Make an Original Sonic-Logo Sketch for a Commercial Pitch
Make an Original Sonic-Logo Sketch for a Commercial Pitch
There are two easy ways to end a commercial pitch. The first is to drop a sting everyone already knows onto the last frame, which makes the cut feel finished and gives the room nothing to discuss except how expensive the real thing will be. The second is to put any impact there — a whoosh, a low boom, a rising reverse — which also makes the cut feel finished, and which you cannot move, retime, or reuse anywhere else in the spot, because it isn't a mark. It's a full stop.
The third way takes about an hour and leaves you with something arguable. You compose a short original shape, in a plain sound first, and you find out where it goes. Three decisions stack up: the moment it occupies, the musical shape, and the sound it's made of. Keep those three separate in your session and each one can be heard on its own, which means each one can be changed on its own. That's the whole trick, and it's the reason a rough sketch is more useful in a pitch meeting than a polished sting nobody can edit.
The work below is a proposed exercise, not a demonstration. The commercial is invented, the timings are stipulated, and no audio was recorded or listened to in the writing of this piece. You supply the session.
Choose the moment the mark must occupy
A mark usually does one of three jobs in a spot: it introduces the brand, it closes an action already on screen, or it returns between passages. Those are different jobs with different requirements, and the same shape will not serve all three.
The first job wants the mark to arrive where nothing else is competing for attention. The second wants it to land on something the audience can already see happening, so the sound and the picture read as one event. The third — a recurring return — wants it short enough and consistent enough that hearing it twice doesn't feel like a loop, and it usually wants to be placed so it doesn't run into dialogue.
Before you write a note, write down what is already on the timeline. Surrounding speech, product sound and silence all constrain the mark more than any aesthetic preference does. Say the function out loud in specific terms. "It should land as the lid closes and leave the last line intact" is a brief. "It should feel warm and human" is not — and it's the kind of note that ends with a solo cello on the last frame because someone decided that cellos are warm.
Here is the invented spot we'll work through. It runs 20 seconds. A character packs a storage box and says a scratch line; the last word ends at 17.40. Their hands lower the lid; the lid contacts the rim at 18.20, and that contact is a real recorded product sound, a soft plastic thud about 0.15 seconds long. The picture then holds on the closed box from 18.20 to the end, with room tone under all of it.
So the mark's job is the second kind: close an action. Two constraints come with it. It has to relate to 18.20, and it has to leave the spoken line alone.
Notice what an arbitrary impact at the end of an edit actually is. It makes the cut feel finished, which is a real effect and a shallow one. It has no internal relationships, so it can't be moved 0.6 seconds earlier and still mean the same thing, can't be shortened into a recurring transition, and can't survive a change of edit. What you want at the end of the pitch is a small object with parts — which is the next section.
Compose the shape using a plain sound
Plain means plain. One oscillator, no reverb, no EQ, no compression, nothing that flatters. You are listening for shape, and any character you add now will hide whether the shape is any good.
Work in a small session where the decisions stay legible: one track per alternative, named, with the notes on a grid you can nudge in milliseconds. Working this way is the opposite of starting from a borrowed sting and adjusting it, which is how a pitch ends up with someone else's mark and a changed note.
Our stipulation: three pitches, E5, D5 and A4 — the E on the top space of the treble staff, falling a step to D, then dropping a fourth to A. In A minor that's the fifth, the fourth and the tonic: a shape that descends and comes to rest. That's a convention of Western tonal music, not a law of hearing, but it's a convention your room will recognise instantly, which is useful when you want the discussion to be about timing rather than about whether the tune is nice.
Now make two versions with the same three pitches and a different rhythm.
Version 1, measured. Attacks at 0.00, 0.25 and 0.50 seconds from the entry, each note 0.20 seconds long. The whole thing is over by 0.70.
Version 2, press and hold. E5 at 0.00 for 0.10 seconds, D5 at 0.12 for 0.10 seconds, then A4 held from 0.30 all the way to 1.30.
One consequential feature changed: whether the rhythm walks or presses. Everything else held still. If you also change the final pitch — drop it to D and leave the phrase hanging — you have run two experiments at once and you won't know which change did the work.
The two shapes have different endings by construction. Version 1 stops and leaves silence; the mark is a thing that happened. Version 2 is still sounding when it arrives at its landing point; the mark is a thing that's holding.
