Ogi Radić asks- Can finding out what drives online ad awareness help create better ads? March 3, 2010
Posted by Mark Blei in : Staff posts , add a comment
The advent of online advertising more than a decade ago yielded a “new” metric – online ad awareness – a measure of online ad memorability. Dynamic Logic at this point stored at least 4500 campaigns that measured this concept in MarketNorms. It’s high time we’d started thinking if this measure of online campaign noticeability gives us any further insight into its connection with, say, brand favorability or purchase intent. If we can predict what drives ad awareness, then we may be able to gauge the ads effect on persuasion.
The first thing the client always asks about is industry benchmarks. Looking at several different categories in the last 3 years, we do notice some differences in terms of category. Entertainment campaigns are on average more noticeable than, for example, automotive ones. This means that category inherently affects the ads noticeability.
In turn, how about investigating ad awareness by looking at performance per online ad awareness? How are other campaign metrics in campaigns faring in the context of their memorability? Are more memorable campaigns more likely to have increases in persuasion? What about those campaigns where online ad awareness decreases? What does this decrease even mean in practical terms?
Using MarketNorms I aggregated 26 banner campaigns with best increases in memorability, and 26 with highest decreases. Interesting things emerged as findings there. Those campaigns where online ad awareness decreases also had accompanying decreases in brand favorability and purchase intent.
The best performing campaigns had increases in favorability and purchase intent.
This means that if we figure out what a decrease in online ad awareness means, then we can potentially prevent it as it’s detrimental for the brand.
In conclusion, if the message causes people to remember the campaign less after being exposed to it, that’s not good for the brand and ought to be avoided. In creating ads, the client needs to appropriate the ads for the industry, and more technically, to avoid having the ads cause a decrease in brand awareness. Our task here at Dynamic Logic is to figure out what the decrease means, and we’re presently working on it.
Ogi spends his free time reading everything he can get his hands on, going to the gym, and discovering good restaurants around NYC.
Lauren Rodas Discusses Recruitment On Video Ad Networks And Caribbean Office Envy March 2, 2010
Posted by Mark Blei in : Staff posts , add a commentHi Everyone—
My name is Lauren Rodas and I am one of the first Project Managers hired here for the Safecount Puerto Rico team. So far it’s been a little over a year already since our office here in Puerto Rico was created and it’s been great. There are so many interesting events that happen on a day to day basis with being a Project Manager here in Safecount Puerto Rico. It is exciting as a Project Manager to never know what to expect. New things are always happening such as crazy requests from publishers and the different reactions when they see ‘Puerto Rico’ in my signature. A lot of envy and countless “how’s the weather?”. Learning new stuff is a constant thing especially when you’ve been here for only a year. I think the best thing about working in Safecount is the way new improvements come out and change frequently. I’ve been a part of testing with Peanut Labs which will help supplement recruitment with social networks and it’s helped me achieve success in my studies.
Recently, I’ve worked a lot with Video Ad Networks and I have to admit it’s a challenge and I have heard that there are many ways that we are trying to better recruitment on those networks. The in-layer banner survey is one thing I can’t wait to see come out. We also found it very successful to recruit off of an in-video overlay but it’s only an option when the Network has already sold the ad slot. Hopefully with growing new technology, we can move into the direction to be able to recruit well over all Video Ad Networks.
Online advertisement is growing fast and we are all learning how to adapt our current technology to keep Dynamic Logic/Safecount as one of the top names in Marketing Research/Data collection companies.
Michelle Eule discusses the relationship between click though rates and branding metrics March 2, 2010
Posted by Mark Blei in : Staff posts , add a commentMore than ever before, agencies are feeling pressure from advertisers to leverage the “real-time” metrics available in the digital environment to optimize their media plans on the fly. Certainly, one of the many benefits the medium offers is the myriad of metrics, many of which are available in real-time, and the ability to make quick changes to media delivery.
Typically, the metrics used for optimization online are behavioral in nature, such as click-through or interaction rates. While such metrics give an appropriate indication of performance for campaigns with direct response goals, they are not a good fit for campaigns that have longer-term branding goals. A study conducted by Dynamic Logic and DoubleClick in 2009 showed only a very weak positive relationship between click though rate and branding metrics. People may interact with ads for reasons unrelated to the brand message, such as interesting games. With such a weak correlation between the metrics, using behavioral data to optimize a branding campaign gives, at best, an incomplete picture of campaign performance and, at worst, a deceptive picture that can lead to bad decisions.
