Routing a tour with streaming data, carefully

Streaming geography changed touring for developing artists. Before it, routing a first tour outside your own city was guesswork plus whoever the agent knew. Now there is a map.

The map is useful and it is not a booking sheet. Here is how to use it without getting caught out.

What the data does and does not measure

Listener geography tells you where people played your music. That is genuinely informative and it is a step removed from the question you are asking, which is who will buy a ticket, arrange an evening and travel to a room on a Tuesday.

The gap between those two shows up in predictable ways.

Algorithmic listening inflates markets. If a large share of your plays in a city came from placement, those listeners did not choose you. They will not notice an announcement.

Cities are not catchments. Streaming data is reported by city, and a venue serves a region. A modest city figure surrounded by populated area can outperform a bigger city that people leave for shows.

Streaming skews young and urban relative to ticket buying in many genres, and the gap varies by genre in ways your own history knows better than any general rule.

The hierarchy of routing signals

Previous ticket sales in a market beat everything. If you played somewhere and sold, that is not a proxy, it is the answer.

Then followers, because a follow is a decision. Then engaged listeners, then raw listener count, then anything derived from an algorithmic placement.

The practical version: build a shortlist from followers and over-indexing markets, then apply the confirmation rule below before anything reaches the agent.

Confirm with something that is not streaming

One signal is a hypothesis. Two independent signals is a market. For live specifically, the confirmations worth having:

• Ticket history, yours or a comparable act's.
• Radio airplay in that region. Airplay means the record is physically in the air where people live.
• Local press. Coverage from outlets based there, which also tells you who to approach for the announcement.
• Landing page clicks by country, which is your own first-party data rather than a provider's estimate.
• Mailing list concentration, if your email platform reports it at country level.

A city with over-indexing streams, regional airplay and previous local press is a genuinely evidenced date. A city with only a large listener count is a guess with a chart attached.

Weight recency

Listener geography usually comes with a current figure and a peak. A market at 40 percent of a peak set two years ago is a historic market, not a live one, and it will not sell like its peak.

Similarly, read geography at least three weeks after any large placement ends. During and immediately after, the map reflects the playlist's audience rather than yours, and routing a tour off a playlist's geography is an expensive way to learn that.

What to hand the agent

Not a list of cities. A short document per candidate city:

• Followers and listeners, current and peak, with the dates.
• The over-index figure, so a smaller city with disproportionate strength does not get dismissed.
• The confirming signals: airplay, press, previous sales, link clicks.
• Anything you know that is not in the data. A promoter relationship, a support slot offer, a local scene connection, a friend who runs a venue.

An agent's judgement plus evidence beats either alone, and the conversation is more productive when both sides can see the same numbers. What does not work is arriving with a route already drawn from a listener ranking, because the first three questions an experienced agent asks will be about things the ranking cannot answer.

After the tour, close the loop

The most valuable dataset for routing the next tour is the last one. For each date, record the market, the capacity, the sell-through, the shape of the curve and what the pre-announce signals looked like.

Two runs of that and you have a house model of what a market needs to look like before it sells for your project specifically, which is worth more than any generic benchmark. Almost nobody does this, and it is the cheapest advantage available in live music data.