Here's where shape and placement start arguing with each other, which is the useful part. If the mark should land on the lid at 18.20:
- Version 1, ending at 0.70, has to start at 17.50 — 0.10 seconds after the last word ends.
- Version 2, whose held note ends at 1.30, has to start at 16.90 — 0.50 seconds before the last word ends.
Both land on the contact. One clears the line entirely and is gone in less than a second. The other gives you a sustain across the picture hold and puts 0.20 seconds of that sustain underneath the release of the last spoken word.
Take Version 2 as the provisional choice, because a 1.80-second picture hold is easier to sit on with a held note than with three short attacks, and because the two quick presses read as the small double motion of a lid going down. Keep Version 1 in the session at 17.50, muted, named, un-deleted. You will want it if the overlap turns out to matter.
Do not check your shape by putting it next to a famous brand sting to see whether it holds up. That comparison pulls you toward borrowing, and it answers a question you aren't asking. Compare your alternatives with each other.
Change the sound character while preserving the motif
Freeze the pitch and rhythm. Both of the following routes play the same three attacks and the same hold, in the same order, with the same entry point. Only the sound changes.
Route A is the plain tone you already have. It is the control, and its job is to be boring.
Route B is an original struck object. Record an empty metal tin tapped with a fingertip, close and dry, in a quiet room. Then the honest problem shows up: a struck tin rings where it rings, and it probably doesn't ring at E5, D5 and A4. You either move each hit slightly to reach those pitches — which changes the object's character a little, because pitch-shifting a transient is a sound decision, not a tuning chore — or you go find an object whose natural ring sits near the notes you need. What you don't do is let the object pick a different motif. If the tin sings a lovely B flat, that's not a reason to rewrite the shape; it's a reason to tap something else.
The two routes differ mainly in attack and decay, and that's exactly where the relationship you chose can get lost. A plain tone starts the instant you tell it to and stops when you tell it to. A struck object begins with a broadband click and rings on past the note's written end.
Those are quantities you can check, not moods. If the held A4 is attacked at 17.20 and the note ends at 18.20, a struck version that rings for one second puts its last audible moment around 19.20; a two-second ring reaches 20.20, which is past the end of the spot. At the front of the hold, the ring from the D5 and the click of the A4 can overlap each other enough that the falling step E–D–A stops being a step and becomes a smear with three bumps in it.
Keep the plain version in the same folder as the struck one, at matched level. If the struck version is louder, the comparison becomes a loudness contest and everyone will prefer it for the wrong reason.
Write down two separate things while you still remember them. What you composed: the three pitches, the two rhythms, the entry points, the date. What you recorded or licensed: who played it, what the object was, where and when, and which library or preset if you didn't record it yourself. If you can't say who owns a sound, don't put it in pitch materials.
A phrase from the practitioner account I have on this — Davide Carbone's What is Sonic Branding? on samplify.studio, dated 13 September 2026 — is worth keeping nearby: the composer describes separating musical shape from sound design and trying a short identity in a simulated use context. The note I have records that mechanism only. It's a practitioner's self-report; it reports no outcomes, no fees and no recall figures, and the linked recordings and videos on that page weren't part of what I have. Take it as a reason to run the three decisions separately, which is what you're already doing, not as evidence that any particular motif will be remembered.
Hear the sketch exposed and inside the proposed edit
Export two kinds of file, and don't skip the first kind.
The exposed version is the mark alone: no voice, no product sound, no music bed, a beat of silence before and after each alternative, all of them level-matched. This is the file that lets someone say "the second one" without having to disentangle the mark from the edit. An isolated export is also the only honest way to show what you composed, because a contextual version hides the composition under everything else.
The contextual version is the mark placed in the 20-second edit, with everything else untouched. Same picture, same scratch line, same lid thud, same room tone. If you find yourself re-mixing the spot to make the mark sit, you've stopped comparing placements and started producing, and the comparison is gone.
Then listen for four things: the entrance, the tail, the overlap, and the space left around speech.
