Numa desaceleração económica, devo cortar o orçamento de marca?

Gerir a tesouraria com rigor é obrigatório, mas a marca é o corte cuja fatura não chega agora. A evidência de eficácia mostra que quem corta a voz numa crise cede quota, que depois recompra mais cara na recuperação. Corte, mas o que for dispersão, não a posição.

O que diz a evidência sobre manter publicidade numa recessão?

A análise de Peter Field ao IPA Databank sobre a recessão de 2008 mostra que a quota de mercado responde mais fortemente à share of voice na crise: um excess share of voice acima de 8% rendeu crescimento de quota com índice 167, contra 46 para quem subinvestiu.

Porque é que recuperar quota de mercado depois de cortar é mais caro?

Porque a quota entregue não fica à espera: a memória arrefece, a presença apaga-se, outro ocupa o lugar. Recomprá-la custa preços de recuperação, mais altos que os de manutenção. A análise PIMS mostrou o fosso: mais 1,7 pontos de quota na retoma para quem manteve a voz, contra 0,6 para quem cortou.

Manter a marca na crise dá retorno garantido para a minha empresa?

Não, e quem o promete vende certeza que a evidência não tem. Os dados do IPA e do PIMS são correlação em grandes anunciantes, enviesada por seleção, não uma garantia para uma PME. Mostram um padrão forte e repetido. O retorno de um negócio concreto mede-se caso a caso.

Qual a diferença entre cortar custos e cortar a marca?

Cortar um custo administrativo é parar de pagar uma conta: o dinheiro que não sai não pede mais nada. Cortar a marca é entregar uma posição relativa, que um concorrente ocupa e que se recompra mais cara na retoma. Um alivia a tesouraria; o outro vende um ativo barato.

Cutting the brand in a crisis: the cut that comes back charging interest

In a slowdown, the first line of the budget to be questioned is almost always the same. It is not the rent, it is not the salaries, it is not the inventory. It is the brand. It is cut because it seems the most painless cut: at the end of the month the saving shows up in the spreadsheet and nothing visible is missing. That is precisely the trap. Of all the cuts within reach of anyone running a company in economic uncertainty, the brand is the only one whose bill does not arrive now. It arrives in the recovery, and it arrives with interest. This article is about that decision, in the concrete context of 2026, about what the evidence on effectiveness says to those who make it, and about where that evidence stops.

The moment the question becomes urgent

No one cuts the brand in the abstract. They cut it when money is tight, and money is getting tight. In the Spring 2026 Forecast, published on 21 May, the European Commission revised the growth of the Portuguese economy downwards, to 1.7% this year, and projects inflation still above comfort, at 3.0%. The report describes a sequence of shocks early in the year, severe storms in January and February, a sharp rise in energy prices in March and April, and a preliminary stagnation of activity in the first quarter. It is against this backdrop, of weaker growth and costs still pressing, that the brand line returns to the board's table with a cross beside it.

And it is a legitimate question. In a slowdown, managing the treasury with rigour is not an option, it is an obligation. What is proposed here is not to spend against the current on faith. It is to understand why this specific cut behaves differently from all the others, before signing it.

Why cutting the brand is not like the other cuts

Cutting administrative expenses, travel or fixed costs is stopping payment of a bill. The money that does not go out asks nothing more. Cutting the brand is another thing: it is surrendering a position. And in a slowdown that position is measured in relative terms, not absolute ones.

The concept that organises this is called share of voice, the slice of a category's advertising investment that belongs to a brand. It was formalised by John Philip Jones in the Harvard Business Review in 1990: brands whose share of voice exceeds their market share, the so-called excess share of voice, tend to gain share over time. The implication in a crisis is the point that changes everything. When the whole category falls silent at the same time, those who keep investing see their share of voice rise without spending another euro, because the denominator shrank. Those who cut do the opposite: they disappear from the field while a competitor who held on buys, with the same budget, more presence than it would buy under normal conditions. Cutting the brand does not save a bill. It cedes ground, and it cedes it to whoever stayed.

The proof: those who hold the voice buy the recovery's position

This is not agency rhetoric. It is the most consistent finding in the advertising effectiveness literature on recessions. In 2020, Robert Brittain and Peter Field gathered it in a report for the Australian and New Zealand markets, with Field running a specific analysis on the IPA Databank, the British archive of effectiveness studies. Field isolated the campaigns that were in the market during the 2008 and 2009 recession and compared them with those that ran two years before or two years after. The conclusion is direct: market share responds more strongly to share of voice in a recession than in normal times.

In numbers, taking as a base of 100 the effect of the same strategy in normal times: campaigns with excess share of voice above 8% during the recession produced annualised share growth with an index of 167, that is, 67% more effective. And the penalty for under-investing worsened: campaigns with excess share of voice of zero or less fell to 46, less than half the effect of normal times. In profit the pattern repeats, with an index of 114 for those who invested above 8% and of 43 for those who under-invested. The recession amplifies both sides of the scale: it rewards those who keep the voice and punishes those who silence it, more than it would under normal conditions.

