AI/ML team or COE often works across the enterprise advising on business opportunities,building and testing models,monitoring their performan
30、ce,and sharing expertise.As with big data,eventually these activities will become a core part of what is done and the members of the special teams currently in place may roll back into their home organizations.10Copyright 2023 IIA All Rights ReservedOrganizational ComponentsEspecially as an analytic
31、s organization grows,it will be necessary to begin to break the organization into various teams with specific charters and specialties.The advice that follows is most appropriate for larger analytics organizations with at least 50-60 members.Organizations with a few dozen or less total analytics and
32、 data science employees are probably best served to keep everyone in one group with everyone pitching in to help in multiple areas.In an organization that size,there just arent enough resources to have people focused on only one area or activity.Well cover 4 of the 7 components discussed in the full
33、 research brief:The Data Management TeamThe Business Intelligence and Report Automation TeamThe Operational and Ad-Hoc Analytics TeamThe Executive Insights Team11Copyright 2023 IIA All Rights ReservedOrganizational ComponentsCont.The Data Management TeamWill differ from operational data management a
34、s it is owned by IT.First,this team must help with data governance,particularly as it relates to the data that the organization utilizes most frequently Next,is the ongoing validation of data quality to ensure that information feeding into analytical processes is clean enough.Finally,is managing any
35、 organizational-specific data management tools and platforms used for executing analytical processes.This is typically decentralized after operational/ad hoc analytics teams and often at the same time as data architecture.The Business Intelligence and Report Automation TeamBuilds,tests,and maintains
36、 reports and dashboards for the company.Through automation,ensures data transparency and optimizes performance on the analytics team.Works closely with the data management and architecture teams and owns new reporting needs.This is typically the second group to be decentralized after operational/ad
37、hoc analytics teams.12Copyright 2023 IIA All Rights ReservedOrganizational ComponentsCont.The Operational and Ad-Hoc Analytics TeamFocuses on basic analytical tasks.Functions supported include HR,Marketing,sales,risk,and operations.Serves as a pipeline for new analytics opportunities at the enterpri
38、se level.Team members are typically less technical but more business savvy-often seen as“analytics translators”for the org.They are also typically the first group to be decentralized.The Executive Insights TeamMost common in companies where analytics has matured.Typicallysmall team targeted directly
39、 at supporting senior executive requests.Highly consultative,versatile,strong communicators and presenters.13Copyright 2023 IIA All Rights ReservedBecome an IIA Client to Access the Full Brief IIAs Research&Advisory Network(RAN)clients get access to the full Organizing Analytics and Data Science Org
40、anizations Pt.2 research brief online and as a PDF which contains a complete guide to putting your companys organizational approach together.Research&Advisory Network(RAN)clients also have direct access to the experts that developed this content and framework with on-demand inquires.14Copyright 2023 IIA All Rights ReservedDont go it alone.Let IIA guide your |We work with clients to build and grow their analytics organizations.Benefit from our unbiased,unrivaled network of analytics experts,practitioners and thought leaders.3.6V1