Optimizing Org Charts with ChatGPT-4: From hand-drawn to optimized
Using ChatGPT-4 to read a whiteboard’s hand-drawn org chart, this article explores 10 actual use cases leveraging generative ai’s enhanced visual , analytical, and generative capabilities, making it a powerful tool for strategic design tasks.
“OrgAI Design” augments human processes in several ways:
OrgAI Design is evolving organizational design, and tools like ChatGPT-4 are central to this transformation.
Use case 1: “OrgAI Design” Reading a hand-drawn organization chart on a conference room white board
Here’s how it works. Starting with a white board discussion of a company’s org chart, pictured here. I took this photo with a cell phone in an org design discussion.
It’s a real org chart. The lines are as messy as the handwriting, there are annotated questions, some boxes refer to individuals, some to entire functions, and the handwriting riddled with abbreviations and annotations. It could be any company, in this case a higher education institution. You could also use a screen shot of a computer-generated org chart if your starting point was a tool like OrgManager, Workday, SAP, or similar product.
Prompt 1: “Here is a hand-drawn org chart of a company structure. (And in ChatGPT-4 click the icon to upload the photo you took of the white board.)”
It’s important to verify that everything in the chart was understood by the ai. A quick way to do this is to ask chatGPT to draw the organization in text form, which it will do by creating an outline of the organization. Just ask it something like “Can you draw the org chart in a text outline format so I can make sure it is interpreted correctly?”
Use Case 2: Identifying duplication and overlap
Prompt 2: “I need to find ways to optimize this organization’s structure. Here is a hand-drawn org chart of the company. Where do you see potential duplication or overlap between functions, and how could the organization be reorganized to remove overlap ?”
Response 2: (Note: Seven total recommendations were made. For brevity, just suggestions 2-4 are shown below. No other contextual information was given to chatGPT, just a photo of the whiteboard in the room where we were talking. This demonstrates it can read the org chart, interpret the lines, deduce the industry to make contextually-relevant insights, and understand not only the functions but their current and potential interaction potential with each other if positioned differently.)
Certainly! Based on the organizational chart provided, here are areas with potential duplication or overlap and suggestions for reorganization to eliminate these redundancies…
2. Communication Roles:
- Overlap: The roles under “VP Communication” like “Design Dir.”, “Media Relations Sr. Dir.”, “Social Media Dir.”, etc., might have overlapping tasks.
- Solution: Consolidate similar roles. For instance, “Media Relations” and “Social Media” can be under a single “Communications & Media” department. This promotes cohesive branding and messaging.
3. Community & Alumni Relations:
Overlap: “Alumni Affairs” and “Community Outreach” under “VP Advance” might have some common functions, especially in institutions where alumni play a significant role in community activities.
Solution: Consider merging them into a “Community & Alumni Engagement” department. This would streamline outreach efforts and possibly foster stronger alumni-community ties.
4. Operational Functions:
- Overlap: “Plant Operations”, “Maintenance”, and “Campus Safety” under “VP Operations” may have intersecting responsibilities when it comes to campus infrastructure and safety.
- Solution: Group them under a “Campus Infrastructure & Safety” department. This ensures that all aspects of the physical campus are coordinated and overseen with a unified vision…
To remove overlap effectively, it’s essential to conduct a thorough review of each department’s functions and responsibilities, engage with department heads and staff to understand the nuances and intricacies of their roles, and prioritize the organization’s goals, ensuring that any restructuring aligns with its mission and long-term vision.
Consider that the above analysis and findings occurred in approximately 10 seconds. This is faster than a person could even read the org chart, let alone think through the question being asked. Further human observation may find additional potential redundancies, but this list is a good starting point for consideration.
Use Case 3: Identifying collaboration roadblocks in the structure
In Org Design, a common analysis factor is the structure’s potential roadblocks to collaboration. The next prompt invited generative ai to identify collaboration barriers, and then goes further asking ai to recommend mitigation strategies the team could consider.
Prompt 3: “What potential bottlenecks or barriers might there be to collaboration and communication in the current organization structure? And how could these be mitigated through org redesign or other methods?”
Response 3: (Note: Five bottlenecks were identified in the current structure, and ten recommendations were made to mitigate them. For brevity, only a subset of the 10 are included below. This demonstrates a good understanding of the underpinnings of effective collaboration in the human world. It covers topics from span of control and overload, to siloing, unclear expectations, and unclear boundaries and where those may be inherent to the structure of an organization.)
