The tension every shelter knows
Domestic violence shelters get asked to “show outcomes” constantly — by funders, by boards, by their own strategic plans. But most of what counts as impact data in this sector means asking survivors to relive parts of a story they’ve already told two or three other people that same week: intake staff, a caseworker, maybe a funder-mandated survey.
There’s a better way to build real outcome data, and it doesn’t require adding a single new question to any survivor’s intake or exit process. In fact, It is built entirely from routine administrative information that shelters are already generating as a normal part of service delivery — no re-traumatization required.
In a national context where, per the 19th Annual DV Counts Report from the National Network to End Domestic Violence, U.S. domestic violence programs served 79,088 people in a single 24-hour national census and still couldn’t meet 14,095 requests for help. 60% of them housing-related, which shows the gap between what shelters do and what they can prove they have done is not a small operational detail. It is a resourcing problem with real consequences for the next person who calls.
Housing stability at exit and follow-up
The first metric is a simple yes/no status check at a standard follow-up point — often three or six months after program exit — which is typically already part of most shelters’ aftercare practice in some form. As expected, there isn’t additional narrative required nor further conversation about what happened. It is just a status marker such as whether the client is still stably housed, or not, and tracked consistently across every client who exits the program.
This single data point, monitored reliably over time, tells a funder something an output number like “number of clients served” never can: whether the program’s work holds up after a client leaves the building. It is also one of the lowest-burden metrics to collect, because most shelters already have some version of an aftercare check-in built into their process. At the end, the shift is simply in recording the answer consistently, every time, rather than informally.
Completed connections to next-step resources
The second metric tracks whether a referral to legal aid, income supports, or longer-term housing was made and completed — not attempted, completed. This is administrative data that case managers are often already generating as a normal part of a client’s exit plan; the work here is in capturing it systematically rather than letting it live only in individual case notes.
This metric matters because it captures something output data misses entirely: the difference between offering a resource and a client actually connecting with it. A shelter that can show a high completion rate on next-step referrals is demonstrating something concrete about the quality of its transition planning and without asking a single survivor to narrate anything about their experience.
Shelter re-entry rate, framed carefully
The third metric which is how many clients return to shelter within a defined window, needs more context than the first two, and it is worth naming that up front. A higher re-entry rate can reflect systemic housing gaps and a lack of affordable options in the community as much as it reflects anything about program performance. This is not a metric to present without that framing attached.
Tracked internally, with that context built in from the start, it is one of the most honest systems-level indicators available to a shelter, precisely because it resists a simple “good number, bad number” reading and forces an honest conversation about what’s actually driving the pattern.
Start with one metric, not three
A funder would rather see one metric tracked reliably across every client than three tracked inconsistently. If your shelter is starting from zero on structured outcome tracking, pick the metric closest to what your team already does as part of its regular workflow, and build consistency there before adding the others. Housing stability at follow-up is often the most natural starting point, since the aftercare check-in it depends on is already common practice in many programs.
Consistency, more than completeness, is what makes a metric credible to a funder reviewing data across multiple grant cycles. A single number tracked the same way every time will do more for a shelter’s credibility than three numbers tracked sporadically.
Why this approach matters
None of these three metrics asks a survivor to answer a single new question. They’re built from information shelters are already collecting as part of routine service delivery — the shift is in how consistently that information gets recorded and reported, not in what gets asked of the people the program serves.
BData Solutions works with domestic violence and homelessness-serving organizations to build outcome frameworks exactly like this one — trauma-informed by design, built from data you’re already generating, and structured to hold up under funder scrutiny.
