Guide
How to measure server list and creator ROI
A server list slot or a creator video costs real money, and the numbers you get back rarely say whether it paid. Votes, views and a busy evening feel like results, but none of them tells you how many new players stayed or spent.
This guide sets up four numbers you can work out for any source, walks through an example, and covers how to read the result without fooling yourself. All it needs is a spreadsheet and a way to know which address each player first joined through.
Before you pay: set the deal up to be measured
- Give the deal its own address, such as
list1.example.netoralex.example.net. If you have not done this before, tracking where your players come from covers the DNS and the traps. - Write down the start date, the end date, and what you paid. A slot bought by the month and a flat fee for one video are both a cost and a window.
- Pick a follow-up period before you look at any numbers, for example the 30 days after each player's first join. Fixing it up front stops you from choosing whichever period looks best.
- If the creator also has a store code, note it. It gives you a second view of their purchases.
The four numbers
Work these out per source, over the same window, so two deals can be compared even when one sent ten times the traffic. Two helpers go next to them in the sheet: buyers, the number of new players who bought anything, and payback, which is revenue divided by cost.
New players
Players whose very first join came through the address during the window. Not total joins and not unique players: regulars who saved the address would inflate both. A source only gets credit for players it introduced.
7-day return rate
Of those new players, the share who came back at any time in the 7 days after their first join. Count only players whose 7 days are over; someone who joined yesterday is pending, not lost. Leave very short visits and failed logins out of what counts as coming back, or bots and connection tests will look like loyal players. Day 1, Day 7 and Day 30 retention explains the windows in more detail.
Revenue per new player
Net store revenue from those new players inside your follow-up period, after refunds and chargebacks, divided by the number of new players. Count the period from each player's own first join, so someone who joined on the last day of the window gets the same chance to spend as someone who joined on the first.
Cost per returning player
What you paid, divided by the new players who came back within 7 days. This is the number most likely to change a decision. Cost per join rewards sources that send crowds who leave after one look; cost per returning player only counts the players you kept.
A worked example
Hypothetical numbers
Everything in this example is made up to show the method. It is not data from any real server, creator or server list.
Say you ran two deals in the same month: a sponsored slot on a server list for 30 days, and one video from a creator for a flat fee. Each had its own address. Once every new player's 30 days were over, the sheet looked like this:
| Server list slot, 30 days | Creator video, flat fee | |
|---|---|---|
| What you paid | $150 | $200 |
| New players | 600 | 220 |
| Came back within 7 days | 72 (12%) | 77 (35%) |
| Buyers | 9 | 14 |
| Net revenue, first 30 days | $135 | $310 |
| Revenue per new player | $0.23 | $1.41 |
| Cost per new player | $0.25 | $0.91 |
| Cost per returning player | $2.08 | $2.60 |
| Payback (revenue / cost) | 0.90x | 1.55x |
The server list slot looks better on the first measures. It is cheaper per new player, $0.25 against $0.91, and even a little cheaper per returning player. But its players did not stay or spend: 12% came back, and the slot earned back 90% of what it cost. The creator sent fewer people, but 35% of them came back, and they spent more than the fee.
Neither answer is final. With 9 and 14 buyers, one large purchase could move either payback figure a long way. The honest reading is that the creator is worth another deal, and the slot is worth one more month only if it gets cheaper or you improve what new players see when they arrive.
Reading the result without fooling yourself
- Compare sources over the same window and the same follow-up period. Revenue measured over 60 days will always look better than revenue measured over 14.
- Look at buyers as well as revenue. One big buyer and twenty small ones can add up to the same total, and only one of those is likely to repeat.
- Give creator videos a longer tail. A video keeps sending players for weeks after it goes up, so a window that closes after a few days undercounts it.
- Server list positions change. Compare the same slot across months, not a top slot against a lower one.
- Expect the address and a creator code to disagree. The address credits players who joined through it; the code credits purchases where someone typed it. Use both, and never add them together.
- Watch for cheaters. A source whose players keep getting banned is not a cheap source, whatever its cost per join.
Doing it in a spreadsheet
One row per deal is enough. These columns hold everything above, shown here with the example numbers:
source,address,cost,new_players,returned_7d,buyers,net_revenue_30d
Server list slot, 30 days,list1.example.net,150,600,72,9,135
Creator video, flat fee,alex.example.net,200,220,77,14,310Then add four formulas to each row:
- Return rate:
returned_7d / new_players - Revenue per new player:
net_revenue_30d / new_players - Cost per returning player:
cost / returned_7d - Payback:
net_revenue_30d / cost
Fill a row in only once every new player's 7 days and follow-up period are over. Until then, mark it pending, the same way a fair retention chart does.
Where MCAnalytics fits
MCAnalytics fills in most of the sheet. The Growth page shows every address with its new players, 7-day return rate and revenue per new player, and a join funnel follows each address from first join to a return within 7 days and a purchase within 30 days. Tebex purchases are matched on the buyer's premium UUID and credited to their first address, with refunds and chargebacks taken off, and a creator codes report sits next to it.
One part stays yours: MCAnalytics does not record what you paid for a deal yet, so cost per returning player and payback are a division you do with the numbers it shows. See source tracking, Tebex analytics and creator codes.