The current organization structure presents a few potential bottlenecks or barriers to collaboration and communication:
1. Multiple Direct Reports to President: The President has several direct reports, including the various VPs and advisors. This can lead to potential communication bottlenecks and difficulty in managing and overseeing all these functions directly.
2. Fragmentation of Similar Functions: For instance, there are separate roles for “Media Relations Sr. Dir.” and “Social Media Dir.” under “VP Communications”. This could lead to potential overlap in responsibilities or lack of coordinated communication efforts.
3. Lack of Clear Reporting for Some Roles: For instance, functions like “Community Education”, “Regulatory”, and “Plant Operations” do not have clear hierarchical reporting structures, which might make it unclear who is accountable for decisions and performance.
4. Potential Siloing: With many distinct departments and roles, there’s a risk of departments working in silos, where they don’t communicate or collaborate effectively with other departments.
5. Unclear Roles in Student Affairs: There are multiple roles related to student services, including “Admissions”, “Fin Aid”, “Student Life”, and “Career Services”. While they fall under the “VP Student Affairs”, the distinction of responsibilities might not be clear.
Mitigation Strategies:
1. Consolidate Reporting Structures: Reduce the number of direct reports to the President by consolidating similar functions under a single VP or director.
2. Merge Similar Functions: As suggested previously, merging roles like “Media Relations Sr. Dir.” with “Social Media Dir.” can lead to more cohesive communication strategies.
3. Clear Hierarchies: Ensure every role and department has a clear reporting structure, which will provide clarity in decision-making and accountability.
4. Cross-Functional Teams: Create cross-functional teams or committees that bring together representatives from different departments to work on shared goals or projects. This can help break down silos and foster collaboration.
…
6. Clarity in Roles and Responsibilities: Clearly define and communicate the responsibilities of each role, especially in departments like Student Affairs, to avoid overlap and confusion.
…
10. Review and Iterate: Regularly review the organizational structure and make iterative changes based on feedback and observed challenges to ensure optimal collaboration and communication.
By addressing these potential bottlenecks and barriers proactively, the organization can foster a more collaborative, efficient, and communicative environment.
The speed or response, again around 10 seconds, gives the design team the ability to focus on insights and recommended actions, rather than being mired in lengthy analysis discussions. Again, the team should circle back and see what else they may find, but the above list is impressive as a starting point.
Use case 4: Benchmarking, modeling and options generation
Another factor in org design is benchmarking. In this case, the university benchmarked below where it aspires to be. It’s committed to protecting its teammates, engaging long term planning and natural attrition to reach desired benchmark performance.
Prompt 4: “When our organization is compared to industry benchmarks, our staffing levels look to be about 10-15% high compared to the top quartile companies. We would like to get there through natural attrition over 1-2 years to eliminate impact on our team mates. What changes should we consider over that period to the org structure to be in line with the leaders in our industry?”
Response 4: (Full response is included below, as each suggestion for consideration demonstrates the application of various workforce optimization concepts.)
Certainly! Using the org chart you’ve provided as a reference, here are possible changes to achieve a 10-15% reduction:
1.Merge Overlapping Departments:
Combine “Community Outreach” and “Alumni Affairs” into one department titled “Community & Alumni Relations.” This will consolidate two roles into one, streamlining responsibilities.
2.Centralize Marketing Efforts:
Merge roles under “VP Communication” such as “Design Dir.”, “Marketing Dir.”, “Media Relations Sr. Dir.”, and “Social Media Dir.” into a single “Communications & Marketing Department.” A unified team can share resources and responsibilities, possibly eliminating one or two positions.
3.Redefine Leadership Roles:
The “Assoc V.P. Operations” can potentially be absorbed into the “VP Operations” role. This would reduce redundancy at the leadership level.
4.Streamline Student Services:
“Admissions”, “Fin Aid”, and “Career Services” could come under a singular “Student Services” department. This consolidation might lead to the reduction of one leadership or administrative role.
5.Optimize Academic Roles:
Merge “VP Academic Affairs” with “Dean’s Council”. By doing this, one leadership role might be reduced.
