How to Do Market Research for a Mental Health App

Market research for a mental health app runs through five steps, in order: define your segment by condition, acuity, and care context; size demand with public data instead of a borrowed headline stat; map competitors by category, not feature list; identify who actually pays; and check whether your differentiation survives contact with that payer. Skip the ordering and you'll size a market that doesn't match the product you're about to build.

Quick Answer: Define your segment (condition + acuity level + care context) before you research anything else. Then size demand from public health data, map competitors across consumer/employer/provider/payer categories, identify who pays, and confirm your differentiation holds up against that specific buyer — not against every app in the category.

Prerequisites: Pin Down Condition, Acuity, and Care Context First

Every other step in this guide depends on three decisions you have to make before you open a single data source: which condition you're addressing, which acuity band you're serving, and what care context your product sits in. Get these wrong and your demand numbers, your competitor list, and your payer research will all be for a market you're not actually entering.

Condition is the specific problem — anxiety, depression, insomnia, substance use, ADHD, stress management — not "mental health" as a category. "Mental health app" is a shelf label, not a market. A sleep-focused CBT-I tool and a peer-support community for eating disorder recovery share a category page in the app store and almost nothing else: different users, different competitors, different buyers.

Acuity is how severe the person's need is, and it's the variable founders skip most often:

• Prevention and everyday stress — people managing normal life load, not in treatment, not seeking it.
• Sub-clinical and mild symptoms — real distress, no diagnosis, may never see a clinician.
• Diagnosed and in active treatment — working with a clinician; your app is an adjunct, not the care.
• Acute or crisis — immediate risk. This band needs a human, not an app, and no consumer product should be marketed or researched as if it can handle this alone.

That last point isn't a footnote — it shapes your research from the start. If any realistic slice of your users could be in acute crisis, your market research already has to account for the fact that the product must always provide a clear, visible path to real emergency help: in the US, the 988 Suicide & Crisis Lifeline, or 911 / local emergency services. You cannot ethically survey, interview, or market to that segment as though the app itself is the solution to a crisis — and a research plan that treats it that way will produce numbers for a product you shouldn't build as described. If you haven't already, work through the fuller guide to validating a mental health app idea alongside this one — market sizing is one gate in that process, not the whole thing, and the acuity mapping there goes deeper than this article does.

Care context is where your product sits relative to formal treatment: standalone wellness tool, adjunct-to-therapy companion, or a product designed to be prescribed or reimbursed as part of clinical care. This single decision determines which competitors you're actually up against and who you'll eventually ask to pay — both covered below.

Write one sentence naming your condition, your acuity band, and your care context before moving to Step 1. If you can't write that sentence yet, the research below will be measuring the wrong thing.

Step 1: Define Your Segment in Specific, Checkable Terms

A usable segment definition names a condition, an acuity band, a care context, and a geography — vague enough and every subsequent number you pull will be uncheckable by anyone but you.

Compare two segment statements:

• Vague: "People with anxiety who want a mental health app."
• Specific: "Working adults with mild-to-moderate, non-crisis anxiety symptoms, not currently in therapy, in the US, looking for a self-guided coping tool."

The second version tells you exactly which competitors matter (self-guided consumer tools, not clinical DTx), which data sources are relevant (general population surveillance data, not treatment-seeking cohorts), and which payer route applies (consumer, not necessarily insurer). The first version could describe five different products with five different markets.

This is also where the wellness-versus-clinical fork matters most. A prevention-focused, self-guided tool and a product designed to carry clinical evidence and pursue reimbursement are different businesses with different research questions, different regulatory exposure, and different timelines to revenue. If you're not sure which side of that line your idea sits on, digital therapeutics vs. wellness apps walks through the distinction in detail — read it before you finalize your segment, because it changes what "competitor" and "evidence" mean for the rest of this process.

Step 2: Size Demand Without Borrowing a Number You Can't Verify

Size demand by combining general population data with your own primary signals — never by repeating a prevalence statistic you found in someone else's pitch deck or blog post without checking its source, year, and definition.

Nearly every mental health startup deck cites the same handful of headline figures, usually without a citation. That's a red flag, not a shortcut: a number stripped of its source, survey year, and exact definition (lifetime prevalence? past-year? past-month? diagnosed vs. self-reported?) tells you almost nothing about the segment you defined in Step 1. Two studies measuring "anxiety" with different instruments and different time windows can produce very different figures — that's a data-quality problem, not a market-size answer.

