Marketing Tactegy

The Full Measure on MeasureUp (4 of 4)

MeasureUp is a great conference for all things marketing ROI and return on marketing investment ROMI put on by the IIR.  They put on some great events and I highly recommend you take a look at their site for some upcoming events.  To summarize this series of 4 posts the Measureup conference there were a few major points that I came away with:

1. Marketing Mix Modeling may have peaked. Marketing Mix models have provided great insight over the years, but lately many have been shown to only deliver short term insights. When used in a competitive, mature category, such as, for the Oreo brand, MMMs typically deliver short term insights in that temporary price reductions can increase sales. Unfortunately, they also lead to loss of brand value. MMMs don’t include the long term brand value that can be destroyed through continuous price actions. Consumers simply get trained to always buy on ‘deal’.
2. Social media although is growing and is important for many brands, it still has a long way to go to be a major contributor for some of the top consumer brands. For the rest of us, though social media can still drive a lot of brand engagement, it can be measured and will continue to grow in importance.
3. Analytics and measured insights are becoming more and more important in marketing decision making. Marketers are finally embracing analytics and using them to their advantage. There is still a long way to go, but it is progressing swiftly.

There were many more insights and I hope next year you’ll have a chance to participate as well.  Just let me know if you would like further information on any of these topics.

The Full Measure on MeasureUp (3 of 4)

Here are continuing remarks and key notes from the MeasureUp conference.  This is the third in a series  (You can also read my Twitter stream at #ROMI)

1. Melanie Davis, Director of Marketplace Insights at American Express had some great insight as to how American Express uses modeling to support their brand.  They analyzed their presences on Twitter and Facebook with the goal to track online buzz compared to their competitors (primarily bank credit card issuers) using share of conversation and sentiment.  It was modeled using Structured Equation Modeling to determine how online buzz drove their funnel (Awareness, Engagement and Consideration, Conversation, Loyalty and Usage).  This is a great example of using the community engagement funnel (described in our social media white paper.  Download now.  One of the other interesting findings was that even though there was a lot of buzz it was relatively small compared to the number of purchased impressions made using traditional media.
2. Building a business case for marketing, social media and ROI were discussed in my session. With Ted Molter, Director of Marketing at the San Diego Zoo and Wild Animal Park and Nick Popielski, Director of Marketing at Atlanta Gas Light Resources.  Both had compelling stories on how they were able to drive analytics and numbers into their marketing organizations with some astounding results.  Nick Popielski showed how ROI was the only way to get the attention of the C-Suite and provide a solid foundation for the marketing business case.  Understanding the causes of churn led to significant growth in their net customer base.  Ted Molter spoke primarily about online marketing ROI showing how online media including social media was able to drive high conversion rates and ROI.  The numbers reached into the 1,000% range for ROI.
3. Ed Keller of Keller-Faye provided some great insight on their TalkTracker tracking survey.  This service provides both on and offline measurements of word of mouth.  Social media must be looked at from an online and offline perspective when comparing online and offline word of mouth conversations.  77% of them still take place offline according to studies and research conducted by Keller-Faye.  They estimate that over 3.3bn brand conversations take place every day in the US alone.  The top talked about brands are:
   • Coca Cola
   • Verizon
   • Walmart
   • Pepsi
   • AT&T
4. Ellen Stone, SVP of Marketing at the Bravo Network gave probably the most complete discussion on calculating ROI for social media.  There is no question that social media marketing can be very effective at driving buzz and delivering value for Bravo network.  They were using a number of fascinating tactics to drive buzz and deliver incremental viewers to their programming.  This was especially true for the launch of new programming.
5. Unfortunately I wasn’t able to capture the name of the presenter from Publicis, the agency for Denny’s restaurants.  My apologies.  Nevertheless he had some great statistics from their superbowl campaigns.  Denny’s super bowl offer of a free breakfast after the 2008 Super Bowl was wildly successful driving value and sales for the brand.  Unfortunately the advertising campaign was not sustained after the first quarter and it lost its momentum that could have been easily and cost effectively sustained.  Here are a few interesting facts:
   • Add-ons and late arrivals paid for the cost of the ad and the food at the 1,500 Denny’s restaurants
   • 2 million guests enjoyed the meal
   • Restaurant locater on the web was a key metric to determine interim success.  The ad delivered over 14,000,000 hits to website in 12 hours.  It generated over 2,300 news hits, 500 newspaper articles.  It was even mentioned by Jay Leno and Jon Stewart.  “Denny’s” became the 2nd most googled word.
   • Many guests heard about the free meal through text, calling, Facebook and mass emails to friends
   • It took over a million chickens to lay the eggs required.
   • Through this ad, Denny’s was able to meet its financial goals for the quarter.

