The release week data checklist
What to check on release day, day three, day seven and day fourteen, which numbers are not readable yet, and the two decisions release week data should actually inform.
Danny Angove
Co-founder, Backline · 13 August 2026 · 4 min read
In short
- Release day tells you about distribution and setup, not performance. Judging a track on day one is judging the delivery.
- Day three is the first honest read on quality, because save rate and follower conversion have enough volume to mean something.
- Day seven decides the paid spend question. Day fourteen decides whether the track gets a second campaign or joins the catalogue.
- Third-party streaming figures usually trail reality by around two days, so release week charts read as if the release happened later than it did.
- Write down your expected numbers before release day. Without them, whatever happens will be interpreted as roughly what you expected.
Release week generates more bad decisions per hour than any other period in a campaign, because the pressure to react is highest exactly when the data is least readable.
Here is a schedule that matches what the numbers can actually support.
Release week, by what the data can support
Illustrative- 1Before: write the expected numbersFirst week streams, save rate, new followers, link clicks. Four figures, two minutes.
- 2Day 1: plumbingLive everywhere, correct metadata, links resolving on mobile, your own tracking firing, playlist entries confirmed. No performance judgement.
- 3Day 3: ratiosSave rate, follower conversion per thousand listeners, early skip behaviour. The first honest read on quality.
- 4Day 7: the spend decisionAre the day 3 ratios holding at higher volume? Spend into markets already responding, or hold.
- 5Day 14: the verdictWhat stuck in audience terms, which markets over-responded, and whether this earns a second campaign.
Before release: write down the numbers you expect
One line, four figures: first week streams, first week saves or save rate, new followers, landing page clicks. Based on the last comparable release, adjusted for whatever is different.
This takes two minutes and it is the difference between analysis and rationalisation. Without a written expectation, any outcome reads as roughly what you thought would happen, and you learn nothing that carries into the next release.
Release day: check the plumbing, not the performance
Day one numbers are dominated by delivery, timezone boundaries and whichever playlists refreshed. What day one is genuinely good for is catching things that are broken while you can still fix them.
- Is the track live on every platform you claimed, with the right artist profile attached?
- Do your links resolve to the right destination on each platform, from a phone as well as a desktop?
- Is the metadata right? Correct spelling, correct features, correct artwork, ISRC present.
- Is your own tracking working? A click on your landing page should show up in your own numbers within the hour.
- Did the release actually enter Release Radar and any pitched playlist you were told about?
Any of those found on day one is a saved campaign. None of them requires looking at a stream count.
Day three: the first honest read
By day three you have enough volume for ratios to stabilise, and ratios are what tell you about quality.
Save rate. The proportion of listeners who saved it. Higher than your last release is the single best early sign, and Spotify for Artists is the place to read it because most third-party feeds do not carry it.
Follower conversion. New followers per thousand new listeners. If listeners are strong and this is weak, you are being played to a passive audience and the numbers will fade when the placement does.
Skip behaviour where you can see it. Where a platform exposes it, early skips point at an intro or a mix decision rather than a marketing problem, which is useful to know before you spend money widening the audience.
What is still not readable on day three: totals, chart position and anything involving geography, because early geography reflects playlist audiences rather than yours.
Day seven: the spend decision
Day seven is when paid media should be decided, and the question is narrow. Are the day three ratios holding at higher volume?
- Ratios holding and streams climbing: spend, and spend into the markets that are already responding rather than the ones you wish were.
- Ratios softening as volume rises: hold. You would be paying to widen an audience that is not converting, and the same money works better on the next release or on the markets that did convert.
- Volume low but ratios strong: this is the interesting case. A small, high-quality response is usually worth a modest, tightly targeted push, because you know the track lands when it reaches the right person.
By day seven you can also read geography honestly for the first time, once the initial playlist audiences have washed through.
The day seven spend question
IllustrativeIt is day seven of the Harbourlight single. Are the early ratios holding, and should we put the remaining budget behind it?