The entrance is relative to the line. In our stipulated layout, Version 2 entered at 16.90 has E5 at 16.90, D5 at 17.02, and the held A4 from 17.20 to 18.20. The last word ends at 17.40. So the sustain and the release of the final word occupy the same 0.20 seconds. That is the specific thing the exposure pass is for, and I can't tell you the answer, because it hasn't been listened to. Maybe the note supports the line and the line's ending lands inside the note. Maybe 0.20 seconds of tone over a consonant is enough to blur the last word, in which case you move the entry later, or shorten the hold, and both of those are arrangement decisions rather than reasons to rewrite the melody.
The tail is relative to the picture. Both stipulated versions end at 18.20, and the picture holds to the end of the 20-second spot, so both leave the same 1.80 seconds of room tone after them. A struck version that rings changes that arithmetic, and the file worth checking is the one against the picture hold, not the exposed one.
There is also a collision at 18.20 that has nothing to do with music: the plastic thud is a diegetic product sound sitting at the exact instant the held note arrives. Two sounds at one moment. If the thud and the note's arrival fuse into a single event, the closing may read more strongly than either alone; if the note covers the thud, you've lost the product sound you spent money recording. You have options: start the hold 60 milliseconds early so the picture lands inside the note, move the mark 0.20 seconds late — clear of the thud's own 0.15 seconds — so the product sound speaks first and the music answers it, or let them sit together and accept that the moment is one composite hit. Choose by listening to the geometric difference, not by assuming.
Retain the session. One file with both shapes, both timbres and both entry points, so anyone can solo any layer, and so the version that loses today is still there when the client changes their mind in three weeks.
None of this tells you whether the mark works as a brand identity. The comparison lets collaborators judge concrete things: whether the mark lands on the same instant as the lid, whether the spoken line survives, whether the sound character is worth the specificity it costs, and whether the ending has room to breathe. It does not establish recognition, recall or distinctiveness. It does not clear anything with anyone. Trademark review, final mastering and audience testing are separate jobs, and as of this sketch none of them has happened.
Where the sketch stands
The provisional mark: E5, D5 and A4 in A minor, the two quick presses followed by a hold, entered at 16.90 so the hold lands on the lid contact, rendered once in a plain tone and once from a struck object you recorded yourself, with the measured alternative kept at 17.50. Four files: two exposed, two in context. One session, unflattened.
The question still open for the composer is the one the exposure pass exists to answer. Does 0.20 seconds of sustained tone underneath the release of the last word read as support, or does it read as a smudge over the line? If it's a smudge, you don't need a new melody. You need the other rhythm, entered later, landing on the same instant, giving up the sustain to keep the words. That is a decision you can hear, argue about and change — which is the point of sketching the mark instead of buying one.
Frequently asked questions
What three decisions should stay separate when sketching an original sonic logo?
The moment the mark occupies, the musical shape, and the sound it is made of. Keeping them separate in the session lets each one be heard and changed on its own.
How do you choose where the mark should land in the edit?
Write down what is already on the timeline — surrounding speech, product sound, and silence. Then say the mark's function in specific terms: introducing the brand, closing an action on screen, or returning between passages. In the invented spot, the job is to close an action, so the mark must relate to the lid contact at 18.20 and leave the spoken line alone.
Why compare two rhythms using the same three pitches?
Holding pitches constant isolates what the rhythm does. In the proposed exercise, both versions use E5, D5, and A4; Version 1 is measured and ends at 0.70, so landing on 18.20 means entering at 17.50. Version 2 presses and holds, ending at 1.30, so it enters at 16.90. The version 2 hold is provisional because it sits on the 1.80-second picture hold and its two quick presses read as the lid's small double motion.
What should be checked when changing the sound character from a plain tone to a struck object?
Freeze the pitch and rhythm, then listen for attack and decay. A struck object begins with a broadband click and rings past the note's written end, so a held A4 attacked at 17.20 and ending at 18.20 may still be audible at 19.20 after a one-second ring, or past 20.20 after a two-second ring. The ring from D5 and click of A4 can also overlap and turn the falling step into a smear. Keep the plain version at matched level, and do not let the object pick a different motif.
What can the exposed and contextual listening passes establish, and what do they not establish?
They let collaborators judge concrete things: whether the mark lands on the same instant as the lid, whether the spoken line survives, whether the sound character is worth its specificity, and whether the ending has room to breathe. They do not establish recognition, recall, or distinctiveness, and they do not clear trademark review, final mastering, or audience testing. The proposed exercise's timings are stipulated, and the piece states that no audio was recorded or listened to in its writing.