Baby Name Search
Search 2,000 baby-name entries using text, usage, origin, style and meaning cues. Read qualified profiles and compare the language history behind each result.
Openbaby name tool
Use Search and Generate correctly: understand dropdown matches, suggestion scores, text-query limits and the current local shortlist preview.
Search 2,000 baby-name entries using text, usage, origin, style and meaning cues. Read qualified profiles and compare the language history behind each result.
OpenType a baby-name shortlist and compare reasons for each finalist. Use practical family discussion notes; entries are an unsaved preview with no voting or sharing.
OpenContent revised 2026-11-02 · Evidence and limits are stated below.
The workbench above has three modes: Search, Generate and Shortlist. Search narrows the supplied catalogue; Generate ranks suggestions using selected catalogue labels; Shortlist displays a locally typed list. These are different tasks. If you are looking for a particular spelling or word, choose Search. If you want a few candidates that match some of your preferences, choose Generate. Use the result as a starting list, then verify the facts relevant to your family.
The field labelled “Name or meaning” is visible in both Search and Generate, but the current Generate calculation does not use that text. Typing Olivia or light while Generate is active will not make the suggestion list search for that word. Switch to Search to use the text field. This distinction is easy to miss because the controls share a layout; it matters when a result appears unchanged after you type.
Generate uses the Gender, Origin, Style and Meaning dropdowns. It does not offer a first-letter, surname, syllable-count or pronunciation filter in this workbench. Sound remains a check you make with the intended full name. The displayed suggestions come from the supplied catalogue rather than newly invented names. The same catalogue and selections produce the same ranking; this is not a random-name button or a forecast of what will become popular.
A suggestion can be useful even when it matches only part of your selection. That is why Generate and Search must be interpreted differently. Generate awards points for matching selected labels and can retain partial matches. Search requires all active filters to match. When a specific feature is essential, use Search to narrow candidates, then examine the source and spoken form before choosing.
| Condition in Generate | Points | Important boundary |
|---|---|---|
| Matches selected Gender usage | +24 | Girl/Boy also accept entries labelled Unisex; Unisex requires that catalogue label. |
| Matches selected Origin | +10 | Matches a catalogue origin label, not a new etymology investigation. |
| Matches selected Style | +12 | Style labels are descriptive browsing categories. |
| Matches selected Meaning | +16 | Uses the catalogue meaning text or meaning-keyword labels. |
| Catalogue popularity band is top-10 | −7 | An existing catalogue band; not a verified latest annual ranking. |
| No dropdown preference selected | +1 | Starter score; the top-10 subtraction still applies. |
For an illustrative candidate matching Gender and Origin but not the other selected labels, the starting match contribution is 24 plus 10, or 34. If its catalogue band is top-10, the score becomes 27. Another candidate matching all four dropdowns can receive 62 before that subtraction. These are worked arithmetic examples, not claims that a particular named child or catalogue entry has that score. A larger number means more of this calculation’s weighted matches, not a better life outcome.
Only candidates with a score greater than zero are retained, and Generate shows at most five. Results are ordered by descending score; equal scores are ordered by the name’s locale-based alphabetical comparison. With no dropdown preferences selected, ordinary starter candidates receive one point, while a top-10-band candidate receives one minus seven and is excluded. The phrase “balanced starter suggestion” is a tool label, not evidence of balanced gender use or a scientifically optimized recommendation.
The reason line displays at most two reasons even when more conditions contribute. Do not assume the visible two phrases are a complete score breakdown. Likewise, a candidate listed after selecting several preferences may satisfy only one or two because Generate adds points rather than enforcing every selection. Switch to Search if an AND-style filter is what you need, then verify the matched catalogue labels in the relevant source.
The popularity subtraction uses the catalogue’s existing band. It does not fetch a current official dataset when you change a control. Do not read “very familiar” as a verified 2026 local rank or treat a missing name as proof that it is rare. The historical trends guide and SSA evidence worksheet show how to check popularity as a separate, dated question.
In Search, the free-text query looks for a case-insensitive substring in the name, meaning, short description, origin labels, style labels or meaning keywords supplied with each profile. It is not an advanced query language. Quotation marks, Boolean operators and regular expressions are not documented controls here. Start with a straightforward spelling or word, observe the result, and then add one dropdown when that helps answer your question.
