Should I Focus on Getting New Users or Keeping Existing Ones Happy?

Around 25% of mobile apps are used once and never opened again. One session, then silence. The user does not complain, does not cancel, does not leave a review. They simply stop. And the team building the product often has no idea it happened, because they were watching download numbers climb and calling it growth.

The real question beneath the acquisition versus retention debate is whether your product is earning the users it already has.

This is the context in which the acquisition versus retention question lives. On the surface it sounds like a budget allocation problem, a slide for the quarterly review. In practice it is a question about whether your product is actually working, and whether spending more to bring people in is making things better or just making the problem bigger and more expensive.

We work on products across sectors and stages, and the teams that struggle most with this question tend to share one thing: they are measuring the wrong things and drawing the wrong conclusions. Download numbers look like growth. Session counts look like engagement. Neither tells you whether a real person found genuine value and came back for more. Getting to the right answer means being honest about what your numbers actually show, and what they are hiding.

Why the Question Has No Universal Answer

A product that launched three weeks ago and a product that has been live for four years face completely different problems. Asking them both whether to prioritise acquisition or retention is like asking someone on their first day of a new job and someone who has held that role for a decade the same question about career development. The answer depends entirely on where they are.

A new product needs users before it can retain them. Retention without acquisition is just a small group of early adopters who may or may not represent the broader market. But acquisition without any regard for retention is a leak with a tap open above it. You can keep turning the tap, and the bucket never fills.

The honest answer is that most products need both, but the balance shifts depending on three things: how long the product has been live, what the retention data actually shows, and what it costs to acquire users relative to what each user is worth over time, all of which are considerations that sit at the heart of a sound app planning and strategy process. Get the balance wrong in either direction and you waste money. Get it wrong by over-indexing on acquisition while retention is broken, and you can destroy the business case entirely.

The Financial Logic: What Acquisition Actually Costs When Retention Is Broken

Retention is both a product health metric and a financial one. A user who churns after three days has a lifetime value close to zero. The cost to acquire them, whether through paid media, referral incentives, or influencer campaigns, does not disappear when they leave. It sits on the balance sheet as a sunk cost, and every new user acquired into a broken retention environment adds another one.

Repeat customers spend up to 67% more than new customers, according to Business.com. That gap compounds over time. A product that holds users through month three and beyond does not just benefit from direct revenue. It benefits from reduced acquisition pressure, from word-of-mouth referrals, and from the kind of brand trust that lowers the cost of everything downstream.

When retention is broken, acquisition spend becomes damage containment, and expensive damage containment at that. The financially rational move, before any acquisition budget is committed, is to understand what happens to users after they arrive and whether the product gives them a reason to stay.

Where You Are in the Adoption Curve Changes Everything

Early-stage products and mature products have opposite problems, and treating them the same way produces the wrong interventions. In the early stage, the priority is learning. You need enough users to generate meaningful feedback, to understand which parts of the experience land and which fall flat. Acquisition at this stage is about signal. You need people in the product so you can watch what they do.

Acquisition at the early stage is about generating signal, not chasing scale.

As a product matures and the core experience stabilises, the calculus changes. Retention becomes the leading indicator of product-market fit, and acquisition spend starts to make economic sense only once that retention baseline is solid. A product sitting at 30% day-thirty retention is not ready to scale acquisition. A product sitting at 65% is in a very different position.

We worked on an art-based auction game with real money prizes that illustrated this clearly. In its healthy phase, day-thirty retention sat in the mid-sixties to low seventies percentage range. That is a product ready to grow. When the supply of available games dropped and retention fell to around 30 to 35%, the instinct from the client was to add new features to win users back. That instinct was wrong. The product was not ready for that investment. It needed the core experience fixed first.

The maturity test

A simple way to sense-check where you are: could you describe, in one sentence, why a user comes back on day seven? If the answer is vague, the product is still in early-stage territory regardless of how long it has been live. If the answer is specific and grounded in user behaviour you can actually observe, you are in a position to build on it.

What Your Retention Numbers Need to Show Before Acquisition Spend Makes Sense

Day-one retention below 50% is a warning sign. The goal is to push that number as high as possible and then carry it through to day thirty and beyond, particularly when acquisition spend is involved and there is a real cost attached to every user who arrives. Spending to acquire users into a product where half leave on day one is a compounding loss.

The benchmark worth watching is not just day one, though. In the top ten personal finance apps, around 71% of daily active users are lost between day one and day thirty. We saw a similar drop-off pattern in a water tracking app we developed, where user silence had been misread as satisfaction. The retention data told a different story.

Before committing acquisition budget, the numbers you want to see are roughly these:

• Day-one retention above 50%, with a clear upward target
• Day-seven retention showing that the initial drop-off has stabilised
• Day-thirty retention that justifies the cost to acquire each user
• Engagement patterns that reflect actual product use, not confused navigation or gamification loops

Run your retention numbers before you finalise any acquisition budget. If day-thirty retention does not justify the cost per acquired user, fix the product first and revisit the spend.

If those numbers are not there, acquisition investment is premature. That is a judgment on the timing, full stop.

How Silent Churn Hides the Real Problem

Most dissatisfied users do not complain. They do not submit a support ticket, leave a review, or tell anyone why they left. They just stop opening the app. This is the nature of mobile behaviour, and it makes silent churn one of the hardest problems to detect from the inside.

The teams most vulnerable to this are ones that have built an implicit assumption into their monitoring: that no news is good news. If nobody is complaining, the product must be working. In mobile, that assumption is almost always wrong. Around 25% of mobile apps are abandoned after a single use, and the users who abandon them rarely say why.