Instead, build your own view from primary sources you can actually check:

• National public health surveillance data — agencies like the CDC and equivalent bodies in other countries publish population health survey data you can query directly and filter by the demographics that match your segment, rather than trusting someone's summary of it.
• National mental health and substance use research bodies — organizations such as SAMHSA and NIMH in the US, or the WHO for international context, publish underlying survey data and methodology, not just headline soundbites.
• Peer-reviewed clinical literature — published studies in psychiatry and clinical psychology journals give you prevalence and treatment-seeking data with a stated methodology you can evaluate, rather than take on faith.
• Primary research you run yourself — search volume for the specific problem language your segment uses, app store review mining on adjacent tools, and direct conversations with people in your segment. The complete guide to customer research for founders covers how to run those conversations without leading the witness, which matters even more here than in a typical consumer category.

Treat every borrowed figure as a hypothesis to check against its primary source, not a fact to cite. If you can't find the original survey, sample, and year behind a statistic, don't repeat it — note the gap and move on to a source you can verify.

Step 3: Map Competitors by Category, Not by Feature List

Group competitors into the category they actually compete in — consumer wellness, employer-distributed, provider-facing, or payer-covered — before you compare a single feature, because a meditation app and a prescription digital therapeutic are not fighting for the same user or the same purchase decision.

Founders coming from a typical SaaS background default to a feature-parity spreadsheet: who has journaling, who has AI chat, who has a therapist directory. That comparison is close to useless in this space because two products with near-identical features can be in completely different competitive sets if one is downloaded directly from an app store and the other is distributed through an employee benefits portal or prescribed by a clinician.

Build your competitor map in two passes:

1. Same category, same acuity band. Who else is trying to reach your exact user, through your exact channel, at your exact care-context level? This is your real, near-term competition.
2. Adjacent category, same condition. Who addresses the same underlying problem through a different route — a workbook-based app versus a coaching subscription versus a peer-support community? These aren't direct competitors, but they're where your prospective users' time and money currently goes, and understanding what keeps people using (or abandoning) those alternatives tells you what you're actually up against.

For each competitor, note not just what they built but how they're distributed and priced — a $10/month consumer subscription and a per-employee-per-month enterprise contract imply entirely different go-to-market motions, even for similar-looking products.

Step 4: Identify Who Actually Pays

Name the payer explicitly before you build anything, because "the user" is only one of at least four plausible answers, and each one changes your product, your evidence bar, and your sales cycle.

In consumer software generally, the user and the buyer are usually the same person. In mental health, that's often not true. A working adult might use an app that their employer paid for as a benefit, that their insurer covers as part of a health plan, or that their therapist recommended as a companion to sessions — with the founder never having built a way to sell to any of those three parties directly.

Work through each candidate payer and ask what it would take to close them:

• The user, directly. Lowest evidence bar, fastest sales cycle, but the hardest category to stand out in and the most churn-prone.
• The employer, via an employee assistance program or wellness benefit. Requires HR/benefits buy-in and usually a per-employee-per-month contract, with utilization data expected at renewal.
• The provider or health system, as a tool clinicians recommend or integrate into care. Requires workflow fit and often technical integration, with a much longer sales cycle.
• The payer or insurer, as a covered or reimbursed benefit. Requires the highest evidence bar — often something approaching clinical trial data — and the longest, most uncertain path to revenue.

Step 5: Run a Differentiation Check Against Your Actual Buyer

Test your differentiation against the specific competitors your actual payer will compare you to — not against every app in the category — because a feature edge that matters to a consumer buyer can be irrelevant to a benefits manager or a health system.

Take your segment definition, your competitor map, and your identified payer from the steps above, then ask: if this buyer is evaluating three options side by side, why do they pick yours? The honest answer usually falls into one of three buckets — a distribution advantage (you can reach this exact segment more cheaply or credibly than incumbents), an evidence advantage (you can prove benefit that competitors can't), or a workflow advantage (you fit into how this buyer already operates more smoothly than alternatives).

If your answer is a feature list, keep pushing. Features get copied inside a release cycle or two. A distribution channel, a body of evidence, or a workflow integration takes competitors much longer to replicate — and if you can't name which of the three you're building toward, you likely don't have differentiation yet, you have a longer to-do list than your competitors.

