Songtradr Review: The Commission Maths, the Credits, and the AI Question

Songtradr is one of the largest music licensing marketplaces in the world, and the two questions our inbox actually asks about it are narrow: what does it take from a placement, and will it accept AI-generated tracks. The short answers are that the commission ranges from roughly 20% to roughly 60% depending on which tier you pay for, and that yes, it will consider AI material — with no explicit ban, judged case by case.

That second answer makes Songtradr unusual. Most large libraries have picked a side by now, and a majority of the premium end has picked "no". Songtradr has not published a prohibition, and its stated position is that a track is assessed on production quality and on whether you can demonstrate clear commercial rights to it. For anyone building an AI catalogue, that is the single most important fact on this page.

This review is about Songtradr specifically: what it is, what the tiers cost, what the commission does to a real placement fee, how the credit-based submission system behaves when you have a lot of tracks, and where the platform is genuinely opaque. The wider question of which sync company to choose belongs to our sync licensing companies comparison, and how a sync deal is cleared at all belongs to our sync licensing guide. We will not restate either.

What Songtradr actually is

Songtradr is a business-to-business licensing marketplace. On one side sit rights holders — several hundred thousand artists, producers and songwriters since its global launch in 2016. On the other sit brands, agencies, production companies and platforms that need music under picture, including names at the scale of Google, Disney, Netflix and Amazon.

Between those two sides the platform runs four things: a searchable catalogue with machine-generated metadata, a Music Wanted board of open briefs you can submit to, a team of human sync agents who pitch catalogue tracks against incoming requests, and a distribution product that puts the same tracks on streaming services.

The company was built partly through acquisition, and the one most people recognise is Bandcamp. Worth knowing mainly because it means you are licensing through a consolidator with an ownership history, not a small independent library. Our Bandcamp alternatives page covers the direct-sales side of that story.

The channel that matters most is the least visible one. In composer reporting, placements far more often arrive because an agent found the track in the catalogue than because a Music Wanted submission was selected. One reviewer described every placement they received as arriving unsolicited, with none of their board submissions ever chosen. That is a structurally different platform from the one the submission interface implies, and it changes what you should optimise for: metadata and catalogue depth, not submission volume.

The tiers, and the discrepancy we have to flag

Songtradr's current published structure is two paid tiers.

Two honest caveats before you use that table.

First, the free tier is gone. Older reviews — including well-regarded ones still ranking — describe a free plan with a monthly credit allowance and a 40% commission. That plan has been eliminated, and distribution royalties now flow entirely to artists on the paid tiers.

Second, our sources conflict on how the free tier's split was even framed. One describes the free tier as taking 60% commission; another describes it as the artist keeping 60% of sync fees, with paid tiers reaching 80–100%. Those are different claims about the same number, and we cannot reconcile them from published material. We are reporting the current Lite/Pro structure above, flagging that the historical figures circulating online are unreliable, and stating the obvious operating rule: read the live terms on the platform before you subscribe. Songtradr has revised its tiers more than once, and a percentage quoted in a review — including this one — is a snapshot.

The commission arithmetic, worked plainly

Percentages are easy to skim past. Money is not. Here is the same $2,000 placement fee under every split that has appeared in Songtradr's history and in the wider market.

The gap between the top and bottom of the Songtradr range is $800 on a single placement — more than sixteen years of Pro subscriptions.

Which leads to the arithmetic that decides the tier for you. Pro costs $30 more per year than Lite and returns 40 percentage points more of every fee. Break-even is where $0.40 of gross licensing income equals that $30 difference: about $75 of gross sync earnings a year. Above that, Pro wins, and it wins by more with every dollar. A single $200 library placement — the low end of the observed range for a corporate or creator use — already puts you $50 ahead on Pro.

So Lite is a strange product. It is not a trial, because it costs money. It is not economical, because it is beaten by Pro at trivial earnings levels. Its only rational use is as a cheap way to keep a catalogue listed while you decide whether the platform does anything for you at all, and even then you should upgrade before a deal closes rather than after.

For context on what those fees look like in the first place: library and marketplace placements commonly land between $20 and $200 per use, direct licensing to independent creators runs from around $25 to $500, and placements above $1,000 exist but are rare and are not what an unrepresented catalogue should plan around. Our AI music income breakdown sets that against streaming, and our streaming royalties page covers the per-play side.

