HOME

SERVICES

BLOG

Schedule call

Categorias

Top Posts

1

Título do post número 1

CATEGORIA • 17 NOV, 2025

2

Título do post número 2

CATEGORIA • 17 NOV, 2025

3

Título do post número 3

CATEGORIA • 17 NOV, 2025

4

Título do post número 4

CATEGORIA • 17 NOV, 2025

5

Título do post número 5

CATEGORIA • 17 NOV, 2025

Instagram

Stephanie Amatus
stephanieamatus Seguir
2 posts
69 seguidores
356 seguindo
Stephanie Amatus she/her

I like thinking about stuff, and sometimes I talk about them.

Post 1

LinkedIn

Stephanie
Stephanie Amatus

Growth & Marketing Strategist

Curitiba, Paraná, Brasil · Dados de contato

Conectar

The 90-Day Turnaround: How to Fix an Underperforming Growth Strategy

Days 1–30: Reconstruct What Changed

Growth strategies rarely announce the moment they stop working. There is usually no dramatic collapse, no single campaign that suddenly destroys performance and no convenient notification informing you that the strategy that worked beautifully eight months ago has reached the end of its useful life.

It tends to happen quietly, with customer acquisition becoming a little more expensive, campaigns that used to carry the quarter still generating results but requiring increasingly more effort to do so, sales needing more leads to close the same number of deals, organic growth losing momentum and conversion slipping by a few percentage points. Since everyone can still point to something that is technically working, it becomes surprisingly easy to spend months adapting to worse performance without realizing that the strategy itself is deteriorating.

Eventually, someone looks at the numbers over six or twelve months instead of six or twelve weeks and notices the uncomfortable pattern: the company has been working harder, spending more and producing increasingly mediocre growth.

That is the kind of situation where I think a 90-day turnaround can be useful, not because three months is enough to magically transform every part of a business, and certainly not because growth can be fixed by following a neat 30-60-90-day template, but because ninety days gives you enough time to investigate what changed, make a limited number of meaningful interventions and collect enough evidence to decide what the business should do next.

If I were responsible for those ninety days, my objective wouldn’t be to make every number green by the end of the quarter. I’d want to leave the business with something much more valuable: an understanding of why growth slowed down in the first place and a system capable of noticing and responding to those changes before another quarter disappears.

The first thing I would do is resist the urge to compare current performance only against targets, because targets tell you where you wanted to go, but they don’t necessarily tell you what went wrong on the way there.

Instead, I’d go backwards.

I’d find the period when the strategy was performing better and start asking what was different. Perhaps customer acquisition cost has increased 35% over the past year, which certainly matters, but the percentage itself isn’t particularly useful until we understand what produced it. Media costs may have risen, conversion may have fallen, the audience may have changed, a previously strong creative concept may have exhausted itself or the company may have gradually moved toward customers who are easier to attract but considerably harder to convert.

Growth is the final result of many variables interacting with one another, so a deteriorating headline metric can hide several completely different stories underneath it.

My job during those first weeks would be to reconstruct those stories.

I’d compare acquisition sources over time, looking beyond how much traffic or how many leads they generated to understand what eventually happened to those people, then connect that information with conversion, sales cycles, average deal value, retention, repeat purchases and whatever else determines the economics of that particular business.

The question guiding the analysis would be deceptively simple: what changed between the growth we used to have and the growth we have now?

Look for deterioration, not just bad numbers

One distinction I find particularly useful during a turnaround is the difference between something that has always performed poorly and something that used to work and has gradually deteriorated, because those situations point toward very different problems.

If a channel has never produced profitable customers, the assumptions behind using it may have been wrong from the beginning, whereas a channel that produced profitable customers consistently for two years and then stopped suggests that something in the environment, the audience or the business itself has changed.

Competition may have increased, acquisition costs may have risen beyond what the unit economics can comfortably tolerate, customer behavior may have shifted, the offer may have become less competitive or the company may have changed something internally without realizing how much that decision would affect the rest of the customer journey.

