When Ads Build Value 當廣告建立價值時 by Nigel Foote, Starcom EMEA Managing Partner

When we talk value in media agencies, traditionally we talk about 'media value' – squeezing media costs in a deal or over-delivering audiences to a campaign – as opposed to building 'brand value'.
當我們(媒體代理商)談到價值時，傳統上我們談的是“媒體價值”－壓低媒體交易成本或者傳遞給超過預估的閱聽眾數量－而非建立“品牌價值”。

That needs to change. Marketing today is in crisis: the battle for consumers' attention; the death of loyalty, the rise of retailer power, and intense margin pressure all mean that marketers need the focus of all their partners on delivering measurable returns. In the media business, that means helping clients to deliver a sales plan as much as a media plan.
這個想法需要改變了。今日的行銷正在轉折點：消費者注意力的戰役；忠誠度不再、零售商的權力高漲與強烈的利潤壓力全部意味著行銷人員必須要他們全部的伙伴專注在實行可測量的投資報酬。在媒體業而言，這代表著幫助客戶實行銷售計劃與媒體計劃同等重要。

THE IMPORTANCE OF BENCHMARKING COMMUNICATIONS PRACTICE 為行銷溝通建立基準的重要性
We should be obsessed with doing things well in addition to doing the right thing; and focusing on return on objectives, not just on a narrow definition of return on investment. At Starcom we believe that one of the best ways to do this is to conduct marketing mix modeling for an individual brand. However, timeline and financial pressure can militate against customised analytics – communications planners often need immediate guidance. In which case, the next-best solution is to have a set of case studies – to learn about normative averages and reasons for differences, and to provide a match when we don't have a tailored model.
我們不僅要做對的事，也應該著眼在把事情做好；並且專注在目標回報率而非狹義的投資報酬率。星傳相信其中一個最佳的方法是為個別品牌塑造行銷組合模型化。然而，時間與財務上的壓力不利於客製化的分析－行銷溝通企劃人員往往需要立即的指導。因此，其次的解決方法為過一套的個案研究－來學習平均的基準與造成差異的原因，進而提供在無法量身訂做模型時的一個對照。

The Starcom Brand Library is just such a database. Covering over 150 brand cases, and drawn from three continents over the last ten years, it is a unique collection of quantified advertising effort and impact measures as well as qualitative analyses of the brand situation. A consistent adstock modelling approach has been applied to each brand case, and there is ad/brand manager analysis of the brand situation across 29 brand, category and communication factors. The Brand Library offers a broad spread of packaged and non-packaged goods brands, although, as is often the case with the application of such analysis, there is a bias to large-market-share/TV-dominant brands.
星傳的『品牌圖書館』就是一個上述的資料庫。涵括來自三大洲在過去十年中超過150個品牌的案例，它的獨特收集在於不僅有量化的廣告活動與廣告impact的測量，也針對品牌情勢有質化的分析。每一個品牌都有持續性的廣告資產模型化，與來自於廣告/品牌經理對品牌、品類與傳播等29個因素的品牌情勢分析。『品牌圖書館』包含了消費性商品品牌與耐久性商品品牌，雖然大多數情形下這種分析都有相同的狀況－偏重於市場佔有率大與電視廣告為主的品牌。

Across these cases we have learnt that advertising builds brand value. However, the more interesting questions are as follows.
How strong are ad effects?
How quickly do they act?
How durable are those effects?
How is ad value affected by ad clutter?
What are the overall learnings on frequency and flighting?
從這些品牌案例中我們學習到廣告如何建立品牌價值資料。可是，下面這些問題是最有趣的：
廣告效果有多強?
廣告多迅速會有反應?
廣告效果可以維持多久?
廣告噪音如何影響廣告價值?
在廣告頻次與波段方面的整體學習?

LESSONS FROM THE BRAND LIBRARY 來自『品牌圖書館』的學習
1. How Strong? 廣告效果有多強?
In determining how strong ad effects are, we take the measure of what happens, on average, to sales when we add 100 gross rating points (GRPs). The average volume share uplift is 1.4% per incremental 100 GRPs, although there is a wide variation in the scale of incremental effects across such a diverse collection of brand cases (as one might expect).
在決定廣告效果的強度時，我們測量了平均每100個總收視點對銷售的影響，發現每增加100個總收視點，平均的銷售量上升為1.4%，雖然在如此多元化品牌資料收集下，增加的效果在我們的量尺中有相當大的差異性（如所預期的）。