The credit system, and what it does to a large catalogue

Submissions to the Music Wanted board are metered. Rather than uploading against every open brief, you spend credits, and historically a single submission has cost between 5 and 20 credits depending on the opportunity type. Tiers grant a periodic credit allowance plus a small number of free submissions; the exact allowances have changed across the platform's revisions, which is the same caveat as above.

The consequence is worth stating carefully, because it cuts against the instinct of anyone who has built a catalogue with generative tools. More tracks do not buy you more shots at the board. Credits do. If you have 300 finished tracks and a monthly allowance that funds a handful of submissions, your catalogue size is irrelevant to that channel — you are picking a small number of briefs and choosing which tracks fit them.

There are only ten to fifteen opportunities posted in a typical month anyway, and selection rates are not published. So the board is a lottery with an entry fee, and the sensible way to treat it is as a supplement rather than the strategy.

The strategy is the catalogue itself. Since agents search on metadata, the returns concentrate in tagging quality: mood, tempo, key, instrumentation, genre, use case, whether there are vocals, and whether an instrumental version exists. Volume helps only insofar as it is discoverable volume. A 300-track catalogue with lazy tags is smaller, functionally, than a 40-track catalogue tagged properly. Register your recordings properly too — our ISRC guide covers why cue sheets need them, and ASCAP vs BMI covers the performance-royalty back end that a broadcast placement generates on top of the sync fee.

The AI policy, read closely

Here is where Songtradr differs from most of the large libraries, and it is the reason this page exists on this site rather than a general music-business one.

Songtradr publishes no explicit ban on AI-generated music. Acceptance is decided case by case, and the two criteria that surface consistently are production quality and clear commercial rights. The platform's own product is built on AI — its metadata tagging generates thousands of descriptive data points per audio file rather than relying on human tagging — which is at minimum a cultural signal about where the company sits on machine involvement in music.

Three things follow from that, and they are not all in your favour.

It is a genuine opening. Compare it with the premium end of the market, where AI exclusions are written into the terms, and with the streaming side, where Deezer tags fully-AI tracks and excludes them from editorial and algorithmic surfaces, and CD Baby rejects fully-AI material as policy regardless of what a classifier says. A large library that will assess your track on whether it is any good is a materially better position than most of the alternatives.

"Case by case" means the quality bar does the filtering instead. A policy that does not exclude you by category will exclude you by comparison. Your track is being weighed against professionally produced library music by people who listen to submissions all day. Audible AI tells — smeared transients, unnaturally even dynamics, odd stereo imaging, metallic overtones on sustained material — get a track dismissed long before anyone reads a form. Our AI music mastering guide covers the production side of clearing that bar.

Clear commercial rights is a real requirement, not boilerplate. You need to be licensing output you generated under a tier that grants commercial use, and you need to be able to say so. Add the PRO position on top: ASCAP, BMI and SOCAN aligned in October 2025 to register partially AI-generated compositions while holding that wholly AI-generated ones are ineligible. A track that cannot be registered collects no performance royalties from a broadcast placement, which can quietly be worth more than the sync fee itself.

And the obvious point, stated once: a permissive AI policy is not permission to answer an AI-use question dishonestly. If a submission form asks, answer it. Songtradr is one of the few large libraries where the honest answer does not disqualify you.

What Songtradr does not tell you

A fair review has to include the parts we could not verify, because their absence is itself the finding.

Selection rates are unpublished. Nobody outside the company knows what fraction of submissions or catalogue tracks get placed. Third-party reviews cite figures below 5% of composers securing placements, sourced from testimonials rather than from data, and we are repeating that only to note that it is an estimate with no methodology behind it.

Rejections come with no feedback. Composers are not told why a track was not selected or what would improve the next one. For a platform charging both a subscription and per-submission credits, that is a meaningful gap: you are paying to enter a process that will not tell you how it works.

Timelines are long and vague. Selection can take months. Payment after a placement is reported in the range of six to nine months. Neither of those is unique to Songtradr — sync is a slow-paying business everywhere — but it should determine how you plan around the income.

Account and payment disputes appear in the record. There are complaints about accounts being terminated and funds withheld, with artificial streaming cited and limited disclosure or appeal. We have verified no individual case and cannot judge how representative they are against a base of hundreds of thousands of users — treat them as a reason to keep your own records, not as an established pattern.

