Name Combiner

Blend names into something new. Enter two names for a couple or a baby, or up to ten for a team or a band. Every result is scored for how easily a person can actually say it.

Names to blend

    Enter at least two names. Everything runs on your device and nothing is uploaded.

    Refine the results

    Different naming traditions allow different consonant combinations. Setting this stops the scoring penalising sounds that are perfectly normal in that language.

    Blending methods

    How to get good results

    Most name combiners cut each name in half and glue the halves together. That works perhaps one time in five. The rest of the time you get something like Jenmar or Chrilis, which no one would ever use, because cutting a name at the wrong point destroys the sounds that made it a name in the first place.

    This tool cuts at syllable boundaries instead, which is why the results are usable more often. But you will still get better output if you know what the controls do.

    Start with two names, not six

    The fewer names you enter, the more of each one survives in the result. Two names give the engine enough material to fuse properly. With six names, each contributes roughly one syllable, so the output reads more like a coined word than a blend. If you have six candidates, try them in pairs first and blend the winners.

    Let the score do the filtering

    The number on each card is not decoration. It combines a pronounceability measure, which checks the result against the consonant clusters English actually permits, with a blend quality measure, which checks that every name you entered left a recognisable trace.

    As a rough guide: above 85 and the name reads naturally enough that most people would assume it was real. Between 60 and 85 you have something usable that may need a spelling tweak. Below 50 there is usually a specific problem, and the small grey line under each name tells you what it is.

    Use the filters rather than scrolling

    If you need a name that starts with a particular letter, set it. If you are naming a company and need something that fits comfortably in a logo, cap the length at eight characters. Narrowing before you browse is faster than reading a hundred results.

    Syllable structure of the name Jennifer Jennifer divides into three syllables: jen, ni and fer. Each syllable has an optional opening consonant group, a vowel at its core, and an optional closing consonant group. Jen is j plus e plus n. Ni is n plus i with no ending. Fer is f plus e plus r. j e n n i f e r syllable 1 syllable 2 syllable 3 opening consonants vowel core closing consonants Consonants between two vowels go to the syllable that follows, as far as English allows. So it is jen-ni-fer, never jenn-if-er.
    How the engine reads a name before it blends anything.

    The six blending methods, and when each one works

    Each method is a structurally different operation, not a variation on the same one. You can switch them on and off individually under Refine the results.

    The six blending methods compared on one pair of names Aiden and Sophia blended six ways. Syllable splice gives Soden. Sound overlap gives Sophiden. Alternating gives Aiphia. Vowel swap gives Oidan. Head and tail gives Aiphia. Nested gives Aiphiaden. Aiden + Sophia, run through every method Each method is a different operation, not a variation on one. Syllable splice whole syllables, in turn So + den Soden Sound overlap fused at a shared sound Sophi + den Sophiden Alternating one syllable from each Ai + phia Aiphia Vowel swap frame kept, vowels borrowed _d_n + o, a Oidan Head and tail opening of one, end of other Ai + phia Aiphia Nested one dropped inside the other Ai…den + phia Aiphiaden
    Real output from the engine. Switch methods on and off in Refine the results to see the effect on your own names.

    Syllable splice

    Takes whole syllables from each name in turn. Aiden and Sophia become Soden or Sophiden. Because syllables are the unit people actually perceive, spliced names keep the most recognisable pieces of the originals. This is the method to leave on if you want the source names to stay obvious.

    Sound overlap

    Finds sounds the two names already share and fuses them at that seam. This is how the best-known real-world blends were built: Brad and Angelina share a bran transition that produces Brangelina. When two names have a natural overlap, this method finds it, and the result is usually the smoothest one on the page. When they share nothing, it produces very little, which is why it is worth keeping other methods switched on alongside it.

    Alternating

    Takes syllable one from the first name, syllable two from the second, syllable three from the first again, and so on. It works best when both names have a similar number of syllables. With a one-syllable name and a four-syllable name it tends to produce something lopsided.

    Vowel swap

    Keeps the consonant skeleton of one name and replaces every vowel with vowels from the other. Ross and Rachel produce Rochol. This is the most surprising method and the one most likely to produce something that feels like a genuinely new word rather than a mashup. It is also the one most likely to produce nonsense, so it pairs well with a high minimum score.

    Head and tail

    The traditional approach, but cut at syllable boundaries rather than at an arbitrary character count. Reliable and slightly conservative. If the other methods are giving you results that feel too strange, turn everything else off and use this one alone.

    Nested

    Drops the core syllable of one name inside the other, keeping both openings intact. It is the least predictable method and produces longer results, which makes it more useful for band names and fantasy names than for baby names.

    What actually makes a blended name work

    The difference between a blend people adopt and one they quietly ignore is almost never cleverness. It is whether the name survives being said out loud by someone who has never seen it written down.

    The first sound carries the most weight

    People remember the beginning of a name far more reliably than the middle. A blend that keeps one name's opening intact tends to feel like a real name, while a blend that scrambles both openings tends to feel invented. If you want the result to read as a name rather than a product, keep a strong opening consonant.

    Consonant clusters are where blends break

    English allows a specific and fairly small set of consonant combinations at the start of a syllable: br, st, thr, spl and a few dozen others. It allows a different set at the end. Blending two names frequently produces a cluster from neither set, and that is the moment a name becomes unsayable. The scoring here checks every cluster in the result against both lists, which is why some perfectly reasonable looking outputs score badly.