The active Search dropdowns are combined: a candidate must satisfy the query, the selected gender rule, the origin, the style and the meaning condition. Origin and style use catalogue-label matching; the meaning condition checks the supplied meaning text or keywords. The Gender rule treats catalogue-labelled unisex entries as eligible for Girl or Boy as well, while choosing Unisex specifically requires that label. These rules describe the tool’s labels, not everyone’s identity or a complete history of use.
Search results are ordered first by the catalogue popularity bands and then by the name’s alphabetical comparison. The page previews a limited number of cards and may show a heading such as “8 of … matching names.” That heading distinguishes the cards currently displayed from the full number of matches in the supplied set. It does not mean eight is the size of every search result or that the preview is a current popularity leaderboard.
A card may link to a published name profile, while a catalogue-only preview can remain static. A non-clickable suggestion is not a broken promise of a full profile; it can be an entry without a published profile route. Keep an external source record for an important candidate rather than assuming every name in the preview has a fully developed source-linked article. For exact histories that overlap in spelling, the origin guide explains how to select the relevant branch.
| Pass | What to do | What to record |
|---|---|---|
| 1: Explore | Choose one dropdown in Generate and observe the candidate list. | Which preference you selected; appealing exact spellings. |
| 2: Narrow | Switch to Search; add essential filters one at a time. | Which condition removes a candidate and whether it was essential. |
| 3: Verify | Open a published profile or a suitable language source; test the full name. | Source branch, intended pronunciation, family priority and unanswered question. |
Begin with one preference that genuinely matters. Adding four dropdowns before you see any candidates can make it hard to understand why a result appears. After each selection, note the active controls and the changed result. In Generate, look for partial matches rather than assuming every suggestion satisfies the entire set. If you want a particular word in the meaning, try it in Search and examine where the text match came from.
If Search returns no matches, remove one optional condition and observe what returns. A zero-result search tells you about the supplied catalogue and the combination of controls; it is not proof that no suitable name exists in the world. If Generate returns fewer than five suggestions, the positive-score rule and available catalogue labels may be limiting the list. Do not manufacture a factual origin or meaning just to make a candidate fit a dropdown.
Once you have a few candidates, pause the tool and check them fairly. Say each with the complete surname, keep any diacritics and surname parts, and record the pronunciation your family intends. Use the initials and surname checklist for written combinations and the rhythm guide for spoken comparisons. A short spelling is not automatically a one-syllable name.
Keep facts and preferences apart. A name may match the tool’s style label but have an uncertain etymology; another may have a well-documented history while failing a personal sound preference. Neither situation is resolved by adding points. The complete naming workflow helps you turn candidate discovery into an actual family decision, with “pending” notes for checks you still need to complete.
Shortlist accepts names separated by a comma or a new line, trims surrounding spaces and displays each distinct typed option. Its meters are fixed at zero in the current implementation. There is no vote button, share link or collaborative vote collection in this workbench. Treat it as a local list preview rather than a working family voting service. A list that looks like a poll does not establish that other people can access it or cast votes.
The typed search, dropdown selections and poll list are held in the current component’s browser-page state. This component contains no submission, network request or persistent-storage call for those interactions. Opening another route can make ordinary page requests, and reloading the page starts the component again. If a note matters, copy it to your own record before leaving. Do not rely on the workbench as durable storage for an important shortlist.
The downloadable guides and blank printable sheets are public resources shared with every reader. Your filled-in local document, screenshot or paper copy is separate from that public blank file. If you send your completed notes to another person, decide what you want to include and check the destination. The privacy explanation distinguishes local tool interactions, normal website requests and files you choose to save or share.
Before ending a session, keep the exact spelling, the preference that led you to it and one next check. You can use the five-candidate printable sheet without treating a tool score as a final verdict. The tool can help you find possibilities; source evidence, intended use and your family priorities are what make those possibilities ready to compare.
Switch to Search to use the Name or meaning text field. Generate does not use that query.
No. Generate adds weighted matches and can retain partial matches; Search requires all active conditions.
24 + 10 − 7 = 27. The score is tool arithmetic, not a percentage or a verified current rank.
No. This implementation displays typed options with fixed zero meters and does not collect votes.
No. It describes that supplied catalogue and filter combination. Remove an optional condition and check a suitable source.
Open the UTF-8 reference notes and explained checks. Save your own candidate record separately.