6.Centralize IT Functions:
Consolidate “Web Services” and “Analytics” under a unified “IT & Analytics” department. This could result in the elimination of one specialized role.
7.Shared Services for Operations:
Group “Plant Operations,” “Campus Safety,” and “Maintenance” into a “Campus Operations & Safety” department. Combining these may reduce one managerial or oversight position.
8.Advisory Role Redefinition:
Assess the roles of “Staff advisor” and “Fac. advisor”. If their duties overlap significantly, consider combining these roles into one comprehensive advisory position.
9.Role Expansion:
The role of “Grant writer” could potentially expand to also handle some events, possibly reducing the need for a separate “Events” position.
10.Centralize Financial Operations:
Group “Financial Services,” “HR,” and “I.T.” into a singular “Administrative Services” department. This consolidation might lead to the elimination or merging of one of the leadership roles overseeing these functions.
Given the complexity of organizational changes, it’s crucial to evaluate the impact of each change on operational efficiency, team morale, and the overall goals of the organization. Implementing these changes would likely achieve a reduction close to 10-15%, but exact numbers would depend on the specifics of each role and the current staffing levels in each department.
To verify that the generative ai tool understands and is considering the full org chart, it can be asked to draw the original org chart occasionally, and even a new working copy incorporating changes it has suggested making. It will output the org chart outline format as text. If you wish to remove some recommendations or insert others not picked up by its initial suggestions, you can instruct it to make edits, deletions, or additions to the working model and display the new revised model.
Use case 5: Cost Modeling and Estimation
Another common need in org design is calculating the cost of additions or consolidations. Because the whiteboard did not contain cost information one could upload a file with salary data, but that would risk adding that data to the large language model. Organizations and ai providers are building private chatGPT environments to create secure areas for employees to use such tools while protecting a company’s intellectual property. In this case the tool was just instructed to assume an average typical salary gleaned from publicly available data on the internet.
Prompt 5: “Using typical average market rates for these positions, can you compare the cost before and after these changes?”
Response 5: (Note: Obviously, this type of analysis would be carefully performed and validated by a finance analyst, but the capability to do an estimation during the brainstorming process could be helpful to the design team to get a general idea of outcomes.)
Estimating the cost changes before and after the restructuring involves making several assumptions. I don’t have real-time salary data, but I can provide a broad overview using average market rates for these positions as of my last update in 2022. Remember, the figures I’ll provide are just illustrative and may not reflect current rates or specific regional differences.
Before the Changes:
(These are illustrative figures and might differ from current or regional rates.)
1. President: $250,000
2.VP (each VP role): $180,000 x 4 = $720,000
3.Directors (each): $120,000 x 7 = $840,000
4.Assistant Dean, Shop Supervisor, Career Development, Business Engagement: $90,000 x 4 = $360,000
5.Managers (Financial Services, HR, I.T., Web Services, Analytics): $85,000 x 5 = $425,000
6.Grant Writer, Events, Community Outreach, Alumni Affairs, Librarians, Lectures & Exhibits: $70,000 x 6 = $420,000
7.Plant Operations, Campus Safety, Maintenance: $65,000 x 3 = $195,000
8.Admissions, Fin Aid, Career Services, Student Life, Academic Achievement, Enrollment Mgt: $60,000 x 6 = $360,000
Total Before: $3,270,000
After the Changes:
1.President: $250,000
2.VP (each VP role): $180,000 x 4 = $720,000
3.Administrative Services Manager: $90,000
4.Campus Operations & Safety Manager: $80,000
5.Communications & Marketing Department Manager: $130,000
6.Community & Alumni Relations Manager: $75,000
7.Grant Writer & Events: $80,000
8.Student Services Manager: $70,000
9.VP Academic Affairs & Dean’s Council: $200,000
10.IT & Analytics Manager: $90,000
11.Other roles (Student Life, Academic Achievement, Librarians, Lectures & Exhibits, Assistant Dean, Shop Supervisor, Career Development, Business Engagement, Div Chair & Faculty): $80,000 x 9 = $720,000
Total After: $2,415,000
Difference:
$3,270,000 - $2,415,000 = $855,000 saved
By merging roles and eliminating certain positions, the organization could potentially save around $855,000 based on these illustrative figures. However, actual savings would depend on the precise salaries in your region, other overhead costs, and severance or retraining costs associated with restructuring. It’s also crucial to consider the long-term implications of these changes on organizational performance and employee morale.