What we have determined is the number of 'contextual' characteristics that are associated with the variation in such incremental sales effects. Essentially, sales impacts tend to be incrementally greater for newer brands, for those facing low competition and for those whose brand market share is smaller than average. Conversely, brands with larger-than-average brand share, whose messaging is emotional rather than 'news', and those facing tough competition would be expected to have a weaker sales impact per 100 GRPs. Of course, the beauty of such normative data is its interactive interrogation by a number of different criteria to obtain the best brand 'match' to aid planning thinking.
我們找出了‘內容關聯’特性與增加的銷售效果變化之間的聯結。基本上，新品牌、低市場競爭性品牌與市場佔有率低於市場平均值的品牌會有比較大的銷售impacts。相反地，市場佔有率高於市場平均值的品牌（他們的訊息對消費者是情感性的而非‘新聞’）與面對強悍市場競爭的品牌，他們每100個總收視點對銷售的影響則比較弱。當然，這個常模資料美妙的地方就在於透過一系列不同的基準的交叉詢問找出最速配的品牌來協助企劃思考。

2. How Durable?廣告效果可以維持多久?
Turning to the question of durability, we need to examine the adstock of a particular brand case, where adstock is taken to be the duration of sales impact that advertising has over time, and the 'half-life' for the brand – the time at which half the total sales impact occurs. The importance of this question lies in the implications it has for how an advertiser can afford to have hiatus periods – the number of weeks on/off.
至於廣告效果可以維持多久？我們必須要檢視品牌的廣告資產adstock，廣告資產代表的是廣告對銷售產生影響持續的時間，與品牌的半衰期－廣告對銷售的影響殘留達原先整體效果一半產生的時間點。這個問題的重要性在於廣告主可以承擔多長的空檔時間－即廣告上檔/下檔的週數。

A shorter half-life means that the ad impact for that brand doesn't last long, so it requires more continuity of activity, as opposed to a longer half-life where ad effects persist longer, meaning that advertising activity can be less continuous. Looking at the Brand Library data, there is significant variation among the 150 analysed brands. A 10–12 week half-life for effects is not unusual – and, as before, a number of 'contextual' elements are critical when estimating whether a campaign will have a shorter or longer half-life, as Table 1 shows. Our conclusions are that for older brands or established messages, more continuity of advertising is required. Conversely, for new brands or new ideas, less continuity is needed. (see Table 1)
半衰期短代表品牌的廣告impact不持久，所以需要較連續的活動；而相反地半衰期長則廣告效果較持久，意味廣告活動可以較不連續。檢視品牌圖書館中的資料，150個品牌有顯著的差異：有10-12週半衰期的品牌並不罕見－同時，如表1所示，這些’內容關聯’因素是決定一個廣告活動半衰期長短的關鍵。我們的結論是：成熟品牌或確知的訊息需要連續性的廣告；相反地，新品牌或新觀念的廣告可以較不連續。
TABLE 1: DURABILITY OF AD EFFECTS廣告效果持久度


3. Ad Clutter廣告干擾
So what about our third and final core question; namely, what is the impact of clutter? What happens to sales when we add 100 GRPs to the plan? Higher-clutter markets (like Spain) apparently require more weight for a branded campaign to 'cut though' than others. The contrast between Spain (a high-clutter market) and the UK (a relatively low-clutter market) is shown in Figure 2.
廣告干擾的影響為何？當我們的計劃中增加了100個總收視點之後對銷售有何影響？高廣告干擾的市場（如西班牙）明顯地需要加重的品牌音量來突破重圍。圖2即說明了西班牙（高廣告干擾市場）與英國（低廣告干擾市場）的對比。
FIGURE 2: EFFECT OF HIGH CLUTTER高廣告干擾之效應


OVERALL LEARNINGS ON FREQUENCY AND FLIGHTING 廣告頻次與廣告波段的整體學習
Using the Brand Library, Starcom has a clear read on one of the perennial questions in media planning: 'What kind of frequency do we need to plan for this type of brand and campaign?' Traditionally, debates in this area have focused on the alternatives of 'effective frequency' versus 'recency' theories – that is, looking for some kind of 'trigger' frequency (say 3+ on TV) versus looking to maximize reach while consumers are 'in the market'/near purchase. This debate is a false one – there is no general catch-all.
應用品牌圖書館，星傳對長久以來困擾媒體企劃的’為這一個品牌與活動，我們需要何種頻次?’這個問題有清楚的理解。習慣上，這個領域的爭論集中在’有效頻次’與’近期’理論上面－也就是說，找到能‘引起反應’的頻次（如電視上至少三次）相對於當消費者接近購買時盡可能地與他們接觸。這個辯論是無意義的－因為並沒有通則。