    Vowel balance decides the rhythm

    Names that alternate consonants and vowels evenly read fastest. Roughly 35 to 50 per cent vowels is where most comfortable English names sit. Below about a fifth, the name becomes a mouthful. Above about two thirds, it starts to feel formless. The rhythm portion of the score measures how often the name switches between consonant and vowel.

    Length has a ceiling

    Four to nine letters covers almost every name that gets used in practice. Longer blends get shortened by other people whether you like it or not, and you rarely get to choose which shortening sticks. If your result is eleven letters, look at what the natural nickname would be and decide whether you can live with it.

    What people use a name combiner for

    Couple names

    The most common use, and the one with the lowest stakes. Two partners' names blended into one, for a shared social handle, a wedding hashtag, a joint email address or a house account. Because nobody has to live with the result permanently, this is where the more adventurous methods pay off. Set the maximum length to about eleven and let vowel swap run.

    Baby names built from family names

    A common approach when parents want to honour two grandparents, or blend both their own names into a child's. Here the constraints are much tighter, because the child will carry the result for life. Keep the maximum to three syllables and nine characters, turn off vowel swap and nested, and check every shortlisted name against three practical tests: can a stranger spell it after hearing it once, can a stranger say it after reading it once, and what does the obvious nickname sound like.

    Business and brand names

    Blending a founder's name with a category word, or two founders' names together, is one of the oldest naming methods there is. Blended names have a real advantage for businesses: they are far more likely to be available as a domain and a trademark than a dictionary word. Cap the length at around twelve characters, then take your shortlist through trademark and domain checks before you get attached to any of them.

    Band, team and group names

    Blending the members' names is a reliable starting point when a group cannot agree on anything else. Longer results are fine here, so raise the length ceiling and leave nested switched on. Three to five inputs works well.

    Pet names

    Short is essential, because you will be shouting it across a park. Two syllables at most, eight characters at most, and preferably ending in a vowel, which carries further and is easier for an animal to distinguish.

    Usernames and handles

    A blended name is far more likely to be available on a platform than a common word, because you are effectively coining something new. Raise the length ceiling to fourteen, leave every method on, and generate a long list. You will need the volume, because you are checking availability across several platforms and most candidates will fail somewhere.

    Fiction and fan work

    Writers blend names to build characters who sound like they come from two cultures, and fandoms blend character names to label a pairing. Both benefit from generating a large list and reading it aloud rather than scanning it, because the ones that work are usually obvious the moment you say them.

    A short checklist before you commit

    • Say it out loud to someone who has not seen it written. Do they say it back correctly?
    • Write it down for someone who has only heard it. Do they spell it the same way you do?
    • Search the name on its own. Does anything unfortunate come up?
    • Check it in the other languages your audience speaks, especially if the name is short.
    • Work out the obvious nickname or shortening, and decide whether you can live with it.
    • For anything commercial, check the trademark register and the domain before anything else.

    How this tool differs from the others

    Most name combiners perform a single operation: take the first half of one name and the second half of the other. They cut at a fixed character position, they produce fewer than a dozen results, and they offer no way to tell a good result from a bad one.

    This one breaks each name into syllables using the Maximal Onset Principle, the same rule linguists use to decide where a syllable boundary falls. It then applies six structurally different recombination methods, and scores every candidate against English consonant cluster rules and vowel balance before ranking them. Results that are simply two names stuck together are filtered out by default, because those are not blends.

    You can read the full scoring method, including the cluster lists and the weighting, on the how our tools work page.

    Common questions

    How many names can I combine at once?

    Up to ten. Two names produce the cleanest blends because there are only two sets of sounds to reconcile. Three or four still works well. Past five, the tool switches to taking one syllable from each name rather than trying to fuse them, because a genuine ten-way phonetic blend would be unpronounceable.

    What does the number on each result mean?

    It is a 0–100 score combining two things: how easily the name can be said aloud, and how well it actually blends its sources. A name that scores 90 uses sound combinations English speakers already know, and draws recognisably on every name you entered. A name in the 40s usually has an awkward consonant cluster or leans far too heavily on one input.

    Does the tool store the names I type?

    No. The entire blending engine is JavaScript that runs inside your browser. Nothing you type is sent to a server, logged, or saved anywhere. If you close the tab, it is gone. The only exception is the shareable link, which puts the names in the URL so you can send them to someone, and that only happens when you press Combine.

    Can I use a combined name for a business?

    The name itself is yours to use, but a blended name being available as a word does not mean it is available as a trademark or a domain. Before committing, search your national trademark register, check the domain, and check the handle on the platforms you care about. A name can be perfectly original and still collide with an existing registered mark in your industry.

    Why do some results look nothing like the original names?

    The vowel-swap method deliberately keeps only the consonant frame of one name and borrows all its vowels from another, so Ross and Rachel become Rochol. It produces the most unexpected results. If you want blends that stay closer to the source names, turn off vowel swap and leave syllable splice and sound overlap switched on.

    Can I get shorter results?

    Yes. Open Refine the results and lower the longest-result setting, or reduce the maximum syllable count to two. Shorter blends tend to score higher for pronounceability but lower for blend quality, because there is less room for both names to leave a trace.