Using web-sourced numbers may not match this company’s actual salary ranges, or contain other problems. But for the sake of exploring generative ai’s functionality in this space is sufficient. If a number were too out of line, the design team could tell chatGPT to use a different number for a specific job and update the analysis.
There is a problem in the above answer, in that the amount of savings is greater than the requested 10-15%. This is a good example of why it’s important to carefully verify everything produced by the generative ai, although the same applies to the work of any human member of the design team as well. Both people and machines can make errors.
Use case 6: Cost modeling and Critical Role Analysis
The next prompt challenges generative ai’s ability to correct itself, and adds a qualitative challenge requesting the ai develop a prioritized refactoring of savings opportunities by analyzing the relative difficulty of its possible changes.
Prompt 6: “That’s more savings than the 15% needed. Can you remove the suggested changes that would be most difficult and explain why you removed those, and rerun the analysis to achieve approximately 15% reduction?”
Response 6: (Note: We can ask it to explain it’s rationale for each decision to aid the design team’s discussion of suggested options and help in comparing the team’s own further ideas. )
Certainly. Let’s prioritize by maintaining critical functions and roles that directly impact the institution’s mission and by considering potential challenges and disruptions.
Suggested Roles to Keep and Why:
1.VP Academic Affairs & Dean’s Council: Given the importance of academic affairs in a university setting, merging these might overburden the VP and negatively affect academic operations.
2.IT & Analytics Manager: Given the increasing reliance on technology and data, it’s crucial to have separate, specialized roles in IT and Analytics. Consolidating might dilute the focus on each critical function.
3.Student Services Manager: Given that student services directly impact the core mission of serving students, it might be more challenging to consolidate this role with other roles and maintain the same level of service quality.
Revised Estimated Cost After Restoring Above Roles:
1.President: $250,000
2.VP (each VP role): $180,000 x 4 = $720,000
3.Administrative Services Manager: $90,000
4.Campus Operations & Safety Manager: $80,000
5.Communications & Marketing Department Manager: $130,000
6.Community & Alumni Relations Manager: $75,000
7.Grant Writer & Events: $80,000
8.IT & Analytics separate roles: $85,000 x 2 = $170,000
9.VP Academic Affairs & Dean’s Council separate roles: $180,000 + $70,000 = $250,000
10.Student Services Manager: $70,000
11.Other roles (Student Life, Academic Achievement, Librarians, Lectures & Exhibits, Assistant Dean, Shop Supervisor, Career Development, Business Engagement, Div Chair & Faculty): $80,000 x 9 = $720,000
Revised Total After: $2,635,000
Difference:
$3,270,000 (Before) - $2,635,000 (Revised After) = $635,000 saved
This revised difference represents a savings of approximately 19.4%. To hit closer to the 15% savings target, you might consider adjustments like salary freezes, reduced bonuses, or temporary cutbacks in other expenses, rather than further structural changes to the org chart. It’s a balance between achieving the savings target while minimizing disruption and preserving essential functions.
No definition of ‘difficulty’ was given in the prompt. In this case the ai inferred what was meant and offered clarification in its response setting difficulty = critical functions that impact mission and consideration of challenges and disruptions. you could insert a a more granular human-in-the-loop process here by having the ai first assess each of its suggestions, explain them, and giving feedback to ensure the right roles were prioritized.
领英推荐
Of note is the response’s ability to use ranges and estimations. It was initially asked for a range of 10-15%. Even in the request revision prompt, the term “approximately 15” was used. This time the ai recommended 19%, something acceptably closer to 15%.
It also raises the point that if cost saving is the goal, there are other avenues to be considered before staff disruption. Note that you could ask generative AI to help explore, evaluate, compare, and plan any of the alternate cost saving options mentioned. We’ll stay the course as this article is focused on exploring ai’s capabilities to augment org design.
Use case 7: Qualitative Multidimensional Analysis
Now let’s explore a much more complex analytical question, incorporating multiple dimensions including skills adjacency, career progression, operational risk, and value of each role to the organization.
Prompt 7: “Because we are committed to our teammates’ job security and will do this through natural attrition over 2 years, we don’t know exactly which positions will be vacated. Can you suggest the top 5 most critical roles in the revised org design and identify the most similar feeder jobs based on skill adjacency so we can begin developing upskilling and succession plans? Also include thoughts on how the evolution of ai may factor in to changing jobs. Include an explanation of why you chose those feeder jobs.”