While branding metrics are certainly the ideal criteria to use for optimizing a branding campaign, there’s also a danger in making changes to the media delivery too early in a campaign. Branding metrics that are more attitudinal in nature, like Favorability and Purchase Intent, generally take time to develop. Our normative data show that multiple exposures to a campaign, which help reinforce a brand’s message, ultimately help build lower funnel brand metrics. Often, the best creative or site placement in a campaign doesn’t show its full impact until consumers have seen it multiple times. If an agency makes changes to site placements or creative rotation too early in a campaign, they may be moving impressions away from a critical campaign asset before giving it a chance to make an impact.
A client interesting in using sophisticated models for planning and optimization could leverage Dynamic Logic/Millward Brown’s marketing sciences group to build tools that incorporate survey-based branding data with normative information and external data sources. For clients that are looking for a simple and inexpensive solution that provides a quick read on branding metrics to make optimization decisions early in a campaign, Dynamic Logic recently launched a new solution called Adometer. The metrics used in Adometer are recall-based, such as brand recall or message recall, which are the building blocks to generating an attitudinal shift in consumer opinion. These recall-based metrics, meanwhile, aren’t as influenced by frequency levels as lower-funnel attitudinal metrics like brand favorability, and therefore offer better guidance early in a campaign. They simply offer an early gut-check as to whether consumers even notice the ad at all, recognize what brand is being advertised, and understand the message or key benefit the brand is trying to communicate. Comparisons in performance can be made between creative themes and between site placements so that the agency can reallocate impressions towards the best performing ones. The Adometer solution uses short surveys, usually 4-5 questions long, so results can be collected quickly and delivered in real-time via an online interface. Because no control group is needed, there is also no need to worry about balanced audience profiles.
If a client is looking to understand their campaign’s performance across the full funnel of branding metrics, and needs deeper insights into why a campaign is performing well or poorly, then AdIndex is a better solution. However, for an early read into what’s working and what isn’t while it’s early enough to make changes, Adometer may be the right solution.
Michelle Eule is Managing Director, AdIndex Solutions at Dynamic Logic. In this role, Michelle is responsible for maintaining and enhancing Dynamic Logic’s suite of solutions for analyzing the in-market branding impact of digital advertising. She also acts as a consultant on strategic accounts and customized projects. Michelle joined Dynamic Logic in July 2002. Prior to joining Dynamic Logic, she managed a research laboratory for a professor at Columbia Business School, with whom she studied marketing and organizational behavior from a social psychological perspective. Michelle received an MBA at NYU’s Stern School of Business and a Bachelor of Arts in psychology from Barnard College. Michelle has been a speaker at many industry events including conferences hosted by The ARF, OMMA, AMA, Ad:Tech, and OPA, and has been a guest lecturer at Baruch College and NYU. Michelle was a judge for the 2008 IAB MIXX awards and is a member of the Research Committee for the Advertising Council.
Gustavo Sanchez talks about opening new markets ( and comes clean about his obviously secret moonlighting gig as a crime fighter) February 15, 2010
Posted by Mark Blei in : Humor, Staff posts , add a commentMy name is Gustavo Sanchez. I have been a project manager for Safecount for little over a year now, working out of Puerto Rico offices. Recently I had the opportunity to work on the first Ad-Index study in Puerto Rico.
We worked with a major ad agency here on measuring the online components of a major ad campaign for a large advertiser. Our main study objective was to assess the effectiveness of the online campaign at increasing awareness, ad awareness, sponsorship association and purchase intent. Since our product is completely new to Puerto Rico (The Puerto Rico office mainly supports the New York office) it was a very rewarding experience to work on. We worked with US publishers on this study, but the challenging part was introducing our product to some local sites, specifically local news outlets ENDI.com and Wapa.tv, since they were not familiar with survey recruitment. Even though we faced some issues, recruitment was remarkably positive, and all people involved were really responsive to our demands.
Some of my interests outside of Safecount include music, family and friends. And comic books.
Rebecca Rahmanian has a confession to make February 15, 2010
Posted by Mark Blei in : Humor, Staff posts , add a commentHello! My name is Rebecca Rahmanian. I work as a Research Analyst in the New York Office ..
I wouldn’t admit this to most people, but I have a bit of an obsession with celebrity gossip. I like to read a lot of celebrity news columns and blogs such as People, Perezhilton, TMZ, The Superficial, Dlisted, and so on. For me, it’s mostly about the fashion. But I can’t lie; I’m also intrigued to know what’s going on in their lives. Which, at times, can be more interesting then gossip overheard at the Dynamic Logic water cooler.