We encountered this with the water tracking app. The absence of complaints had been read as a positive signal. When we looked at the actual retention data across early days and weeks, the picture was entirely different. Users were leaving quietly and consistently, and the product team had no mechanism to catch it because they were not proactively tracking the right things or asking users directly how the experience was going.

What proactive monitoring looks like

The fix is not complicated, but it does require deliberate effort. Track retention at day one, day three, day seven, and day thirty. Monitor engagement levels over time, not just totals. And ask users directly, through well-timed in-product prompts, how things are going. Do not wait for them to volunteer the information. They will not.

Build retention checkpoints into your product reporting at day one, three, seven, and thirty. Treat a drop between any two points as a signal worth investigating, not a number to smooth over in the quarterly deck.

When Ignoring Retention Destroys What Acquisition Built

The auction game project is the clearest example of this we have seen. When the product was in good health, retention sat in the mid-sixties to low seventies. The core experience was working: enough games were available, users were playing, and the product had a reason to open. Then the supply of games declined. Users had less to do. Retention dropped to around 30 to 35%.

The client's response was to push for new features. More bells, more mechanics, more reasons to engage on the surface. But the root problem was not a lack of features. The product had lost the thing that made it worth opening in the first place. Adding features to a product with a broken core experience does not fix the core experience. It adds complexity around a hole.

We raised the concern repeatedly. Eventually, when retention hit that low point and the evidence was undeniable, the team refocused on the core user journey: making sure enough games were available and that the fundamental experience was working again. Retention recovered to the mid-sixties to low seventies, but rebuilding user trust after a sustained period of decline took considerably longer than fixing the supply problem itself.

The lesson is not subtle. Acquisition cannot build a user base that retention is actively dismantling. Every user brought in during the low-retention period was effectively acquired for nothing. The spend was real; the growth was not.

What Genuine Retention Looks Like Versus Retention That Flatters the Dashboard

Session length going up sounds like good news. Daily active users holding steady sounds like good news. Neither of those things necessarily is. A user staying in your product for a long session might be finding real value. Or they might be confused about how to complete a task. Or the product might be holding them through notification pressure and gamification loops that serve the platform, not the user.

The removal test is a useful way to check which is which. Pick a feature that drives engagement and ask what would happen to your metrics if you turned it off for a group of users. If the honest answer is "we would never do that, " that reaction is diagnostic. It tells you the feature exists to protect a metric, not to serve a user, and that somewhere the team already knows this.

Similarly, ask whether users can reduce notification frequency without losing meaningful product value. If they can opt out of half your notifications without missing anything that matters, the current cadence is serving the dashboard, not the person.

Genuine retention looks like users returning because the product solved something for them. It shows up as low uninstall rates, positive voluntary reviews, and referrals that come without incentives. These are harder to manufacture and harder to fake, which is exactly why they are worth measuring.

How Ethical Product Design Affects the Retention Baseline

There is a meaningful difference between a product that keeps users through manipulation and one that keeps them through value. Ethical products, those that are transparent about what they do, honest in their onboarding, and restrained in their use of behavioural nudges, tend to see around 23% higher retention rates than those built on manipulative mechanics. That figure is not incidental. It reflects something real about how trust compounds over time.

When a user feels tricked, the churn is rarely immediate. They often stay for a while, partly out of inertia and partly because switching has its own friction. But the trust has already gone. What follows is the kind of disengaged presence that inflates monthly active user numbers while contributing nothing to the product's real health. They are counted but not retained in any meaningful sense.

Value-driven onboarding, clear communication about what the product does and does not do, and a notification cadence that respects user attention all contribute to a retention baseline that holds under scrutiny. When you eventually invest in acquisition, you are bringing users into something that earns them rather than traps them. That is the foundation on which acquisition spend actually works.

Before scaling acquisition, audit your onboarding for manipulative patterns. Dark patterns and misleading defaults inflate early engagement and accelerate eventual churn. Fix them first.

Making the Call: A Framework for Prioritising One Over the Other

The decision is not always clean, but these questions make it clearer.

Between funding rounds, we made the decision to implement much better retention tracking and to proactively ask users how things were going. That decision came from recognising that investor confidence had been used as a proxy for product health, which is a comfortable proxy and a misleading one. The product's actual retention had not been properly understood. Getting that understanding in place before the next acquisition push was not cautious, it was the only rational move.

If your retention data is solid and you can articulate why users return, acquisition investment is ready to work. If it is not, that investment will accelerate a problem rather than build on a foundation.

Conclusion

The acquisition versus retention question does not have a universal answer, but it does have a logical sequence. Retention comes first, not as a moral position but as a financial one. A product that cannot hold the users it already has will not benefit from more of them. It will simply spend more to lose more.

The auction game taught us that clearly. Recovery from broken retention is possible, but it takes longer than the original decline, and it costs more than fixing the problem would have. The water app taught us that silence is not approval. The absence of complaints is often just a gap in your monitoring.

What we look for before recommending any acquisition push is a retention baseline that holds at day one, day seven, and day thirty, engagement data that reflects genuine use rather than confusion or manipulation, and a clear answer to the question of why users come back. When those things are in place, acquisition spend has somewhere to go. When they are not, it has nowhere to land.

The products that grow well are the ones that earned their users before they went looking for more. That sequence is straightforward, just less exciting to say than "let's grow" and harder to defend in a room full of people watching download numbers go up.

If you are trying to make this call for your product, let's talk about your retention strategy.

Frequently Asked Questions