Response 7: (Note: In data science this might be some kind of multidimensional analysis, and require the encoding of qualitative data which inherently loses valuable context and takes considerable time. Due to those limitations, this is often relegated to less structured conversation.)
Absolutely. Here are the top 5 most critical roles in the revised org design, based on their impact on the institution’s core mission and operations:
1. President: As the top executive, this role provides the overall strategic direction, leadership, and vision for the institution.
? Feeder Job: VP Operations. The VP Operations has a broad view of many of the institution’s primary functions, making them well-suited for the strategic leadership the President role requires.
? Reasoning: The VP Operations has experience in managing large teams, understands the inner workings of the institution, and has probably been involved in strategic decisions.
2. VP Academic Affairs: Responsible for the academic integrity of the institution, curriculum, faculty matters, and student academic support.
? Feeder Job: Dean’s Council or Assistant Dean.
? Reasoning: Individuals in these roles already have a deep understanding of academic affairs, curriculum development, and faculty relations. They’re experienced in academic administration and can transition smoothly to a VP role.
3. VP Communications & Marketing: Manages the institution’s branding, public relations, and outreach efforts.
? Feeder Job: Communications & Marketing Department Manager.
? Reasoning: The role already involves many of the same responsibilities, just at a different scale. Upskilling here would focus on strategic decision-making and broader organizational leadership as well as the increasing automation capability to perform communication and marketing tasks.
4. IT & Analytics Manager: Ensures the institution’s technological infrastructure is robust, secure, and supports both current needs and future growth.
? Feeder Job: Web Services or Analytics roles.
? Reasoning: Individuals in these roles have a grasp of technology and data. Moving forward, individuals in these roles are likely to be staying abreast of ai advances which will increasingly present new opportunities needed in the future of academic instutions. Upskilling would involve broadening their understanding of overall IT infrastructure, cybersecurity, and strategic tech planning.
5. Student Services Manager: Directly responsible for student welfare, ensuring they have access to necessary resources and support.
? Feeder Job: Student Life or Academic Achievement roles.
? Reasoning: These roles are already closely related to student welfare and success. Upskilling would involve management skills, strategic planning, and possibly broader student welfare initiatives.
For each feeder job, the upskilling plan should focus on filling the gap between the current responsibilities of the feeder job and the demands of the target critical role. Upskilling might include leadership training, courses on strategic planning and decision-making, cross-functional training, and mentorship programs.
Again the response is in about 10 seconds. Asking generative ai to explain it’s reasoning is helpful let the design team understand and challenge the reasoning as it considers this starting point and other roles it may determine are more critical.
The large language model has a broad understanding of these concepts based on what it’s learned from articles, books, public curricula, news, blogs and many other sources. However it does not understand the environment of this organization outside of that context, only humans have this knowledge. But the above is helpful to get the conversation going.
Qualitative Role Evolution Modeling
There is a lot going on in this last question. This is an innovative and complex approach that combines the analytical capabilities of ai with qualitative insights, some form of multidimensional analysis, and future forecasting. Let’s break down the tasks required to answer this:
1. Identifying Critical Roles: the ability to evaluate and identify critical roles in a typical university’s organization structure based on historical and existing knowledge. Determining their criticality over the next 2 years would require forecasting and a clear understanding of the changing dynamics of higher education.
2. Skill Adjacency and Feeder Roles: provide insights into roles that typically serve as feeders based on skill similarity. This includes knowledge of career progression paths and the skills that are relevant to those moves.
3. AI Impact on Job Tasks: included here is a qualitative analysis of the potential impact of AI on various job roles based on existing knowledge. This is speculative analysis, which prior to openai’s public release of ChatGPT we all thought was safely protected in what we naively called “Uniquely Human Skills” (creativity, compassion, communication, and collaboration).
4. Upskilling Needs: Based on the identified changes in job tasks and roles, suggesting potential areas for upskilling requires some degree of comprehension about skill leveling, differences in various job descriptions, and the ability to cluster them and map shortest paths between them.