Last May, our researchers were given the opportunity to conduct research in their field of interest. At the time I was working on two beauty campaigns, both involving celebrity sponsorships so exploring the effectiveness of such endorsements was an easy pick for me. Although celebrities are commonly used in advertising, their influence on the brand and its consumers seems somewhat unclear. In my research I hoped to discover how a celebrity’s presence in an online ad impacts a consumer’s recognition of the brand and their purchase decision.
With the help of our MarketNorms Database, I was able to isolate campaigns including celebrities against those with unknown humans and identify their effect on Aided Brand Awareness and Purchase Intent. My findings indicate that the endorsement of a brand by a celebrity can be persuasive. Yet, their presence may demand too much attention, taking away from the advertised brand. Thus the appearance of Ellen Degenerse Catherine Zeta Jones or Tiger Woods may attract internet viewers to the ad, but not necessarily generate awareness for the advertised brand. As a result, if the primary goal of a campaign is to increase awareness of a brand, a celebrity’s presence may not be helpful.
Meanwhile, consumers may find a celebrity’s endorsements to be sincere, leading to an increased likelihood to purchase the brand in the future. As a result, the presence of a celebrity in an online ad campaign which seeks to increase sales may be more effective.
While celebrity endorsements can have both positive and negative effects on an ad campaign, the one constant is that at the very least, they will generate a second look by avid gossip mongers like myself.
Amanda (Zuniga) Baskett talks about the final product January 21, 2010
Posted by Mark Blei in : Places you wish you were, Staff posts , add a commentThere are many aspects of my job that I really enjoy. However, there is nothing that tops the end of a project when all the data has been cleaned and weighted and it is sitting there waiting to be understood. Sometimes I think I must be a pretty sad person to be so excited about this (my older brothers reassure me that this is true), but then I also tell myself that this is the reason that I’m a researcher.
I love to tell stories – which isn’t surprising given my journalism and English background. In this job data serves as the basis for my story and the words are intertwined to shed light and explain what the data reveals.
In some cases, there are end results that are perplexing at first glance. However, by taking a deeper dive into the data as well as understanding what environment the campaign is running in helps to clarify the results.
For instance, the final results in one of my recent projects were negative especially given that the mid-way results showed positive movement. Instead of simply looking into various data cuts and writing a report based simply on the numbers I began my investigation with the numbers and quickly moved onto scouring secondary resources to help me explain what happened.
In the end I determined that the overall impact of the campaign was masked by negative results from a particular timeframe. Negative press surrounding the product category combined with respondents’ exposure to the ads triggered severe responses to the brand. Thus the overall poor results were not due to ineffective creatives or off-target messaging, but were driven by outside factors beyond the creative, media and brand teams’ control.
Nonetheless, a key lesson was learned. The results and insights have led the client to further investigate whether it makes sense to put a hold on a campaign if presented with a similar scenario in the future.
As a researcher we are not only tasked with understanding what the
data reveals, but to also shape a story that is meaningful for the client. There are often projects were there are different nuggets of data which you can mold an entire story around. It then becomes important to recognize what story would be most beneficial to fashion for your client. As we are all aware a media agency and a brand team have differing interests and markers of success. At the end of the day it is the combination of unraveling the data, finding the golden nuggets, and crafting a client-minded story that makes me enthusiastic about my job as a researcher.
In my spare time I like to eat and run – everyone always teases me for taking vacation days off to run a marathon or doing relay races through the mountains. I guess it is not everyone’s version of vacation. However, I know how to relax too the picture below is from my recent honeymoon in Hawaii.
Meg Reardon Talks about the power of looking “Beyond The Click” January 19, 2010
Posted by Mark Blei in : Staff posts, Uncategorized , 1 comment so farHere is a greaty example of the power of our AdIndex solution being able to tell the story of what happens ‘beyond the click.’
I recently completed an AdIndex study for a CPG client. The data from the media company regarding click-through rates and interaction was spectacular, indicating that click-through was much higher than the industry average and that interaction with the client’s ads drove great increases in consumer’s perceptions of and relationships with the brand.
Had that been the only data provided to the client, as is often the case without the additional insight provided by an AdIndex study, the campaign would have been considered a huge success and perhaps would have been replicated in the future.