Historical comparison gives us clues that a snapshot cannot, which is also why I would be cautious about immediately benchmarking the company against competitors or industry averages. External benchmarks can provide useful context, but the business’s own history often gives us a cleaner investigation.

If conversion fell from 6% to 3.8% while traffic quality remained relatively stable, I care much more about understanding what happened during that decline than discovering that the supposed industry average happens to be 4.2%.

Something changed inside this particular system, and that’s what I want to find.

 

Days 31–60: Decide What Actually Deserves Intervention

After a month of investigation, there is a good chance you will have discovered more problems than you can reasonably solve during the remaining sixty days.

That’s normal.

Every business contains inefficiencies, outdated processes, imperfect campaigns and opportunities nobody has had time to explore, but the purpose of a turnaround isn’t to eliminate every weakness in the organization. It is to identify which ones are currently placing the greatest constraint on growth and concentrate enough resources there to produce a meaningful change.

Suppose the analysis reveals that paid acquisition has become 25% more expensive, landing-page conversion has fallen 8% and customer retention has deteriorated dramatically enough to account for most of the gap between projected and actual revenue. All three deserve attention eventually, but treating them as equally urgent would spread the team across several problems when one is clearly doing more damage than the others.

This is where I would prioritize according to impact, confidence and effort, asking how much each problem actually affects growth, how confident we are that we’ve identified what is causing it and how quickly we can test an intervention without creating new problems somewhere else.

Then I would choose a small number of bets, not because there aren’t other good ideas worth exploring, but because a turnaround needs enough focus for us to understand whether our decisions actually caused anything.

Fix the economics before increasing the volume

One principle would guide most of my decisions during this phase: I don’t want to scale something whose economics are already deteriorating.

If every dollar invested in a channel used to return five and now returns two, doubling the budget because the company needs more revenue may technically increase sales while simultaneously making the underlying problem considerably larger. The same logic applies when sales converts only a small percentage of qualified opportunities, when a subscription business is experiencing unusually high early churn or when an ecommerce company keeps increasing acquisition while repeat purchase rates continue falling.

More volume can hide inefficiency surprisingly well, particularly when everyone is watching revenue without paying enough attention to what it costs to produce it.

That’s why the middle thirty days should be less concerned with making the business larger and more concerned with making the existing growth system healthier. Depending on what the first month revealed, that might mean improving conversion, adjusting the offer, changing qualification criteria, fixing onboarding, rebuilding an acquisition campaign, repositioning the product or improving the relationship between marketing and sales.

There is no universal tactic I would prescribe here because the entire point of spending the first month diagnosing the system is to avoid prescribing a solution before understanding the problem.

Run experiments that answer meaningful questions

This is also where I would become fairly strict about what qualifies as an experiment.

Growth teams can easily become trapped in endless micro-optimization because small changes are easy to launch, easy to measure and wonderfully convenient to put inside a report, but a turnaround needs experiments connected to meaningful questions about the business rather than a collection of things someone thought might be worth trying.

If customers are increasingly choosing a competitor because they perceive implementation as easier, perhaps we test an offer that reduces implementation risk. If a high-value customer segment converts significantly better after encountering a particular case study, perhaps we restructure that journey so the evidence appears earlier. If sales repeatedly hears that prospects don’t understand why the product costs more than its alternatives, we may have a positioning problem worth investigating rather than simply a sales-script problem.

The purpose of an experiment isn’t merely to improve a number. It should leave us understanding something about the customer, the offer or the growth system that we didn’t understand before.

That way, even when the experiment fails, the business has learned something useful.

Then leave the dashboard for a while

At some point during the first month, I would stop looking at charts and go talk to people, because data is remarkably good at showing us what happened and considerably less capable of explaining why human beings behaved the way they did.

Salespeople often notice changes in the market before those changes become obvious in aggregate data because they hear when prospects begin mentioning a competitor more frequently, when an objection that used to appear occasionally suddenly shows up in every conversation or when leads that look perfect inside the CRM consistently turn out to be wrong for the business.