Data readget_streaming_statsget_smart_link_statsget_ads_performanceget_audience_geography
Volume is up and the conversion ratios are holding, which is the pattern that usually justifies spending.
Streams over days 4 to 7 are 41 percent above days 1 to 3, so the release is still growing rather than decaying.
Followers gained per thousand new listeners is 19 across the week, against 21 in the first three days. That is a small softening at nearly double the volume.
Landing page click-through is steady at 12.4 percent, with Spotify taking most of it.
Two markets are over-responding relative to the project's existing base: Dublin and Rotterdam.
If you spend, those two markets are where the evidence is. I would not widen to the largest markets yet, because nothing in this week's data says they are responding.
Illustrative example using a fictional project.
Day fourteen: the verdict, and the catalogue handover
Two weeks in, three questions.
Did anything stick? Compare monthly listeners and followers with the pre-release level. Streams decay, audience does not, so audience is the number that matters.
Which markets responded? Not the biggest markets, the ones that over-responded relative to your existing base. Those are the candidates for a live date, and they are the input to routing.
Does this get a second campaign? A track with holding ratios and a defined market deserves one. A track without them goes into the catalogue, and the catalogue is where most streaming volume lives anyway.
Streams decay, audience does not
IllustrativeThe two week lag warning
One practical point that catches people out. Third-party streaming data typically runs about two days behind reality, so on the morning of day three you are looking at day one. During release week that offset materially changes what you think you are seeing, and it is the main reason a release-day surge appears to land mid-week.
Check whether your tooling corrects for it. If it does not, mentally shift the chart, and never compare a corrected series against an uncorrected one in the same review.
What not to do in release week
Do not change the plan on day one. Do not send the label a first-day number without context. Do not compare a Friday release's first week against a Wednesday release's first week without adjusting the window. And do not judge a track on totals when the ratios are telling you something more useful.
Common questions
- What should I check on release day?
- Delivery rather than performance: that the track is live on every platform under the right artist profile, that metadata and artwork are correct, that your links resolve properly on mobile, that your own click tracking is recording, and that the release entered Release Radar and any pitched playlist. Day one stream counts are dominated by distribution and timezone effects.
- When can you tell whether a release is working?
- Day three for quality, using ratios such as save rate and new followers per thousand new listeners. Day seven for the paid spend decision, once you can see whether those ratios hold at higher volume. Day fourteen for the verdict on whether anything stuck in audience terms.
- Why does my release day spike appear a couple of days late?
- Because most third-party streaming feeds trail reality by around two days, so a chart built from them shows release day activity later in the week. Check whether your tooling shifts the series to correct for that lag, and never compare a corrected series with an uncorrected one.
Sources
- 1Spotify for Artists, Reviewed August 2026. Data in Spotify for Artists
- 2Spotify, 2026. Loud and Clear: takeaways
Danny Angove
Co-founder, Backline
Danny Angove is a co-founder of Backline. He works on the music business side of the platform: what managers, labels and independent artists actually need to see, and which numbers are worth arguing about.
Backline does this for the projects you run
Streaming, audience, social, advertising, website, search, ticketing and press data in one dashboard per project, with an AI assistant that answers questions about your own connected data. Invite-only.
Keep reading
Streaming
Playlist adds are not growth
A playlist add raises the number of people who could hear you. Whether it becomes an audience depends on two ratios, and most of the time it does not.
Danny Angove · 4 min read
Streaming
Comparing releases properly: align on day zero, not the calendar
How is the new single doing compared to the last one? Answering that requires re-anchoring each track to its own release date, and being honest about the tracks that cannot be compared at all.
Danny Starr · 4 min read
Data engineering
Reporting lag: the two days nobody accounts for
Third-party streaming totals describe a day that already passed. Uncorrected, every chart and every correlation is offset, and the error is invisible.
Danny Starr · 3 min read
