The twelve numbers worth knowing about every project
A short list of music industry KPIs that survive contact with a real week, why each one earns its place, the popular metrics that belong in a footnote, and where each reads in Backline.
Danny Angove
Co-founder, Backline · 15 August 2026 · 5 min read
In short
- Twelve numbers is about the limit of what a small team can actually hold in its head. Everything else is available when a question needs it.
- Every number on the list is either a level you can act on, a rate that tells you about quality, or a gain over a defined window. Cumulative totals belong in the header, not the review.
- Half the list is not streaming. Saves, sell-through, email engagement and search demand tell you about intent, which is what forecasts a next step.
- Monthly listeners on its own is a weak KPI. Paired with follower growth and save rate it becomes a good one.
- Pick the twelve once, write down how each is calculated, and stop renegotiating them mid-campaign. Backline puts most of them on one project view so the set holds together week to week.
Dashboards fail in two directions. Either they show four numbers, which is comforting and useless, or they show ninety, which is thorough and equally useless because nobody can hold ninety numbers in mind well enough to notice when one is wrong.
Twelve is roughly the number a manager can actually carry. Here is the twelve I would keep, why each one is on the list, and which popular metrics I would demote.
The other reason to keep the list short is that a KPI set only works if the numbers arrive together. Twelve figures spread across nine logins is not a KPI set, it is a morning. Backline exists to collapse that: one project view carrying streaming, audience geography, social, advertising, website and search, smart links, live shows and ticket pacing, press and newsletter, all on the same timeline and the same date window.
Twelve KPIs, and what each type of number is for
Illustrative| Type | Read it as | Where you read it | |
|---|---|---|---|
| Streams gained in window | Gain | Delivery, this period only | Backline, streaming |
| Monthly listeners | Level | Current audience size | Backline, streaming |
| Followers and subscribers | Level | Audience that opted in | Backline, social |
| Playlist reach and placements | Level | Capacity, not delivery | Backline, streaming |
| Save rate | Rate | Did they want it again | Spotify for Artists |
| Follower growth per 1,000 listeners | Rate | Earned or bought attention | Derived from two Backline levels |
| Repeat listener share | Rate | Depth of the audience | Spotify for Artists |
| Email open and click rate | Rate | Health of a channel you own | Backline, newsletter |
| Landing page clicks by platform | Gain | Where the audience actually goes | Backline, smart links |
| Branded search demand | Gain | Unprompted interest | Backline, Search Console |
| Sell-through per show | Rate | Live demand in a market | Backline, live shows |
| Press mentions, last 7 days | Gain | Why the week moved | Backline, news |
The four that describe reach
1. Streams gained in the window. Not the lifetime total. A window figure calculated as the total at the end of the window minus the total at the start, which is the only version that changes when you change the filter. If a tool shows the same number for 7 days and 90 days, it is showing you a cumulative total with a filter drawn on top. Backline reports the gain, so moving the window from 7 days to 90 changes the number, which is the entire point of having a window.
2. Monthly listeners. A level, not an accumulation, and the closest single proxy for current audience size. It moves with playlist placement, so read it alongside the next two rather than alone.
3. Follower or subscriber count across platforms. The audience that has explicitly asked to hear from you again. Slower moving than listeners and much harder to inflate.
4. Playlist reach and active placement count. Reach is the sum of followers of playlists you are currently on, which is capacity rather than delivery. Watch the direction, and treat a drop as a scheduling event rather than a catastrophe. Backline tracks both over time, so you see a placement arrive and leave rather than discovering a month later that reach halved.
Those four belong on one screen, which matters more than it sounds. Monthly listeners jumping while followers stay flat is a completely different story from both moving, and you only notice the difference when the two numbers are next to each other under the same date window.
The four that describe quality
These are the ones that separate a project that is growing from one that is being pushed.
5. Save rate. Saves relative to listeners tells you whether people who hear it want it again. Read it in Spotify for Artists and record the ratio with the rest of the twelve, because the comparison that matters is against your own previous releases rather than an industry average.
6. Follower growth per thousand new listeners. A ratio, so it is comparable across weeks of wildly different size. If listeners spike and this ratio collapses, you have bought attention rather than earned an audience. Both inputs sit under the same window on the project view, so the ratio is arithmetic you can do in your head.
7. Repeat listener share. The proportion of listeners who came back, from whichever platform reports it for you. This is the number that predicts whether a tour will sell in a market.
8. Email open and click rate. Aggregate rates on a list you own. Backline pulls sent campaign performance and subscriber counts from Mailchimp into the same project view as everything else, so a send lines up against the streaming and link response it produced. Mailchimp's published all-industry averages sit around 35.6 percent open and 2.6 percent click, which is a reasonable sanity check if you have nothing else to compare against.