The generative part of generative ai is surprising in it’s creative approach to solving problems contained in questions. This may lead to new analytical methods and approaches. Let’s call the analytical technique used to answer the question “Qualitative Role Evolution Modeling”. It is a blend traditional task-based analysis, skill adjacency, future influence along with AI’s capabilities to process and interpret complex, subjective information. While elements of this approach have been used in various contexts, this seems to be an innovative new approach to what was previously a very manual process.
Use case 8: Structural agility analysis
Increasing and organization’s agility is an often-sought outcome of organization optimization, allowing companies to more quickly innovate, respond to opportunities, act early when conditions start to change, and compete with market disrupters. This question explore’s generative ai’s ability to assist a design team in analyzing how an organization’s structure could improve agility.
Prompt 8: “How easily could the organization react to changes given its existing organization structure? What recommendations might make it more agile in times of change?”
Response 8: (Note: 3 possible weak points are initially identified in this response. 10 suggested mitigations are recommended and are all included here.)
Given the existing organization structure, here are some observations regarding its agility and recommendations to enhance flexibility:
Observations:
1. Centralized Decision Making: The current structure has a significant centralization around the “President” role and the Board of Trustees. While centralization can streamline decision-making, it might slow down adaptability in rapidly changing environments.
2. Departmental Silos: Multiple vertical departments (e.g., VP Operations, VP Communication, VP Student Affairs) may lead to departmental silos, making cross-departmental collaboration challenging.
3. Duplication of Roles: As previously noted, there seem to be overlaps in functions, which can lead to inefficiencies and slow down the response to changes.
Recommendations:
1. Decentralize Decision Making: Empower middle management and team leads with decision-making authority for specific areas. This reduces the dependency on top-tier leadership for every decision and allows for quicker adaptations.
2. Cross-functional Teams: Foster the creation of cross-functional teams that cut across departmental lines. These teams can work on projects or initiatives that require multi-departmental expertise, enhancing collaboration and breaking down silos.
3. Regular Strategy Reviews: Implement periodic strategy reviews to assess the organization’s direction and agility. These reviews can help identify areas that need adjustments, ensuring the organization remains responsive to external changes.
4. Invest in Technology & Training: Modernize the organizational tools and systems to enable faster communication, data sharing, and decision-making. Along with this, provide regular training for employees to adapt to new tools and methodologies.
5. Feedback Loops: Create robust feedback mechanisms to ensure that insights from ground-level employees, who are often the first to witness changes or challenges, reach the top leadership quickly.
6. Flexible Roles: Encourage a culture where employees are trained in multiple skills, allowing them to adapt to different roles as needed. This makes the organization more resilient to unexpected changes.
7. Scenario Planning: Regularly engage in scenario planning exercises. This helps the organization prepare for potential future changes and develop strategies to address them proactively.
8. Open Communication: Maintain open channels of communication at all levels. This ensures that any changes, whether internal or external, are communicated swiftly, ensuring everyone is aligned.
9. External Collaborations: Forge collaborations with external partners, such as industry associations, research institutions, and other universities. These collaborations can provide fresh insights and help the organization stay ahead of industry trends.
10. Continuous Learning Culture: Encourage a culture of continuous learning. Provide opportunities for employees to upskill, ensuring they remain relevant in a changing environment and can bring innovative ideas to the table.
By implementing these recommendations, the organization can become more agile, better positioned to anticipate changes, and respond effectively to shifts in the educational landscape or external environment.
Several of the mitigations extend beyond the narrow scope of questions about organization structure. Because of an large language model’s broad knowledge base, generative ai’s responses often take into account additional recommendations related to the direct question, in this case suggestions that can be built into the organization’s processes and culture to achieve the requested end.
Use case 9: Comparing Industry Structures
A common early step in organization optimization is for a company to compare its organization’s structure to other companies in the same industry. This generates ideas that get design teams thinking outside their box and breaking away from legacy assumptions that are otherwise surprisingly difficult to recognize.
Prompt 9: “What major differences do you see between this organization’s structure compared to other companies in a similar industry that could be helpful as we consider changes?”
Response 9: (Note that it is still helpful to do comparisons with specific peer organizations, but this is a quick way to start brainstorming insights.)
While I don’t have a direct comparison with another specific organization’s structure in the same industry based on the provided image, I can provide general observations on common organizational structures in similar industries, such as educational or non-profit institutions:
1. Centralized Leadership: In many organizations, the leadership is more centralized, with fewer direct reports to the President or CEO. Instead, more roles report to the VPs, making the top tier less crowded and streamlining decision-making.