However, since our AdIndex study provides insights on the impact of a campaign among those that click, but also on the other 98% of consumers who were exposed to the ad but did not click, a critical insight emerged. Based on click-through data, the campaign was working well to drive a specific message because the consumer had time to absorb and process the entire intended message. However, if a consumer did not interact with the ad or visit the client site, we noticed in our AdIndex data that there was a significant amount of confusion regarding the main message of the campaign that was leading to an erosion of brand perceptions.
Meg Reardon Research Director Dynamic Logic
This insight proved invaluable as creative was being drafted for the 2010 campaign. With the insight provided from our AdIndex study, the client was able to re-focus messaging and avoid potential further erosion of brand perceptions due to the confusion revealed in our study.
Just a simple, straightforward example of how important it is to not only understand the behavioral impact of a digital campaign, but also the attitudinal impact on the overall health of the brand.
!
Creative vs. Research – Can’t we all just get along? By Christopher Bian January 14, 2010
Posted by Mark Blei in : Staff posts , 1 comment so far
In the realm of the online advertising world, as with any other advertising discipline, how the creatives look and feel is perhaps the largest contributor to the success of a campaign. As you’re probably aware, the creative process for any advertising campaign can be a delicate one. Often, research can throw a proverbial wrench in the machinery of the creative discipline by “telling” the creative minds to “do it this way.” But I firmly believe this does not need to be the case; research is simply the compass that allows the creative mind to focus its energy towards achieving a specific goal. The key is to understand how to interpret and use the wealth of knowledge that the industry has amassed, as well as thinking forward to define the next set of valuable metrics as the technological capabilities of digital marketing evolve.
When thinking of analytics that can provide guidance for a creative team to work with and help maximize the impact of the campaign, there is more than one method. Copy testing is commonly used to test performance of multiple creative concepts against a panel of chosen consumers. Real-time diagnostic tools are available, such as Adometer, which allows marketers to evaluate chosen metrics or demographics and make tweaks to optimize the creative or media mid-flight. Normative data, which I’ll be focusing on here, has been aggregated over an extended period of time and allows marketers to delve into trends that may help shed light on the current climate. But remember, this type of research tells us what has or has not worked in the past, not necessarily what will work. When it comes down to it, a holistic approach is necessary to properly plan. There are multiple factors one must consider: nuances of the media plan, the type of product or category, goals of the campaign, as well as the metrics necessary in order to define branding success (lead generation, conversion, social action, ad interaction), just to name a few. In order to give a simplified example of how a marketer could use normative data to help guide the creative process, let’s look at interactive ads in the online space.
Here, we’ll explore the differences in impact of interactive creatives and what it’s implications on the execution of your creatives might be. For the purpose of the examples, some or all of the following aspects may be used:
- ability to interact with the ad
- display unit size of 300×250
- types of site the ad will be placed on
- campaign goals of new vs. established products
MarketNorms [Last 3 Years, N=2,377, n=3,697,267] shows that a 300×250 interactive unit has an overall larger impact on ad awareness (4.2% vs. 2.9%)* than a 300×250 ad that was not interactive. Both were featured on portals, which generally reaches a wide audience.
In this instance, the interactivity seems to have played a role in affecting ad awareness, which opens up a greater opportunity for your creative team to introduce an engaging experience with the brand and product.
While the interactive units impacted awareness metrics, there was no statistically significant impact on persuasion metrics. This does not mean that interactive units can’t help your brand drive intent or consideration, but rather, understanding the audiences you are attempting to reach is but one key to developing a successful creative concept. Take for example, the data amongst 300×250 interactive units that were advertised on niche site categories (compared to a typical portal such as MSN) in food/beverage, health/beauty, lifestyle, parenting and shopping.
^ Niche Sites defined as sites within the food, beverage, health, beauty, lifestyle, hobbies, parenting, family and shopping categories
The data showed a statistically significant increase on intent to purchase of 1.2% after exposure. All units advertised were interactive, yet, respondents exposed on niche sites claimed an increase in favorability and intent to purchase. Compared to those exposed on a typical portal, awareness metrics were impacted. We know that portals such as MSN or Yahoo! receive a wide range of visitors, reaching a general audience. Meanwhile, the visitor of niche sites like WebMD.com or Epicurious.com has come for information regarding something specific, and therefore, a higher incidence of relevance to a properly targeted product or service is more likely to occur. The question remains: how can we convert this information into creative output?
