Project Management

Project Management Central

Please login or join to subscribe to this thread
Data Set Management
Hello PM Community!

I have recently been offered an opportunity to take on a Program Manager role to drive the execution of (some portion) of our overall data strategy dealing specifically with data sets (generating, acquiring, & working with other teams on managing access). These data sets will be used for research purposes and to train ML algorithms. There are several other internal teams involved in this strategy and I imagine my role would consist mainly of the following activities.

a) Develop and drive a process to solicit data requests across the organization
b) Develop and drive a process for prioritization of data set needs
c) Be a central point of contact for all incoming data set requests
d) Manage communications across my BU about needs & asks
e) Provide inputs to the process for how we store data and manage our data sets
f) Vendors – be the centralized source of expertise on how we select, manage and work with data vendors. Who to work with, how to vet them, etc.
g) Coordinate with finance / budget team…

Have any of you had experience as a PM/PgM managing data projects/strategies like this? If so, could you share your advice on tools, processes, procedures, and general advice on what I can do to be successful in this role? Do you have any good online sources I should read? Or advice to share? Templates that would be helpful, (i.e. for vendor selection, data prioritization, other)?

Thank you,
Dan
Sort By:
Hi Dan

good luck!

From a program manager perspective, I would refer to the domains of PgM performance:

1. benefits management: is there a business case, which benefits are expected and how will they be measured

2. strategy alignment: any strategy is subject to change, driven from external but also internally from your program, as you identify reality

3. governance: how is your program embedded in the organizational decisions, where does your authority come from, how do you influence the projects within your program

4. stakeholder engagement: besides internal communication how do you deal with external stakeholders, build alliances, sell your program, control politics

5. there is much to say about your program lifecycle: do you have a roadmap, a set of projects, a cost timeline etc
...
1 reply by Daniel Berenholz
May 01, 2020 1:42 AM
Daniel Berenholz
...
Great to hear from you Thomas, I hope you are doing well and thank you for the feedback.

These are all important questions to have answered and I have already asked several of them. In this case, there doesn't seem to be an enterprise strategy confirmed (yet) and data needs have been addressed in an ad-hoc and siloed manner within each product BU. I have more to say on that, but I'll leave it there for now.

I will certainly address all of the topics you shared above with my management team and in the meantime would love to hear some ideas on learning paths that could help me ramp up in the short-term and also as this role develops long-term. Of course, having a Master's in Data Science would be ideal, but something that might provide some tactical guidance would be extremely helpful to get me started. In the meantime, you've inspired me to take a look at the Standard for Program Management. :)

Cheers,
Dan
May 01, 2020 1:03 AM
Replying to Thomas Walenta
...
Hi Dan

good luck!

From a program manager perspective, I would refer to the domains of PgM performance:

1. benefits management: is there a business case, which benefits are expected and how will they be measured

2. strategy alignment: any strategy is subject to change, driven from external but also internally from your program, as you identify reality

3. governance: how is your program embedded in the organizational decisions, where does your authority come from, how do you influence the projects within your program

4. stakeholder engagement: besides internal communication how do you deal with external stakeholders, build alliances, sell your program, control politics

5. there is much to say about your program lifecycle: do you have a roadmap, a set of projects, a cost timeline etc
Great to hear from you Thomas, I hope you are doing well and thank you for the feedback.

These are all important questions to have answered and I have already asked several of them. In this case, there doesn't seem to be an enterprise strategy confirmed (yet) and data needs have been addressed in an ad-hoc and siloed manner within each product BU. I have more to say on that, but I'll leave it there for now.

I will certainly address all of the topics you shared above with my management team and in the meantime would love to hear some ideas on learning paths that could help me ramp up in the short-term and also as this role develops long-term. Of course, having a Master's in Data Science would be ideal, but something that might provide some tactical guidance would be extremely helpful to get me started. In the meantime, you've inspired me to take a look at the Standard for Program Management. :)

Cheers,
Dan
Hi Daniel,

I think @Thomas as detailed pretty well most of the actions that you should follow, nevertheless I'd like to add that if you have to do this you have to develop a Information Security Governance if not exist in your company or if exists work in collaboration with the Information Security Manager, normally all data most comply with external regulations and internal policies and must exist access levels to the data.

There are a lot of information how to develop Information Security Governance plans normally the basic steps are the following:
1 - Formulate a information security strategy
2 - Integrating information security in the actual governance
3 - writing information security policies
4 - Creating Business cases
5 - Influence information Security Governance
6 - Gaining information security stakeholders commitment
7- Information Security Roles and Responsibilities.

This kind of projects implies that we must develop roadmaps of the information across all channels of the company , who is accessing the information, when and if have permissions for that.

I think a step like this should the foundation of your program.

Alexandre
...
1 reply by Daniel Berenholz
May 01, 2020 12:30 PM
Daniel Berenholz
...
Hi Alexandre,

Thank you for your feedback. I’ve updated the title of my post to more accurately describe what aspect of an overall data strategy I will be managing. In this case there are several teams involved, Research, Product Teams, and other team(s) that manage the buildout of the platform that hosts the data sets (among other things) and will also deal more with the security/data governance. In my case, I will focus mainly on the items listed above and will probably interact in a very limited capacity with the data privacy and infosec teams (at least for now).
Daniel -

A data strategy program needs to consider organizational design to support sustainment of the objectives once the program is over. For example, will you have a Chief Data Officer? Will there be defined Data Stewards and if so, what responsibilities will they have?

Kiron
May 01, 2020 5:26 AM
Replying to Alexandre Costa
...
Hi Daniel,

I think @Thomas as detailed pretty well most of the actions that you should follow, nevertheless I'd like to add that if you have to do this you have to develop a Information Security Governance if not exist in your company or if exists work in collaboration with the Information Security Manager, normally all data most comply with external regulations and internal policies and must exist access levels to the data.

There are a lot of information how to develop Information Security Governance plans normally the basic steps are the following:
1 - Formulate a information security strategy
2 - Integrating information security in the actual governance
3 - writing information security policies
4 - Creating Business cases
5 - Influence information Security Governance
6 - Gaining information security stakeholders commitment
7- Information Security Roles and Responsibilities.

This kind of projects implies that we must develop roadmaps of the information across all channels of the company , who is accessing the information, when and if have permissions for that.

I think a step like this should the foundation of your program.

Alexandre
Hi Alexandre,

Thank you for your feedback. I’ve updated the title of my post to more accurately describe what aspect of an overall data strategy I will be managing. In this case there are several teams involved, Research, Product Teams, and other team(s) that manage the buildout of the platform that hosts the data sets (among other things) and will also deal more with the security/data governance. In my case, I will focus mainly on the items listed above and will probably interact in a very limited capacity with the data privacy and infosec teams (at least for now).
I have lot of experience in the field from the time where all these stuff was called just data warehousing. The reason because it is a program is because the whole endeavour is based on architecture and each architecture has components the usually each component is manage as a project. Then, here I fully agree with @Kiron. First of all this will be a solution for the company. So, sorry for put here a link to something I wrote, but perhaps this helps you: https://www.projectmanagement.com/blog-pos...-right-solution
Second, make some research on data warehousing components (the environment is still calling in that way) to understand the architecture and its components then to understand what you have to consider as projects. Returning to @Kiron point, it is critical to understand the strategy to select which components to consider. here is where my article could help.

Please login or join to reply

Content ID:
ADVERTISEMENTS

You may have to fight a battle more than once to win it.

- Margaret Thatcher

ADVERTISEMENT

Sponsors