Short version: Google's YouTube Data API v3 is free, and for most jobs that settles it. The catch is what happens when you want more. Google's data has two prices: free, or not for sale. There is nothing in between, and no amount of money moves a call from the second bucket to the first. Tapline sells the calls in that second bucket at $0.25 to $0.98 per thousand. If Google already gives you what you need, stop reading and keep using it.
What the data actually costs
Every Tapline call costs the same: 2 credits, or 3 for channel info. Google's cost swings between nothing and impossible depending on which call you make. So the comparison that matters is per job, not per plan.
| What you need | Google charges | Tapline charges |
|---|---|---|
| Video and channel metadata | Free, up to about 300,000/mo | $0.25 to $0.98 / 1,000 |
| Keyword search | Free, up to 3,000/mo (100 a day) | $0.25 to $0.98 / 1,000 |
| Transcript, video you own | Free, up to about 1,200/mo (40 a day) | $0.25 to $0.98 / 1,000 |
| Transcript, video you don't own | No: Not for sale, at any price | $0.25 to $0.98 / 1,000 |
| Most-replayed curve | No: Not for sale, at any price | $0.25 to $0.98 / 1,000 |
| Codec and resolution, video you don't own | No: Not for sale, at any price | $0.25 to $0.98 / 1,000 |
Read the middle column top to bottom and the shape of this is clear. Google is not expensive. Google is free until it is a brick wall. The bottom three rows have no price because Google does not sell them to anyone but the video's owner, which is the reason this page exists at all.
What a thousand calls costs on Tapline
One price per call, whatever the endpoint, so a transcript costs the same as a metadata lookup. The rate drops as the plan grows:
| Plan | Monthly | Requests/mo | Per 1,000 | Rate limit |
|---|---|---|---|---|
| Free | $0 | up to 250 once | n/a | 60 requests/min |
| Starter | $49 | up to 50,000 | $0.98 | 5 req/s · 5 concurrent |
| Growth | $199 | up to 500,000 | $0.40 | 25 req/s · 20 concurrent |
| Pro | $499 | up to 2,000,000 | $0.25 | 50 req/s · 50 concurrent |
Against the search cap the gap is wide. Google gives you 100 searches a day. The $49 plan gives you about 1,600 a day, 10 results each. If you are tracking keywords around the clock rather than testing, pick Growth. You will want it for the parallel requests more than the credits.
You can price this yourself before you spend anything. Pull a transcript without signing up, or send a live request and read the JSON that comes back.
Where Google costs nothing and you should stay
We sell a competing API, so read this with that in mind. Most people who go looking for an alternative to the YouTube Data API should stay on the YouTube Data API.
You pay nothing per call, at any volume, as long as you stay inside the quota. Most reads cost 1 unit, so 10,000 units a day is about 300,000 metadata calls a month. That is six times what Tapline's $49 plan includes, for free. If you mostly look up video and channel stats, paying us would be a waste of money.
Google's API is also the only one Google backs, and the only one that can write: uploads, playlists, live chat. A channel's private numbers are Google-only too. If your legal team needs data straight from Google, the choice is already made.
Use Google. Come back when you hit one of the three walls below.
What Google won't sell you
Transcripts for videos you don't own
This is the big one, and Google's docs are clear about it. captions.download is the only official way to get caption text. Its documentation says the method "requires the user to have permission to edit the video." An API key won't do. You need OAuth 2.0 with the youtube.force-ssl or youtubepartner scope. Ask for someone else's video and you get a 403, saying "the permissions associated with the request are not sufficient to download the caption track." captions.list won't help either. It also needs permission, and it only lists the tracks.
Every product that reads other people's transcripts is built on something else. Podcast search, video summarizers, search over conference talks, all of it. More quota won't fix it, and neither will money. This is a permission rule, not a limit you can raise.
Keyword search past 100 a day
Search gets its own quota, and it's small. From Google's getting-started guide: "Projects that enable the YouTube Data API have a default quota allocation of 100 search.list calls, 100 videos.insert calls, and 10,000 units per day combined for all other endpoints."
Paging makes it worse: every extra page of results is another search.list call against the same 100. If your product finds videos instead of looking up IDs you already have, this is usually where it stops. You can't batch around it. There is no bulk search.
Data only the owner can get
The most-replayed curve you can see in the player has no endpoint in the v3 reference at all. Codec, resolution, bitrate, and filesize do exist, in the fileDetails part of videos.list, but Google states "this data can only be retrieved by the video owner." In each case the data exists. The limit is who gets to ask for it.
