Application strategy: metrics and KPIs for the scale phase

There is a dangerous moment in the life of an app: the one when it stops looking for traction and starts scaling. It's dangerous because climbing multiplies everything, including defects that no one noticed while the base was small.

In the validation phase, wrong metrics are cheap. You adjust and follow. In the scaling phase, the same wrong metric costs you dearly, because you are investing heavily to grow a number that may not mean what you think it means. Growing quickly in the wrong direction is the most sophisticated way to go broke.

The thesis of this text, aimed at those who already have a working product and are preparing to accelerate, is straightforward: scaling well is a metric decision before it is a marketing decision. Before you pour money into acquisition, you need to know, with confidence, that each new user is worth more than it costs. Without this, scale is just burning cash with the appearance of success.

On the scale, vanity metric becomes passive

Number of downloads, total registered users, likes, these metrics are comforting and dangerous. They always go up, sound good in presentation and say almost nothing about health.

In the traction phase, you can tolerate looking at them. At the scale, they become an active hazard, because they hide what is breaking as you accelerate. An app can double its installed base and lose money faster with each new user, and the download graph won't account for that.

The first move of those who climb with maturity is to remove the vanity from the main panel. Not because these numbers are useless, but because they should not guide investment decisions. What guides is the metric that answers: is this growth healthy or am I filling a leaky bucket faster?

The economic unit is the KPI that decides the scale

Before scaling, one number needs to be resolved: how much it costs to acquire a user compared to how much they generate over time.

If the acquisition cost exceeds the value the user brings, scaling only deepens the damage. Every real invested in growth returns less than what was received. Many operations discover this late, after having already wasted capital buying a base that would never pay off.

The relationship between acquisition cost and lifetime customer value is the master KPI of the scaling phase. He doesn't need to be perfect, but he needs to be healthy and improving. Scaling before closing this account is betting that volume fixes a bad economy, and it rarely does.

Retention as a precondition, not as a metric among others

Retention, on the scale, stops being just any metric and becomes a prerequisite. An app with poor retention is a leaky bucket: you can pour users at the top, but they trickle down to the bottom, and the acquisition effort is wasted.

Scaling acquisition over poor retention is the most expensive strategic mistake at this stage. The sign of readiness to scale isn't "we're growing," it's "the people who join stay." Without this, more marketing just accelerates the exit.

The metrics that support the decision to accelerate

Some indicators deserve to be highlighted when the objective is to grow safely:

Retention by cohort over time. Not the overall average, which is misleading, but the behavior of each group of users over the weeks. Newer cohorts retaining better than older ones is the green light to accelerate.

Acquisition cost per channel. At scale, channels that used to be small can saturate. The cost per user tends to go up as you exhaust the easy audience. Monitoring this by channel avoids wasting budget where returns have already fallen.

Value generated over the user's lifetime. What each user returns, in revenue or other core business value, over time. It is the other half of the economic unit bill.

Engagement with the action that defines value. Every app has an action that proves that the user "understands" the product. Monitoring the frequency of this action reveals whether the growth is bringing in people who actually use it or just install it.

The risks that scale amplifies

Scaling doesn’t just create opportunities; amplifies risks that were manageable when small.

The infrastructure risk is the most obvious: what could support thousands of users can collapse with millions, and the failure comes at the worst time. But there are less visible risks. Support that worked on word of mouth collapses in scale. Manual processes that could fit into a spreadsheet became a bottleneck.

There is also data and compliance risk. More users means more personal data, and LGPD doesn't become optional because you're growing fast. Scaling collection without scaling governance is accumulating a liability that charges interest. At some point, he wins.

Maturity lies in scaling these dimensions together with the acquisition, not in discovering, by surprise, that the product has grown and the operation has not.

The channel saturation that no one predicts

A specific scale phenomenon deserves attention, because it catches many operations by surprise: acquisition channels saturate. The easiest to convert audience is reached first. As you scale, you start to compete for the attention of those who are more expensive to convince.

This means that the acquisition cost that supported the account in the traction phase tends to get worse exactly when you invest the most. The channel that returned good returns with a small budget may return mediocre returns with a large budget. Those who don't monitor this by channel continue pouring money where efficiency has already fallen, trusting in a historical average that is no longer valid.

The defense is to treat each channel as a curve, not a fixed number. Tracking marginal cost, what the next user costs, not the average user, reveals when a channel has started to saturate. This is the type of reading that distinguishes those who climb with method from those who climb by fright.

Grow because it makes sense, not because it works

The pressure to grow is enormous, especially when there are investors or aggressive goals on the horizon. But climbing under pressure, without the numbers supporting it, is exchanging today's problem for a bigger problem tomorrow.

The strategic question in the scaling phase is not “how do we grow faster?” It's "is this growth, at the speed we want, sustainable?" When the indicators say yes, accelerating is the right decision and the capital pays off. When they say no, speeding up is just losing money more efficiently.

Climbing well is, basically, having the discipline to only step on the accelerator once you know where the car is going. KPIs are that map. Without them, speed becomes a risk.

If your application is at this turning point and the decision to accelerate is still based more on intuition than on numbers, it's worth talking about it before committing to the budget. On the blog there are other texts about product and growth metrics that delve deeper into each of these indicators.