Every brand has different circumstances and objectives, and needs to model its own response data (whether sales or awareness or image tracking) in order to determine properly the right frequency for achieving its objectives.
每個品牌的情況與目標不同，需要模型化它的廣告反應資料（不論是銷售或者知名度或者形象追踪）來正確地決定達成品牌目標對的頻次。

So while there is no single 'truth', the Brand Library databank has allowed us to plot brands (see Figure 3) and campaigns in terms of the scope for adding frequency without adding inefficiency (horizontal axis), against whether there is a need for continuity or not (vertical axis). The quadrant that any brand is in has fundamental implications for the type of plan that is most efficient for that brand in terms of weight (and frequency) planning, and flighting. We have found that for a majority (73%) of brands studied, continuity is king. While this would appear to support the idea of recency, some 52% of brands would also benefit from heavier weights when on air. This leans towards a frequency argument: these are brands that face intensive competition in low-interest packaged-goods categories.
所以當沒有單一通則時，品牌圖書館資料庫讓我們將品牌與活動依增頻次會不會導致無效（橫軸）相對於是否需要連續廣告活動（縱軸）繪製作圖3。每一品牌所在的象限皆有企劃上最有效率的廣告頻次與廣告波段的含意。我們發現研究中多數的品牌（73%）連續性是最重要的，這一點似乎支持近期理論；也有52%的品牌因上檔時有加重的聲量受益，這又偏向支持頻次理論：這些品牌來自競爭激烈但關心度低的消費性商品品類。
FIGURE 3: LEARNINGS ON FREQUENCY AND FLIGHTING (% OF BRANDS IN DATABANK) 廣告頻次與廣告波段的學習（品牌在資料庫整體佔比）


EVALUATION FRAMEWORKS FOR JUDGING COMMUNICATIONS VALUE 判斷傳播溝通價值的評估架構
Should there be a frequency or reach focus? Like most things, it depends. The vast majority of brands benefit strongly from continuity; and most of these can benefit from heavier weights when on-air.
所以我們到底應該著重頻次或者接觸廣度？如同這世上大多數事情一般，必須視情況而定。眾多品牌強大受益於連續性廣告波段；而其中多數因上檔時有加重的聲量而奏效。

However, there are certain categories that do not fit this pattern. With the Brand Library, we are not able simply to isolate what quadrant each brand is associated with, but in addition can determine what category each quadrant is associated with. This gives communications planners a valuable normative comparison for brands that might not yet have response analytics from their own data.
然而,有些品類並不符合上述模式。從品牌圖書館，我們不僅簡單地獨立與每一個品牌關聯的象限，也能找出每一品類相關的象限，讓傳播溝通企劃人員對尚無法有品牌本身反應分析資料的品牌，有一個有價值的參照基準。

Using such a database enables planners to construct the right hypotheses for what a campaign will achieve, given similar conditions to studied cases. This is often an essential support in developing the right evaluation framework to judge a campaign's true value in driving sales and other response measures.
使用如此的資料庫企劃人員在與研究案例有近似條件的情況下可以對廣告活動將獲得的效果形成正確的假設。這通常是發展一正確的評估架構來判斷一廣告活動真正價值促成銷售或其他反應測量的基本方式。

SUMMARY總結
Increasingly, media practitioners have to be champions of brand value (the value of sales and brand equity among consumers) as much as media value. Simply put, understanding how the brand performs in the marketplace allows us to plan and buy the media better.
如同創造媒體價值一般，媒體從業人員逐漸增加地必須成為創造品牌價值(銷售與消費者心中的品牌資產的價值)的戰士。簡單地說，明瞭品牌在市場上如何運作有助於更好的媒體企劃與購買。

While there is no substitute for real-world learning on individual brands, brand-effectiveness databases such as the Brand Library are very useful in helping determine some core planning issues.
當個別品牌真實世界的經驗學習無可取代時，像『品牌圖書館』這種協助判定一些核心企劃議題的品牌效果資料庫非常有幫助。

Media agencies need to drive value growth by harnessing such knowledge management techniques. Only by doing so can the industry be seen to transcend the pure buying of media value and be viewed by marketers as their partners in actively creating brand value.
媒體代理商需要運用這種知識管理技術來帶動價值成長,惟有如此,這個產業才能超越單純的媒體購買價值，主動創造品牌價值進而被客戶視為的事業伙伴。