Quota, briefly
Most reads cost 1 unit each, so the 10,000-unit day goes a long way: videos.list, channels.list, playlists.list, commentThreads.list are all 1. Captions are the expensive ones. captions.list costs 50 and captions.download costs 200, so even on videos you own, a full day buys about 40 transcripts. Search costs 1 unit but draws on that separate 100-a-day bucket, which is what usually runs out first.
One line in the same guide is easy to miss: "all API requests, including invalid requests, incur a quota cost of at least one point." So a typo in a video ID still costs you. On Tapline, failed requests cost 0 credits.
Can you just ask Google for more?
You can apply through Google's quota and compliance audit. You have to pass an audit against the YouTube API Services Terms of Service. Google doesn't publish how long it takes, or what it takes to pass. Requests do get denied, and there's a form for appealing.
You can't buy your way around it either. Google doesn't sell extra YouTube Data API quota at any price. And a bigger quota only raises limits. It won't give you captions for videos you don't own, and it won't add endpoints that were never there. Two of the three walls stay put.
What to use
Most teams end up running both, and that is what we would tell you to do. Keep metadata on Google's free quota. Buy only what Google won't sell.
Stay on Google when
- You look up stats for video and channel IDs you already have. 10,000 units a day is about 300,000 metadata calls a month, free. Do not pay anyone for that.
- You need to write, not just read. Uploads, playlists, and live chat only work on Google's API. Tapline only reads.
- You want your own channel analytics. Nobody but Google can give you a channel's private numbers.
- Your company needs data straight from Google. Google backs its own API and no one else's. If that is a hard rule where you work, you are done deciding.
Add Tapline when
- You need transcripts from videos you don't own. There is no official way in, at any price. It is why most people end up here.
- You discover videos by keyword. 100 searches a day ends most monitoring projects. Tapline Starter covers about 1,600 a day.
- You want the most-replayed curve or format listings. One is not in the v3 docs at all. The other is owner-only. Either way you need a third party.
- You want to raise a limit by paying. Changing plan takes a minute. A quota extension takes a compliance audit and can be refused.
Tapline reads public YouTube pages and returns clean JSON from ten documented endpoints. It is read-only, so it will never upload a video or moderate a comment. It is also independent: we are not affiliated with YouTube or Google, and they do not endorse us. Google's API is the official source and comes with Google's Terms of Service, including a rule that you refresh or delete most stored content within 30 days. That is a different trade-off, not a better one. Talk to your own lawyer before you build on it.
Every number about Google's API here comes from Google's own docs, as of 2026-07-27: the quota calculator, getting started, captions.download, and the quota and compliance audits guide. Google changes quota costs, so treat anything older than three months as stale. Tapline publishes this page and sells a YouTube data API. We have said where Google's is the better pick.
FAQ
Can I get YouTube transcripts with the official YouTube Data API?
Only for videos you own or manage. The captions.download method requires OAuth 2.0 and, in Google's words, requires the user to have permission to edit the video. For anyone else's video it returns a 403 with the message that the permissions associated with the request are not sufficient to download the caption track. This is a permission rule rather than a quota limit, so extra quota does not change it. As of 2026-07-27.
How many YouTube API calls do I get per day?
By default, 100 search.list calls, 100 videos.insert calls, and 10,000 units per day combined for all other endpoints. Most read methods such as videos.list and channels.list cost 1 unit each, so the 10,000-unit pool covers a lot of metadata. The 100-call search bucket is usually what binds first. As of 2026-07-27.
Why does my YouTube API quota run out so fast?
Usually search, or failed calls. Search has its own bucket capped at 100 calls a day and every page of results is a separate call, so paginating a few queries can exhaust the day in minutes while the 10,000-unit pool sits nearly untouched. Google also charges for calls that do not work: all API requests, including invalid requests, incur at least a one-point quota cost.
Can I buy more YouTube API quota?
No. Google does not sell YouTube Data API quota at any price. The only route above the default allocation is a quota extension request, which requires passing a compliance audit against the YouTube API Services Terms of Service and can be denied. A larger allocation also raises limits only; it does not grant caption access for videos you do not own.
Does the YouTube Data API return the most-replayed graph or video formats?
Neither is available to a third-party caller. The most-replayed curve shown in the YouTube player has no endpoint in the v3 reference at all. Codec, resolution, bitrate, and filesize do exist, in the fileDetails part of videos.list, but Google states that this data can only be retrieved by the video owner. As of 2026-07-27.
What is the best alternative to the YouTube Data API?
It depends which limit you hit. For transcripts of videos you do not own, for keyword search above 100 calls a day, and for most-replayed data, a dedicated third-party YouTube data API such as Tapline is the only route, because no official one exists. For video and channel metadata on its own, the free official API is generous and there is rarely a good reason to leave it. Many teams run both.