View All Pages

Finding A Client

Finding a Client

Throughout the semester you will be working with a client to design an information system that addresses some business need of that client. The first step in this process is to identify at least one potential client. Our goal is to have projects selected and approved by February 10 at the very latest.

Client selection criteria

Your client should be a business, non-profit, club, or other organization that has identified a need for additional information systems capabilities to support their work. One way to begin is to look around you for things that need fixing or capabilities that are missing in your own experience with an organization or business.

The need does not have to be clearly defined. Helping the client to define the needs and make a business case for a new information system is part of your work this semester. That said, it helps if the client has at least a broad idea of what the problem or concern is.

The need should be focused primarily on new software, since our work this semester will be on methods for defining requirements for software. So, for example, a business whose identified need is to upgrade their computers or set up a new backup system would not be a good candidate for a project.

We are looking for what might be termed "Goldilocks Links to an external site." projects. Not too big, not too small, but just right. Projects that are too simple do not give you adequate opportunities to develop your analysis and design skills. Projects that are too complex will be difficult to deliver in the scope of a semester course. See below for advice on how to identify a Goldilocks project.

Your client needs to be willing to commit to some meetings and conversations with you about the project. Your client should be prepared to commit 3-6 hours of time in total for interviews and follow up conversations.

Your client needs to have realistic expectations of what they will get out of this project. The prototype you develop will be hosted on a server that is not available for production use (and not available outside of BC). The client should not expect a working system as a result of the project. What you can offer the client is a copy of your project deliverable at the end of the course which will include a business case for the project, a set of requirements, and screen shots of a working prototype. This could be quite useful in helping your client decide on whether and how to proceed with an actual implementation, but your client needs to be comfortable with this as the deliverable. With these cautions in mind, it does sometimes happen that a team has been able to actually deliver a working system to their client (although this involves moving away from the standard MI258 prototyping environment). If later in the semester this looks possible then we should definitely talk about it, since that can be really exciting when it becomes possible. But do be cautious in what you promise.)

Where to look for a client

In the past students have worked with local businesses and BC organizations. You might begin by thinking about your contacts at BC: is there a need in a club you are a member of or the place where you work on campus? What local businesses might be candidates?

Note: past experience suggests that projects involving Office of International Programs (OIP or Hovey House) and Dining run into problems. We don't have a theory as to why but I am told there are enough data points to warrant caution when considering those sites for MI258 projects.

Another source of possible projects is family and friends. Is there a need in a family business or organization? You are free to work with an organization that is not in the Boston area. While face to face meetings with a client are ideal, you can use phone or video chat as an alternative. You may need to arrange for offsite clients to send you various documents. One thing to consider with a remote client is that you may not have the opportunity for a site visit, but this is not a show stopper.

Finally, your team could consider working with a non-profit. Such organizations are often in urgent need of IT consulting. If your team is interested in this possibility and wants assistance in locating a non-profit, please contact me.

Please feel free to reach out to the teaching assistants who can offer you advice based on their own experience identifying clients, and feel free to discuss this with me as well.

Identifying a Goldilocks project

As noted above, it is important to choose a project that is big enough to be interesting and small enough to be doable. It turns out to be easier (usually) to scale back a large project to fit MI258 than it is to add complexity to a project that is too simple, so your initial focus should be on making sure your project is big enough. Here are some suggestions:

Your project should involve some sort of data and should include both creating/updating that data and reading/analyzing that data.

Your project should involve more than one interaction with the computer. For example: there is more than one user of the system and each user has a different set of goals for interacting with the system. As we shall see, use cases are a powerful way to describe and evaluate system scope and, as you will soon see, what I am suggesting here is that your project involves a number (at least 3-4) of distinct and substantial use cases.

Another way to assess the scope of your project is through a "process lens." What business process is your project going to impact? How important is that process and how significant might the impact be on the process? Could your project lead to a redesign of that process? (Later in the semester we will be learning about business processes and how to model, analyze, and improve them.)

Agenda for initial client conversation

Before presenting your project idea to the class you should have had at least one brief conversation with a contact at the client. Preferably this is face to face or a voice conversation but email or messaging/chat is acceptable if necessary.

In that conversation you should describe the MI258 project and clarify what your client's commitment would be (3-6 hours) and what the deliverable would be (business case and design documents, not a working system). You should discuss with your client possible business needs and think about whether the needs are likely to result in a Goldilocks project.

Be sure your contact knows that your project needs to be approved and agree on how you will follow up with the contact.

Project Approval

Once you have identified a potential client we need to ensure that the project is not only interesting and potentially valuable, but also a good fit for the MI258 learning experience. Accordingly, all projects will need to be approved by me before proceeding.

After you present your project idea in class you can expect one of the following responses from me:

  • Approval to explore further. If I think your project is likely to work out as proposed, I will ask you to develop your thinking a bit further. A team assignment will be posted (due on 2/3) explaining what to do next. This will involve a very brief write up of the project idea.
  • Questions to resolve. If I think your project has potential but see some issues, I will give you specific questions to investigate or options to consider. You will want to follow up with your contact to discuss these. I will also give you a sense of how likely this is to work out (follow up with me if you are not clear with this after class on 1/27). Depending on my assessment you may also need to start exploring a second possible client as a backup or alternative.
  • Look elsewhere. If I see significant obstacles to this as a class project, I will ask you to find another client. This is not to say the client or project are without merit, but rather that it is not going to be a good MI258 project. You are welcome to continue to advocate for your project but you will need to immediately begin looking for another project. (I am open minded, but if this is my response to your project I am unlikely to be swayed.)