The four that describe intent
9. Landing page click-through by platform. From your own smart links, which Backline builds and measures, so the click data belongs to you rather than to a third party you also have to log into. Tells you where an audience actually goes when given a choice, which is often not where you assumed.
10. Search demand for the artist name. Google Search Console clicks, impressions and the branded queries people used to reach the project's own site. Search is the most honest signal of unprompted interest, because nobody searches your name by accident. Connect the property once and it reports alongside the rest.
11. Sell-through per show. Tickets sold against capacity, per date. Absolute sales tell you almost nothing without capacity, and a 400-capacity room at 92 percent is a better story than a 2,000-capacity room at 38 percent. Backline holds capacity, sold and the pacing curve per show, and rolls dates up into a tour-level sell-through.
12. Press mentions in the last 7 days. A count, not a sentiment score. Coverage clusters, and knowing that five pieces landed last week explains a lot of otherwise mysterious movement. Backline monitors coverage per project and reports the count as one of its headline tiles for exactly this reason.
Asking for the twelve, in one go
IllustrativeGive me the twelve headline numbers for Vela Court over the last 30 days, and flag anything outside its normal range.
Data readStreamingAudience geographySmart linksNewsletterLive showsNews
Over the last 30 days Vela Court gained 214,000 streams across Spotify, SoundCloud and YouTube, which is 9 percent above the previous 30 days.
Monthly listeners 61,400, up 4,900. Followers 18,220, up 1,310, which is 21 new followers per thousand new listeners and the strongest that ratio has been in four months.
Landing page clicks 4,860, with Spotify taking 63 percent and Apple Music 19 percent.
Newsletter: one campaign sent, 38.1 percent open rate and 3.4 percent click rate on 9,140 subscribers.
Two upcoming shows have ticket data. Bristol is at 78 percent of capacity, Glasgow at 34 percent with six weeks to go.
Outside normal range: the follower conversion ratio, and Glasgow pacing, which has not moved in nine days.
Illustrative example using a fictional project. The assistant reads the project's own connected data, names the sources it used, and flags what sits outside the project's own normal range rather than reciting every figure back.
What I would demote
Popular metrics that make poor KPIs
IllustrativeNone of those is meaningless. They are just poor KPIs, because they either cannot be acted on, cannot be compared across weeks, or move for reasons that have nothing to do with anything you did.
Total streams all time is the clearest example. It goes up forever, it is the same number in every window, and its only real use is as a headline in a deck.
Write down the definitions
The failure I see most often is not the wrong metric, it is two people using the same word for different numbers. A label's streams figure and yours will differ if one counts a trailing 28 days and the other a calendar month. Someone's followers figure will differ if it includes a platform the other excludes.
So write the twelve down with the arithmetic next to them. Which window, which platforms, which source, corrected for reporting lag or not. Two paragraphs, one document, and it will save you an argument a quarter. Keeping that definition next to the numbers helps: a Backline project holds its own context notes, and the assistant reads them, so the answer you get back uses your definitions rather than inventing its own.
Then leave them alone for a campaign. Metrics chosen mid-campaign are chosen to make the campaign look good, and everyone in the room knows it.
Common questions
- What are the most important KPIs for an artist manager?
- A workable set is twelve: streams gained in the window, monthly listeners, followers, playlist reach, save rate, follower growth per thousand new listeners, repeat listener share, email open and click rate, landing page clicks by platform, branded search demand, sell-through per show, and press mentions in the last seven days. Four describe reach, four describe quality, four describe intent.
- Why is total streams a bad KPI?
- Because it is cumulative. It increases forever, it reads the same under a 7 day filter as a 90 day filter, and it cannot fall even when a project stalls. Use the gain inside a defined window instead, calculated as the total at the end minus the total at the start.
- How many metrics should a monthly report contain?
- Roughly the same twelve, with one paragraph explaining what moved and why. Longer reports get skimmed, and every extra metric is another number somebody can dispute without changing a decision. Backline's scheduled reports send that shape automatically, on a cadence you choose, to the recipients you name.
Sources
- 1Mailchimp, Reviewed August 2026. Email marketing benchmarks by industry
- 2Spotify for Artists, Reviewed August 2026. Data in Spotify for Artists
- 3Google, Reviewed August 2026. Search Console API: Search Analytics query reference
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
Analytics foundations
The 7 day problem in music dashboards
A thirty second test tells you whether a music dashboard's date filter is real. Change the range and watch the streaming headline. What it means if the number does not move.
Danny Starr · 3 min read
Analytics foundations
A weekly data review that takes twenty minutes
A repeatable Monday routine for managers running several projects: what to look at, in what order, what to write down, and what to deliberately ignore until month end.
Danny Angove · 5 min read
Streaming
Monthly listeners, followers and streams: what each is worth
Three numbers that get used interchangeably and measure completely different things. What each one responds to, which can be inflated, and which one to quote in which conversation.
Danny Angove · 4 min read