The terms have changed more than once. The eliminated free tier is the clearest evidence. Assume any figure in this review can move, and check the live page.

Songtradr against the field, briefly

This is the short version by design — the full comparison lives on our sync licensing companies page, including which platforms accept AI submissions and which refuse.

The one-line positioning: Songtradr is an open, non-exclusive marketplace with mid-range commission on its Pro tier and unusual AI tolerance. AudioSparx occupies similar ground for AI material. Epidemic Sound and Artlist pay from pools or upfront buyouts and take rights positions in exchange. Musicbed, Marmoset and Pond5 sit at various points between curation and volume, and several of the curated ones do not accept AI-generated music at all.

Non-exclusivity is the feature that should decide this for most readers. Songtradr's standard arrangement does not stop the same tracks earning through streaming distribution or through another library simultaneously, which means the real comparison is not "Songtradr or X" but "how many non-exclusive channels can I service properly". Our music distribution services comparison covers the streaming half of that.

Before any library conversation: the file has to clear the gate

Here is the part that the rest of the internet's Songtradr coverage leaves out, and it is the thing this site measures.

Songtradr's openness to AI does not remove the distributor gate that comes first. If you are running the standard playbook — release the catalogue to streaming for credibility and metadata, then license the same tracks for sync — your files pass through a distributor's classifier before any supervisor ever hears them. In our 50-file benchmark, raw AI exports scored 0/50 at DistroKid, whose rejection threshold sits around 0.78 confidence; TuneCore's sits around 0.82, and raw files managed 3/50; CD Baby's policy rejects fully-AI material regardless of what its classifier reads.

Undetectr is the first and only AI music watermark remover — the one tool built to remove what distributors scan for rather than a repair suite pressed into the job. It clears six artifact layers in a single pass: the SynthID-class embedded watermark, the C2PA provenance manifest, the spectral fingerprint and the secondary layers underneath. It runs in the browser with nothing to install, takes under a minute per track, accepts MP3, WAV and FLAC from Suno, Udio and ElevenLabs Music, and masters to each platform's loudness specification in the same pass — Spotify's -14 LUFS, Apple's -16. SoundMatch checks for fingerprint collisions before release, which matters more in sync than anywhere else, because a collision surfacing after a campaign airs is a different order of problem from one surfacing on a streaming service. It is €39 once for unlimited tracks, with a €19 Starter tier carrying 10 credits, and it outputs for 150+ platforms. In our corpus it passed 49 of 50 tracks through production distributor classifiers; the single failure was an unrelated copyright flag on a remix, which no artifact remover should be expected to solve.

The framing matters as much as the tool. Artifact removal is release preparation. It is not a substitute for disclosing AI use where a contract asks. Those are two separate obligations, and Songtradr is precisely the platform where conflating them is unnecessary — its policy does not punish an honest answer. Run the file through Undetectr so it is technically clean, then answer the AI question truthfully on the form. Our Undetectr review is the sceptical long-form assessment with the full benchmark, and our methodology explains how the corpus is tested. Both carry the same caveat: a remover will not fix a weak song, and against a library assessing you case-by-case on quality, a weak song is the binding constraint.

Verdict: who should sign up, who should skip it

Sign up for Pro if you have thirty or more finished, well-produced tracks, you can tag them properly, and you are treating sync as one non-exclusive channel among several. Around $49 a year is recovered by a single modest placement, the AI policy is the most workable of any large library, and the agent-driven placement channel means a well-tagged catalogue keeps working without you doing anything. Go in with the expectation that the first placement takes somewhere between three months and a year, and that payment lands months after that.

Skip it if you have a handful of tracks, if you need income on a schedule, or if what you actually want is someone to actively pitch you — a subscription is catalogue listing, not representation, and the reporting from users who expected the latter is uniformly disappointed. Skip Lite specifically: it is beaten by Pro at any level of earnings worth having. And skip the whole category for now if your tracks are still being rejected at distribution, because a library conversation that starts with an unclean file does not start.

The fair summary is that Songtradr is a legitimate marketplace with real brand relationships, mid-range economics on the right tier, meaningful opacity in how it selects, and the most useful AI position of any library at its scale. That combination makes it worth $49 as a channel and worthless as a plan. Verify the current terms before you sign — the ones in this review were accurate when we wrote it, and this platform has changed them before.