Consumer, Employer, Provider, and Payer Routes Compared

The table below lines up the four payer routes from Step 4 against the dimensions that actually change your business — who buys, how long it takes, and what you have to prove.

Read the table left to right as a trade-off, not a ranking: the routes that are fastest to first revenue (consumer) ask the least of you upfront and give you the least defensibility later, while the routes that take longest to close (payer) demand the most evidence but are the hardest for a later competitor to dislodge you from once you're in.

Common Mistakes: Treating Mental Health as One Market

The mistakes below share one root cause — collapsing a category that actually contains several distinct markets into a single "mental health app" market, then researching, sizing, and building for an audience that doesn't correspond to any real buyer.

• Citing a headline prevalence statistic with no source. If you can't name the survey, the year, and the exact definition behind a number, it isn't evidence — it's a number that sounds like evidence.
• Sizing the whole category instead of your segment. A total addressable figure for "mental health apps" tells you nothing about demand for your specific condition, acuity band, and care context.
• Comparing features across mismatched categories. Benchmarking a consumer wellness app against a payer-reimbursed clinical product measures two different businesses with the same ruler.
• Assuming the user is the payer. Skipping employer and payer research because "people will just pay for it themselves" is a guess, not a finding — test it explicitly per Step 4.
• Researching demand for an acuity band you can't safely serve. If your product isn't equipped to handle acute crisis, don't size or market to that segment as if it were — that's a viability problem your research has to surface, not paper over.
• Skipping direct conversations in favor of desk research alone. Public data tells you the shape of the population; only talking to people in your segment tells you whether your specific product concept solves a problem they'll act on.

Key Takeaways

• Segment by condition, acuity, and care context before researching anything else — these three decisions determine which data, competitors, and payers are even relevant.
• Any acuity band that includes crisis-level users requires a visible path to real emergency help (988 in the US, or 911/local emergency services) — this is a research and viability question, not just a product-design afterthought.
• Never repeat a prevalence statistic you can't trace to its source, year, and definition — build demand estimates from public health surveillance data, clinical literature, and your own primary research instead.
• Competitors group by category — consumer, employer, provider, payer — not by feature list. Two products with similar features can be in entirely different competitive sets.
• Name the payer explicitly. The user, the employer, the provider, and the insurer are four different buyers with four different evidence bars and sales cycles.
• Differentiation has to survive contact with your actual buyer, not with every app in the category — aim for a distribution, evidence, or workflow advantage, not a longer feature list.
• "Mental health app" is not one market. Every mistake in this guide traces back to treating it as though it were.

Frequently Asked Questions

How big is the mental health app market?

There isn't a single trustworthy answer to that question as asked, because "mental health app" spans dozens of conditions, acuity bands, and payer routes with very different sizes. Instead of quoting a headline figure, size your specific segment — one condition, one acuity band, one care context — using public health surveillance data and your own primary research, per Step 2 above. A number for the wrong segment isn't useful even if it's accurate for something.

Who actually pays for mental health apps?

It depends on the route: consumers pay directly for low-friction wellness tools, employers pay per-employee-per-month for benefits distributed through an EAP, health systems pay for tools that integrate into clinical workflows, and insurers reimburse products that clear a high evidence bar. Most mental health apps default to assuming the user pays; identifying the real payer for your specific product, per Step 4, changes what you build and how long it takes to close revenue.

How do I estimate demand without a market research budget?

Combine free public sources with primary research you run yourself: pull population data from national public health and mental health surveillance agencies, read the methodology behind any peer-reviewed prevalence studies you cite, and talk directly to people in your defined segment about what they currently do about the problem. None of this requires a paid research subscription — it requires checking sources rather than repeating them.

What's different about market research for a wellness app versus a clinical or reimbursed product?

A wellness app's research centers on habit formation, engagement, and consumer willingness to pay directly — a lower evidence bar and a faster path to revenue. A clinical or reimbursed product's research has to also answer whether you can generate evidence a payer or provider will accept, which usually means clinical input and a longer validation timeline before revenue. Deciding which path you're on, using the digital therapeutics vs. wellness apps comparison, should happen before you size demand — it determines which data and which competitors are even relevant.