2. Unified IT and Communication: It’s common for IT-related roles (like “Web Services” and “Analytics”) to be grouped under a single IT or Technology department rather than spread out. This promotes technical synergy and clearer communication.
3. Student Affairs and Academic Affairs: In many educational institutions, “Student Affairs” and “Academic Affairs” are two of the most significant branches. The distinction between these two often involves “Academic Affairs” focusing on curriculum and faculty, while “Student Affairs” handles student life, support, and services.
4. Operational Roles: Often, roles related to campus infrastructure like “Plant Operations”, “Maintenance”, and “Campus Safety” fall under a Facility Management or Campus Operations department, which ensures that all physical aspects of the campus are coordinated.
5. Financial Roles Separation: In many organizations, there’s a distinct Finance or Financial Affairs department that houses roles related to finance, including “Financial Services”.
6. HR Standalone: Human Resources (HR) typically stands as its own department, given its importance in staff recruitment, training, and welfare.
7. Community and Alumni Relations: In some institutions, community outreach and alumni affairs are combined into a single “External Relations” or “Community and Alumni Engagement” department, leveraging synergies between community and alumni initiatives.
8. Specialized Roles: Special roles like “Grant Writer” might be part of a broader “Development” or “Fundraising” department in other organizations, emphasizing fundraising and donor relations.
9. Clarity on Advisory Roles: The role of advisory bodies like the “Cabinet” is more explicitly defined in some organizations, outlining their responsibilities and their interaction with other departments.
10. Decentralized Marketing and Communication: While centralized communication departments are beneficial, some institutions prefer a decentralized approach, with each major department having its marketing or communication sub-unit. This allows for more specialized and targeted messaging.
When considering changes, it’s essential to understand the unique needs, culture, and goals of your organization. While these general industry practices provide a starting point, they should be adapted to fit the specific context and mission of your institution
This could be a good source of ideas to consider. Interestingly some of the observations counter earlier recommendations. For example, the idea of centralizing decision-making runs counter to the suggestion early to decentralize decision-making for greater agility. The same applies to suggestions about consolidating or decentralizing marketing and communication. All are valid alternatives, the design team would need to evaluate the company’s strategy and choose the best-suited models.
Use case 10: OrgAI Design Analysis: Many More Capabilities…
This is just a sampling of the capabilities of generative ai for organization design and optimization. Here is a list of additional sample prompts that were not included in this article.
Four important observations
Human-in-the-Loop or ChatGPT-in-the-Loop? Why not turn generative ai loose and let it automate all the analysis? Because the data needed for organization optimization is not all contained within the org chart. Optimization requires additional data that comes from our human world, and a large language model does not have access to that. Knowing organizational cultures, current company strategies, interdepartmental relationships, and many other factors that must be considered.
Generative AI isn’t for everything. Many problems are better solved with financial, statistical, data science, or other quantitative tools. Choose the right tool for the job.
Trust but verify. The examples in this article, although drawing from a real org chart and process, are presented here to demonstrate the capabilities of OrgAI Design Analysis, and are not meant to be meticulously validated answers. In a few areas errors are called out, but the list is likely not complete. Use of generative ai must always be validated for bias, hallucination, omissions, etc (also true of humans’ work.)
This isn’t a full how-to for org design. These are just a few samples of the kinds of analysis and inspection that design teams often include in their processes.
Also, the artwork for this article? Created with chatGPT4 and DALL-E 3.
Professor and co-founder of Reconfig
5 个月Thank for a thorough post Mike Harmer with lots specific examples. As you note, one thing that the LLMs are missing are company-specific data. Combine that with AI, and it becomes even more powerful.
Senior Vice President Information Technology at Platinum Dermatology Partners
11 个月This is one of the best step by-step examples I have seen. Thank you for sharing it!
Mike Harmer - an exceptional example of the practical application of #GenAI. This is sure to spur other creative thoughts and #innovation #brainstorming.
Chief Medical Information Officer (i) at University of Utah Health Executive Vice Chair Department of Neurology University of Utah School of Medicine
11 个月This is very insightful Mike and highly pragmatic. Appreciate your forward thinking approach as always.