Compensation currency after an outage: what it costs

Something breaks. The game is down for four hours, or a feature was broken for a day, or an event failed to award what it promised. The team fixes it, writes the apology, and grants every affected player a package of currency.

That last step is reasonable and usually right. It is also the part that gets the least thought, because it happens under time pressure and everyone involved is focused on the incident rather than on what a mass currency grant does to the game a month later.

A grant is an unplanned faucet

Every unit of currency in your game entered through some source, and each source is tuned. Quest rewards, daily bonuses, event payouts — these are set at rates that balance against what players spend on, and the balance is what keeps prices meaningful.

A compensation grant is a source that was not tuned, delivered to everyone at once. The size is decided in an hour by people thinking about goodwill rather than balance, and it lands on every active player simultaneously rather than being earned over time by individuals at different points.

The simultaneity is the part that matters. Normal currency income is spread out — players earn at different rates and spend at different moments, so the aggregate balance moves smoothly. A grant is a step change to every wallet on the same day. Whatever the game's currency supply looked like, it is now higher by the grant amount times your active population, and it happened instantly.

If the grant is small relative to what a player typically holds, this washes out quickly. If it is large — and grants tend to be generous, because the team is trying to make amends — the effects last well past the incident.

Where the cost shows up

The bill does not arrive as a line item. It arrives as three effects that are easy to attribute to something else.

The first is a quiet drop in purchases in the days after the grant. Players who were approaching a purchase now have enough without buying. That revenue is not lost permanently in every case, but a portion of it is: some of those players would have bought, spent, and been back in a buying position within the week, and instead they satisfied the want with the grant and moved on.

The second is a change in what things feel worth. If your mid-tier item costs roughly what a player earns in a week, and you have just handed everyone a week's earnings, that item is temporarily much easier to obtain. Players who acquire it during that window did not experience the progression the price was designed to represent. This is not catastrophic, but it does compress a chunk of progression for a whole cohort at once, and it shows up later as that cohort reaching the content ceiling sooner than they should have.

The third is expectation. If compensation is generous and outages are not rare, players learn that downtime pays. That is not cynicism on their part — it is an accurate reading of the pattern. Studios that have been through several rounds of this sometimes find that the community response to a minor issue includes an explicit expectation of compensation, and the anchor for what counts as adequate is whatever was given last time.

Sizing it against something real

The instinct under pressure is to pick a round number that looks generous. A better method takes about the same amount of time and produces a defensible answer.

Work out roughly what an affected player would have earned during the affected window. If the game was down for four hours and a typical engaged player earns a certain amount in a session, the compensation that actually restores what they lost is close to that. That number is almost always smaller than the round number instinct produces, and it is the amount that returns players to where they would have been rather than putting them ahead.

The gap between "restore what was lost" and "make a gesture" is where the decision sits. A gesture on top is often warranted — it acknowledges the inconvenience rather than just the lost progress. But separating the two makes the size deliberate: this much restores, this much apologises, and here is why the second number is what it is.

The other sizing question is who gets it. A grant to every account including dormant ones is the simplest to execute and the most expensive, because it injects currency into wallets belonging to players who were not affected and may not return. Scoping it to players who were actually active during the window is more work and materially cheaper.

See how we model economy changes before they ship →

Alternatives worth considering first

Currency is the default because it is easy to grant, but it is not always the best match for what was lost.

If the failure was a broken event, granting the event's rewards directly is more precise than granting currency of equivalent value. It puts players where they would have been, in the specific way they would have been there, and it does not add to the general money supply.

If the failure was downtime, extending the affected event or content window costs nothing in currency terms and gives players back the thing they actually lost, which was time rather than income. This is frequently the better answer and is skipped because a grant feels more substantial.

If the failure affected a subset, a targeted make-good to that subset — even a more generous one — is usually cheaper and better received than a smaller grant to everyone. Players who were not affected do not need compensating, and including them dilutes the gesture for the people who were.

None of this argues against compensating. It argues for treating the grant as what it is: a change to the game's economy, made quickly, with consequences that outlast the incident. Five minutes spent on the size and the scope, while the fix is being deployed, is usually the cheapest five minutes available.

Talk to us about economy change planning →