ASO When Your App Has Zero Downloads

A new app has no download history, no ratings stack, and no search-terms report worth reading. App Store search still has to decide whether you are eligible for a query, and the only text it can use from you is the metadata you submit. That is the name, the subtitle, and the keyword field. The product page then has to convert the few people who arrive, because nothing else about the listing is reassuring yet.

The work is narrower than a growth plan. Pick a query you can deserve with the build you have. Put it in the 30-character name. Spend the 100-character keyword field on neighbors, not on repeats. Make the first three screenshots answer that same query. Do not buy installs, do not fake reviews, and do not stuff the description. A practical ship list for the rest of launch is the App Store launch checklist. Field rules are on Apple’s Creating your product page .

What zero history actually changes

You cannot learn from App Analytics search terms you do not have. You cannot point at a rating count. You cannot assume a brand search exists for a name you invented last week. Practitioners watch downloads after launch, and Apple does not publish a weight for that attention, so “we will rank once we have velocity” is not a plan you can schedule. The listing has to be coherent on day one.

Coherent means one job, repeated in the places a shopper sees and in the field they do not. A plant app that names itself after watering, subtitles itself about meal plans, and screenshots a social feed is three products. With no reputation, shoppers will not give you the benefit of the doubt. They will tap the result that looks like the sentence they typed.

Pick a query you can deserve

Deserve means the current build does the job, and a screenshot can prove it without a paragraph of excuse. For a plant watering reminder, a deserved query is someone who forgets to water pots and wants a nudge per plant. An undeserved query is “plant identifier,” “garden design,” or “botanist,” if you do not identify plants or design gardens. Popular and undeserved is how a new app collects one-star reviews that then sit on an otherwise empty listing.

Find the wording in App Store autocomplete and in reviews of apps that already chart, not in your feature brainstorm. The method is App Store keyword research without a paid tool. If the phrase is long, do not paste the whole thing into the keyword field when the name already holds some of the words. That trap is long-tail keywords in 100 characters.

Put that query in the name

The name is 30 characters and it is the line searchers see. A brand-only name hides the job from the one place a new app can be obvious. Keep a short brand if you have one, and spend the rest of the 30 on the query. How that compromise works, including what fails recall, is in how to name an app. The subtitle is the second idea, not a second copy of the name.

For the sample app the name is “Dew: Plant Reminders” (20 characters). The subtitle is “When to water each pot” (22). A person searching for a reminder for their plants can read the result. The words already indexed are dew, plant, reminders, when, to, water, each, and pot. The keyword field does not get those words again.

Spend the keyword field on neighbors

Neighbors are the other words a plant owner types that this app can still satisfy: the kind of plant, the trip that causes the problem, the chore around watering. They are not five spellings of “reminder.” Apple combines words across the name, subtitle, and keyword field, and the field is 100 characters, comma-separated, with no spaces after commas. The rules are in the keyword field guide.

The field for Dew, 100 characters: succulent,fertilizer,repot,sunlight,vacation,thirsty,houseplant,moisture,monstera,cactus,soil ph,ivy. “soil ph” keeps its internal space. Nothing in that string repeats plant, reminder, water, or pot as its own token. Vacation matters because the reminder’s real job is the week you are away. If the app cannot store a per-plant interval, do not add a word that implies it can.

The deserved query sits in the name. The subtitle adds the per-pot schedule. The keyword field adds plant types and situations without repeating those words.

The first three screenshots answer the query

Screenshot one is a list of plants with the next watering date on each row, one of them due today. Screenshot two is the interval editor for a single pot, so “each pot” is visible. Screenshot three is a vacation pause, because that word is in the keyword field and the frame has to prove it. A botanical illustration with no dates answers a different search than the one you targeted.

Captions stay short and use the same verbs as the metadata. “Water the monstera today” beats “Never forget again.” You do not have reviews to do the convincing. The frame is the review. If you cannot make these three images from the real UI, the query is ahead of the product. Cut the query, not the honesty.

A small storefront is distribution, not a cheat

Launching in one smaller storefront first can be a sane way to get a few real users, fix the interval logic, and learn the words they type. That is a distribution and support choice. It is not a trick that makes Apple rank you higher everywhere else, and it is not a substitute for metadata. If you are deciding whether a limited release is worth the operational cost, read how to soft-launch an iOS app and treat it as a launch plan, not as an ASO exploit.

Whatever storefront you pick, localize the name, subtitle, and keyword field for that locale if the language is not the one you wrote. A US English keyword string pasted into a storefront you do not speak is not a soft launch. It is an unreviewed listing. Count the localized string the same way, in the character counter, because translations change length.

• One query in the name. Neighbors in the keyword field. No duplicates.
• Three screenshots that prove that query with the real UI.
• No bought installs, no fake reviews, no description stuffing.
• A smaller first storefront only if you will actually support those users.
• Analytics later, once there are search terms to read. Not invented volume on day one.