Customer support knows what frustrates people, customer success often knows why an account is becoming difficult long before it churns and customers themselves know what they expected when they bought, what almost prevented them from buying and whether the experience they eventually received matched the promise that brought them there.

Even the language people use can reveal something important. A company may spend years describing its product around a particular benefit while customers consistently describe its value in completely different terms, and that gap can expose positioning opportunities that no analytics platform was ever going to identify.

By the end of the first thirty days, I don’t necessarily need to have solved the growth problem, but I do need to be able to explain it.

Days 61–90: Decide What Has Earned the Right to Scale

A Turnaround Should Make the Next Turnaround Less Necessary

By the beginning of the third month, I would expect the conversation inside the company to have changed, because we should no longer be discussing growth entirely through opinions and assumptions. We should have evidence about what changed, evidence about which interventions affected it and, just as importantly, evidence about which ideas sounded promising but didn’t survive contact with reality.

Perhaps a new onboarding flow significantly improved activation, repositioning the offer increased conversion among one customer segment while reducing it elsewhere or a channel everyone assumed was essential turned out to generate customers with poor retention, while a much smaller source of acquisition quietly produced the most valuable accounts.

Those discoveries matter because growth strategies improve when resources begin following evidence rather than organizational habit.

The final month is where I would start making those decisions more permanent, deciding which activities deserve more investment, which ones should receive less, which experiments should become standard practice and which parts of the strategy exist primarily because nobody has questioned them in a very long time.

There can be a surprising amount of growth hidden inside that last category.

Turn discoveries into operating decisions

A successful experiment that remains inside a presentation isn’t particularly useful.

If the turnaround reveals that leads contacted within an hour convert substantially better, that insight needs to become part of the sales process. If a particular type of content consistently influences high-value opportunities, that information should shape the editorial strategy. If one customer segment produces dramatically better lifetime value than the others, the discovery should influence targeting, messaging, acquisition, product decisions and potentially pricing.

This is the difference between optimization and organizational learning: optimization improves a number, while learning changes how the company operates because of what the number taught us.

By the end of the ninety days, I would want the business to have a much clearer relationship between the evidence it collects and the decisions it makes, which includes measurement but not measurement for its own sake. I don’t need a dashboard containing forty-seven KPIs glowing beautifully on a screen somewhere in the office; I need enough information to understand whether the growth system is healthy and, more importantly, to notice when something meaningful begins changing again.

Because eventually, something will.

 

No growth strategy works forever in exactly the same form because the environment around it refuses to stay still. Markets mature, competitors improve, advertising becomes more expensive, customer expectations change, products evolve and channels that once felt almost unfairly effective eventually become crowded, which means that even a strategy performing exceptionally well today is built on assumptions that will someday stop being true.

The mistake isn’t having a strategy that eventually underperforms.

The mistake is noticing six months too late.

That’s why I would consider a 90-day turnaround unsuccessful if the company finished the quarter with better numbers but no better understanding of how those numbers were produced. A temporary improvement is useful, but a business that still cannot explain why performance changes will eventually find itself in exactly the same situation again.

The first month should give us a much clearer understanding of what changed and why, the second should allow us to intervene where those changes matter most, and the third should give us enough evidence to decide which interventions deserve to become part of the way the company operates going forward.

After that, the work doesn’t end. It simply becomes less dramatic.

A healthy growth strategy should be observed continuously, questioned periodically and adjusted when the evidence gives us a reason to adjust it, rather than being rebuilt every time a disappointing quarter finally makes an old problem impossible to ignore.

If I had ninety days to turn around an underperforming growth strategy, that is ultimately what I would be trying to build: not a miraculous quarter, not a temporary spike in a dashboard and not another collection of tactics that happen to be fashionable right now, but a company that understands how its growth actually works well enough to recognize when the system begins changing.

Because the best outcome of a turnaround isn’t a strategy that never underperforms again. Markets are far too unpredictable for that.

It’s a business that notices when performance begins deteriorating, understands what the numbers are trying to tell it and knows where to look before a disappointing quarter has the chance to become a disappointing year.