This example of data comparison serves as a supportive normative insight; if interactive ads are to be used, tailor the interactive elements to better align with consumers who are at different points of the purchase funnel, indicated by types of sites the consumer is to be exposed on. For example, the interactive elements for portal ad units could touch on the overarching features and benefits of the product in a simple, concise way in order to introduce the brand to those who are not familiar with the product. For established brands on niche site categories, your creative team will have the challenge of providing a more involved experience to those who are already aware of the brand and closer to any actual purchase behavior. Perhaps a store locator or promotional deal can be included within the ad unit to guide these consumers along the search/purchase process. Furthermore, a combination of both can be used for product extensions or medium awareness brands.
Through the normative data, it’s clear that targeting of media channels that properly align with your brand’s goals is necessary, precluded by the need to understand your brand’s audience on each channel in order to provide more value to the consumer at the creative level. While this may be one and albeit somewhat simplified example, it demonstrates that data can provide a healthy framework on which creative decisions can be implemented, without severely impeding on the creative process.
Every campaign is bound to be different due to the varying aspects that come into play: product category, high end vs. low end products, existing awareness of the brand or parent brand, budget, campaign goals, etc. But research is what can help tie it all together to give your campaign the best shot at success. After all, great creative can live in harmony with strong research trends and principles!
*Difference is statistically significant at 95% confidence, MarketNorms – Q3/2009
Alejandro Molina checks in with some helpful tips January 14, 2010
Posted by Mark Blei in : Staff posts , add a commentHi. My name is Alejandro Molina and I work on the tech team in the NYC office as a QA/Support Analyst. I was once an avid user of ad blocking software. But as time went on, I saw site after site close down due to lack of funds. These days, it’s hard to not see how important online advertising can be for sites that need the money to stay afloat. Even so, it’s sometimes hard to see online advertising in a positive light when I run into intrusive ads that take over the webpage and interrupt what I’m doing. Working at a company that tells these online agencies that such ads should be frowned upon is a good feeling.
In my time here, I’ve learned a few other tips about making online advertising more effective. Make sure your ads are not intrusive. Don’t interrupt the user’s experience if you can avoid it. Make the message relevant to the user. If I’m on a website that targets a specific audience, make sure you are advertising product that really is relevant to that particular audience. Also, make sure your ads are perfectly clear and eye catching without being annoying. Attention spans online only last so long, that if your ad is confusing or takes more than a few seconds to get the point, you’ve already lost half your audience scrolling down the page or clicking to somewhere else.
Keep to at least these points and even a former ad blocker like me will pay attention to your ad.
Lauren Mendelson discusses “Umbrella Branding” January 7, 2010
Posted by Mark Blei in : Staff posts, cute animals , add a commentMy name is Lauren Mendelson and I’ve been at DL for nearly 2 years, working out of the New York City office as a research analyst.
I recently conducted an analysis for a car manufacturer, who was trying to market four new vehicles at once. The goal of this campaign was quite lofty: the car manufacturer wanted to increase awareness of these four new cars as well as to increase intent to buy the models.
This type of online campaign is known as an “Umbrella Branding” marketing strategy. The key to these types of campaigns (according to Millward Brown’s March 2008 POV) is to have a consistent and clear brand identity across the various components. With Umbrella campaigns, when many models or brands are being advertised, research illustrates that only a few brands benefit and others get ignored.
This was the case with this campaign: while the creatives were synergistic and had a similar look and feel, not only to each other, but to offline advertising as well, some models received more attention than others. In this case, a luxury sedan and a minivan were more memorable and favorable than either the SUV or the coupe being advertised.
Additionally, this campaign differed from their past advertising in that it focused on multiple vehicles, which accordingly had different targets, different demographic variables, and different psychographic compositions. By trying to appeal to so many groups, this campaign did not resonate with their main target as well as past campaigns have, which were centered on only one type of vehicle.
Umbrella campaigns are similar to co-branded campaigns in that they often cause the viewer confusion. Often times at Dynamic Logic, we see that co-branding or Umbrella campaigns can have a negative effect; in this case, the creatives did not cause respondents to react in a negative way, but at the same time, the campaign did fall short of achieving all of its goals. With these types of campaigns, it is necessary for each car model to have its own moment of clarity, its own moment to shine.
It will be interesting to see how this car manufacturer decides to advertise their new suite of products. Will they continue to have ads that have a barrage of vehicles, or will they only feature one model at a time? I guess we will have to wait and see!
When not conducting research, I enjoy photography, traveling, and spending time with my new puppy.
