YouTube Analytics export to dashboard
Watch time out of H:MM:SS and onto a chart.
Export the advanced view from YouTube Studio and drop the CSV here. Average view duration is a clock string like 4:12 rather than a number, so it is parsed into seconds before anything is charted. Watch time is already in hours, click-through rate is a plain number in a column whose name ends in a percent sign, and estimated revenue is provisional for the first several days, which the dashboard says out loud.
Prefer the full explorer? Open the app, or start from the generic dashboard builder.
One column is a clock and one column is a guess
YouTube Studio's export is generous with metrics and casual about types. Average view duration is written as 4:12 or 1:04:12 depending on how long people watched, which means the column is sometimes M:SS and sometimes H:MM:SS in the same file. Sort it as text and eleven seconds ranks above five minutes. Watch time, meanwhile, is a decimal number of hours, so the two duration columns in this file use different units and different types.
Then there is revenue. Your estimated revenue is exactly that, an estimate, and it moves for days after the fact as invalid traffic is filtered and advertiser billing settles. Pulling a report on the third of the month and treating last month's figure as final is a habit that produces small, persistent discrepancies against what actually lands. CPM has the same provisional quality and is only meaningful on rows that carried monetized playbacks at all.
How to export from YouTube Analytics
- Open YouTube Studio Go to studio.youtube.com and check the channel avatar at the top right. If you manage several channels through one account, the switcher is there and the export follows whichever channel is active.
- Go to Analytics, then Advanced mode Analytics in the left menu, then the Advanced mode button at the top right. The default overview does not export the per-video detail; advanced mode is the screen with the full sortable table on it.
- Choose your rows and columns Pick the Video tab for per-video rows or the Date tab for a daily series, then use the column picker to add impressions, click-through rate, revenue and CPM. The export contains exactly the columns you have chosen.
- Set the date range The picker offers presets and a custom range. Leave the last two or three days out if you care about revenue accuracy, since those figures are still moving and will change under you.
- Export current view as CSV The export icon at the top right offers comma-separated values, which downloads a zip containing chart data and table data. Take the table data file and drop it here.
The zip contains two CSVs: one for the chart, which is the daily series, and one for the table, which is the per-row breakdown. They answer different questions and both load here, but only one of them has a usable date column.
Try with sample YouTube Analytics data (the same columns, 92 rows, no upload needed).
What the dashboard shows
Every panel below is built from a column that is actually in the export, named the way YouTube Analytics names it.
- Views and watch time over time. Date as the axis with views and watch hours as two series, since a video that gets clicks without holding people looks completely different on the second line.
- Average view duration properly sorted. The clock strings parsed to seconds, so ranking videos by how long people stayed finally puts the right ones at the top.
- Impressions and click-through rate. How often the thumbnail was shown against how often it worked, which is the pair that tells you whether a title change is worth trying.
- Net subscriber change. Subscribers gained minus subscribers lost per video and per day, since the gained figure alone hides the videos that cost you an audience.
- Revenue and CPM by video. Estimated revenue ranked, with CPM alongside, and a note on rows recent enough that the number is still settling.
- Retention against length. Average view duration as a share of what you would expect, which is the closest this export gets to a retention curve.
Date,Video title,Views,Watch time (hours),Average view duration,Impressions,Impressions click-through rate (%),Subscribers gained,Subscribers lost,Your estimated revenue (USD),CPM (USD)What to know about a YouTube Studio export
- Average view duration is a clock string. M:SS for short watches and H:MM:SS for long ones, in the same column. Text sorting puts 0:59 above 10:04, and no spreadsheet will average the column for you.
- Two duration columns, two units. Watch time is decimal hours and average view duration is a clock string. Putting both on one axis produces a chart where one line is permanently flat.
- Estimated revenue is provisional. It updates for days afterwards as invalid traffic is removed and billing settles, and it is not final until the monthly close. Reporting a figure pulled on the second is reporting a draft.
- Click-through rate is a bare number. The percent sign lives in the column name, not in the values, so 4.7 means 4.7 percent. That is the opposite convention from most exports and it catches people going the other way.
- CPM only means something where there is revenue. Videos with no monetized playbacks show zero or blank, and including those rows in a CPM average drags it toward nothing. Filter to rows with revenue first.
- Dates are Pacific time. YouTube reports on America/Los_Angeles, so a video published at nine in the evening in London belongs to the previous day in every number in this file.
Parse the clock, keep the two duration units apart, and treat the last few days of revenue as a sketch. What is left is a rich file, and the combination of impressions, click-through rate and view duration in one table is genuinely hard to get anywhere else.
Frequently asked questions
Why will Excel not average my average view duration column?
Because the values are clock strings rather than numbers, and the format changes within the column: short watches are written as M:SS and longer ones as H:MM:SS. Excel reads some of them as times and some as text, and averages the mixture into nonsense. This page converts every value to seconds first, which makes sorting, averaging and charting all behave.
Is my estimated revenue figure final?
Not for the first several days, and not properly until the monthly close. YouTube filters invalid traffic after the fact and advertiser billing settles over time, so the number moves downward more often than up. If you are comparing months, wait until the following month has begun before pulling the figure, and treat anything from the last three days as provisional.
How do I calculate net subscribers?
Subtract Subscribers lost from Subscribers gained. YouTube reports them as two separate columns and leads with the gained figure in most of its own displays, which makes a video that brought two hundred subscribers and drove off a hundred and eighty look like a success. The net column is where you find out which videos are actually growing the channel.
What is the difference between views and impressions?
An impression is a thumbnail shown to someone on YouTube where it was at least half visible for a moment. A view is somebody actually watching. The ratio between them is the click-through rate column, and it is the number that tells you whether a title and thumbnail are doing their job, separately from whether the video itself holds an audience.
Why does my export have two CSV files in it?
YouTube Studio zips up a chart data file and a table data file. Chart data is the daily time series for whatever metric the chart was showing. Table data is the breakdown by whatever rows the table listed, usually per video. Only one of them has a genuine date column, so load the chart file for trends and the table file for comparing videos.
Is this connected to my YouTube channel?
No. There is no Google sign-in, no API access and no permission grant of any kind. You export the file from Studio yourself and drop it onto this page, where it is parsed by JavaScript inside your browser tab. Revenue figures for a channel are not something to hand to an unfamiliar web service, and there is no server here to hand them to.
Related tools
Open a YouTube Analytics export in one drop
92 rows across a quarter of uploads, clock-string durations, watch time in hours, a couple of videos losing subscribers and provisional revenue on the recent days.
Open the sample YouTube Analytics dashboard