I don't know where you're located, but in the US, the HMIS was intended to serve this purpose. It hasn't for a variety of reasons, but especially because it isn't easy to use so only trained intake specialists use it. Beyond just poor design, the legal requirements for privacy and non-discrimination make intake and data tracking a fraught activity. Also homeless people will frequently avoid giving accurate biographical information to avoid denial of services. So those are two major obstacles you will face.
Tracking homeless individuals is an incredibly challenging task, even without considering the tech. For example, my girlfriend works in macro-level social work primarily focusing on the homeless population. (She and I are data geeks so we talk about this a lot). She currently works at a shelter where they are trying to identify the homeless citizens who cost hospitals and medicaid the most money, because there's some insane distribution of costs such that something like 1% of the patients account for 90% of the costs. She can't get any cost data from hospitals though, because they refuse to disclose it, even anonymized. Medicaid only has part of the data at best and the government officials there haven't been responsive to her requests. So she can only get the cost data by getting individual clients to sign consent forms. Which is methodologically troublesome because she only has access to the clients that happen to come to her shelter and many of the highest cost patients are specifically those who don't go to shelters and instead end up in the hospital due to exposure/hypothermia/etc. Point being, she can't get the necessary data in the first place, never mind that she doesn't have anywhere to store it.
Beyond tracking individuals, the data problems in the field are enormous and I think that may be where you can make the most difference. Even within her one agency, she can't get a lot of the basic data to evaluate programs because the agency has not invested any money in tech for 15 years. Primarily because they can barely afford to stay open. So they've got tons of legacy systems with data scattered around. No one knows where anything is, and even if they did know where to find data, they wouldn't know how to aggregate it or display it in a useful manner.
This is a common problem for community services organizations. They are chronically underfunded and tend to be run by people with little expertise in tech. And the staff tend not to be much more qualified. My girlfriend asked their in-house accountant for a list of expenses related to a specific set of grants the other day. The accountant sent her a part of the general ledger, which is incomprehensible to anyone who isn't familiar with accounting (and it was the wrong part, but that's another story). She also went to a program director and requested some demographic data that was required for reporting under the grants. Turns out he hadn't been collecting it.
Anyway, my point is that it would be helpful for many of these shelters and agencies to have a consolidated information system. So, for example, they can enter information about a grant they've received, including important dates and reporting requirements. And then have some tool to ensure that the program director is aware of and meeting those requirements.
So far, my girlfriend's biggest data/tech success at this agency has been a project I helped her code. The head fundraiser has to fill out a ridiculous number of grant applications every month. They are all in PDF and mostly request the same kinds of information (agency location, purpose, intended use of grant, etc.). So we wrote a program that takes a canned set of responses and automatically inserts them into the PDFs, so that the fundraiser only has to enter answers for non-standard questions. It then auto-updates their grant tracking system. It's cut down on the time they spend processing applications tremendously.
It would work a lot better if there was an integrated information system from which we could pull a lot of that data automatically, because currently it requires the fundraiser to keep the underlying data store up to date manually.
Tracking homeless individuals is an incredibly challenging task, even without considering the tech. For example, my girlfriend works in macro-level social work primarily focusing on the homeless population. (She and I are data geeks so we talk about this a lot). She currently works at a shelter where they are trying to identify the homeless citizens who cost hospitals and medicaid the most money, because there's some insane distribution of costs such that something like 1% of the patients account for 90% of the costs. She can't get any cost data from hospitals though, because they refuse to disclose it, even anonymized. Medicaid only has part of the data at best and the government officials there haven't been responsive to her requests. So she can only get the cost data by getting individual clients to sign consent forms. Which is methodologically troublesome because she only has access to the clients that happen to come to her shelter and many of the highest cost patients are specifically those who don't go to shelters and instead end up in the hospital due to exposure/hypothermia/etc. Point being, she can't get the necessary data in the first place, never mind that she doesn't have anywhere to store it.
Beyond tracking individuals, the data problems in the field are enormous and I think that may be where you can make the most difference. Even within her one agency, she can't get a lot of the basic data to evaluate programs because the agency has not invested any money in tech for 15 years. Primarily because they can barely afford to stay open. So they've got tons of legacy systems with data scattered around. No one knows where anything is, and even if they did know where to find data, they wouldn't know how to aggregate it or display it in a useful manner.
This is a common problem for community services organizations. They are chronically underfunded and tend to be run by people with little expertise in tech. And the staff tend not to be much more qualified. My girlfriend asked their in-house accountant for a list of expenses related to a specific set of grants the other day. The accountant sent her a part of the general ledger, which is incomprehensible to anyone who isn't familiar with accounting (and it was the wrong part, but that's another story). She also went to a program director and requested some demographic data that was required for reporting under the grants. Turns out he hadn't been collecting it.
Anyway, my point is that it would be helpful for many of these shelters and agencies to have a consolidated information system. So, for example, they can enter information about a grant they've received, including important dates and reporting requirements. And then have some tool to ensure that the program director is aware of and meeting those requirements.
So far, my girlfriend's biggest data/tech success at this agency has been a project I helped her code. The head fundraiser has to fill out a ridiculous number of grant applications every month. They are all in PDF and mostly request the same kinds of information (agency location, purpose, intended use of grant, etc.). So we wrote a program that takes a canned set of responses and automatically inserts them into the PDFs, so that the fundraiser only has to enter answers for non-standard questions. It then auto-updates their grant tracking system. It's cut down on the time they spend processing applications tremendously.
It would work a lot better if there was an integrated information system from which we could pull a lot of that data automatically, because currently it requires the fundraiser to keep the underlying data store up to date manually.