A second line of proof, older and of another nature, points in the same direction with a figure that matters to whoever signs the cheque. The Malik PIMS analysis studied about a thousand business units in developed economies, across contractions and the recoveries that followed them. In the first two years of recovery, the units that increased share of voice during the downturn gained 1.7 percentage points of share, against 0.6 for those that cut. Return on invested capital recovered 4.3 points for those who increased, and fell back 0.8 for those who cut.

And here is the honesty missing from those who quote only the good half: the same analysis shows that increasing share of voice during the recession itself has a slightly negative effect on the short-term return. It is not free. Keeping the brand in a crisis is a decision to accept a tighter margin now to buy a recovery position. The return is in the upturn, not in the quarter in which you spend. It is exactly for this that cutting the brand charges interest, and why it is the only one that charges it: the saving is booked today, the bill, larger, is booked later.

The asymmetry: buying share back costs more than holding it

The decisive word is asymmetry. The share surrendered in a cut does not wait, neutral, to be regained when the economy improves. The brand memory allowed to cool, the presence that fades, the place another occupies in the meantime, all of it is bought back at recovery prices, dearer than those of maintenance. Susan Coghill, at the time chief marketing officer of Tourism Australia, summed it up in a way any director recognises: it costs far more to recover share at the end of a crisis than to invest to stay active during it. It is a manager's testimony, not a number. The number is the recovery gap that PIMS and IPA have already shown.

There is a third reading that corroborates the pattern from outside advertising, and it counts for its scale. In 2010, a Harvard Business Review study, Roaring Out of Recession, by Ranjay Gulati, Nitin Nohria and Franz Wohlgezogen, examined 4,700 listed United States companies across three recessions, between 1980 and 1982, 1990 and 1991, and 2000 and 2002. About 80% took more than three years to recover their pre-crisis sales and profit growth rates. Only 9% actually prospered, coming out beating their sector rivals by at least 10% in sales and profit growth. And the winners were not the ones who cut deepest. They were the ones who combined cost discipline with selective investment in the right areas.

What this does not prove

None of this proves that keeping the brand will make more money for your concrete business in the next upturn. Anyone who guarantees you that is selling you a certainty the evidence does not have. The caveats, because without them the numbers are worth nothing.

The IPA Databank is made up of self-selected entries to effectiveness awards, the best work of large advertisers, submitted by the agencies themselves. It is indicative, selection-biased, and measures correlation, not a controlled experiment in which the same company cuts and maintains in parallel. PIMS is modelling on business units in developed economies, from contractions prior to 2000, and it too is correlation. The Harvard study is about business strategy as a whole, cost discipline versus selective investment, not about advertising in particular. And the European Commission projection is exactly that, a revised projection, not an accomplished fact.

There are two further distinctions that separate this thesis from a slogan. Maintaining is not increasing blindly: holding a defensible position is not pouring budget into a slowdown without a thesis to justify it. And the share of voice in these studies is that of large brands with measurable advertising investment, not the reality of a company with limited means and a narrow category. What the evidence offers is a pattern, strong and repeated, not a promise. The return of a concrete business is measured case by case.

Reframing the cut: saving on the account, cost on the balance sheet

Seen through the quarter's spreadsheet, cutting the brand is a saving. Seen through the company's position, it is selling a cheap asset in a buyers' market, to buy it back dear when demand returns. It is the difference between what appears in this month's income statement and what appears in the company's value two years later.

So the right framing is not expense versus expense, it is expense versus asset. Short-term demand generation is a bill: you pay it, it produces, and it disappears when you stop paying. Brand authority is a position that, if it is defensible and sustained with method, is worth more tomorrow than it costs today, and that the competition does not match easily. David Aaker defined, in 1991, brand equity as the set of assets, and liabilities, linked to the brand's name, that add or subtract value. In a crisis, the blind cut turns the asset into a liability: the brand that fades does not stay at zero, it subtracts, and the subtraction is charged in the recovery.

None of this is an argument for cutting nothing. It is an argument for knowing what you are cutting. Not all of the brand budget is a defensible position. Part of it is dispersion, presence spread without a thesis, and that is cut at no cost at all. The difference between the cut that saves and the cut that charges interest is not in the size of the scissors. It is in knowing, before using them, which of the two things you have in front of you. At NP we treat that distinction as what it is, a question of an asset that is measured and defended, not of style, and it is what the Method makes auditable, line by line of the budget, rather than a matter of opinion. We exist to convert perception into a financial asset, and the hour to prove it is precisely this one, when surrendering that asset seems the safest decision and is the most expensive.

Sources

Every number in this article was verified against the primary source or its reference publication. The macroeconomic figure was anchored to the European Commission because it is the primary source we could read directly. Where the source does not support the current reading, or has an interest in the result, we say so in the text. It is the least that is asked of anyone who claims that the brand can be measured.