Still more to come. Stay tuned for the next installment. Let me know what you think of the insights gained.

The Full Measure on MeasureUp (2 of 4)

Here are continuing remarks and key notes from the MeasureUp conference.  This is the second in a series  (You can also read my Twitter stream at #ROMI)

The second day focused more on optimizing traditional media.  The keynote was from Avanish Kaushik, Analytics Evangelist at Google, Inc.  He is also the author of Web Analytics: An Hour a Day  and Web Analytics 2.0 

Avanish is a lively, engaging speaker.  Definitely worth the price of admission.  His primary message relating to web (and of course, Google analytics) is get started, make mistakes, learn and improve.  That is, learn to be wrong but do it quickly.  He provided an impressive array of tools that could be used to quickly analyze the impact of web-marketing on conversion and web design.  There were so many examples of using google analytics to increase conversion it was hard to keep up.  Here are a few:

• How to reduce bounce rates by pointing to design flaws
• How to use experimental design to continue to optimize page layout and design.

With the keys to the kingdom Avinash was able to show that sites could easily be significantly more effective.  One of his thought provocations was, “don’t have opinions, have hypotheses.” 
Avanish also introduced a new term, Hippo decision making.  The Highest Paid Person’s Opinion.  How often is this true when there are no analytics to back up the conversation and decision making. 
His last bit of advice was to continuously ask for feedback from your consumers about their experience with your website with the following three simple questions:

• Why are you here?
• Were you able to complete your task?
• If not, why not?

Here are a few more notes from the other speakers:

1. Scott Deaver, EVP Strategy for Avis Budget Group had a few good insights about his business. Unfortunately I wasn’t able to catch his entire talk but he mentioned a few interesting comments concerning metrics:
2. • Linear metrics
   • On/off metrics
   • Good enough metrics

   He left us with the statement “Don’t fall victim to the tyranny of the measurable.”

3. Dominique Hanssens has a fascinating message in that great marketing doesn’t just drive revenue and profit but it also drives the share price. If marketers can get it right, they can deliver long term incremental share prices. He also pointed out a few potential flaws in the valuation on the stock market. For example, he found that differentiation was not well understood (that is, not reflected in the stock price) but awareness and brand equity was. He also found that customer satisfaction clearly has a long term impact. It would be interesting to find out though whether positive customer satisfaction has a long term impact on share price, but does a negative have a short term impact. For example, how does the negative impact of the recalls for Toyota impact the share price? Is it a long term impact as well or has it hit the stock price in a short term way.
4. Lisa Wellington of Coca Cola and Greg Michaels of Kraft provided some great insights into their methodologies relating to marketing ROI and ROMI. Coca Cola puts the rest of us to shame when we look at how they have a comprehensive system applying globally. They are a big user of Baysian regression, pooled models, conjoint testing, copy testing and scenario planning with highly granular models, looking to understand the relative influence of brand, flavor, pack type, pack size and region. Even in the face of the imprecise nature of the data, they are still able to make great decisions to drive their business forward based on good consumer and modeling insights.
5. Kraft, not far behind showed how modeling has helped them to drive great decision making with their brands. Greg’s point of building a good model now as opposed to a perfect model later illustrates how they are able to use modeling to structure the decision making and conversations marketers have when they endeavor to build their businesses. One of the books he recommends is The Black Swan by Nasim Taleb.
   
   One of the interesting points highlighted by Lisa Wellington related to the great social media divide. Although social media wasn’t discounted for its value to help many brands, the problem is that it is still, for some of the major global brands, such as Coca Cola and American Express miniscule. For Coca Cola in particular, even though it has one of the largest followings on Facebook (5Million, I believe is the quoted number), that represents less than 0.1% of the world population (their target audience). Traditional media is still very valuable and will remain that way for a long time at reaching the masses when and where they want to be reached.

Still some great more insights from the conference. Stay tuned for my next post.

Measuring the effectiveness of social media

One of the hotter topics right now in metrics is how to measure the success of social media.  Many social marketers are looking for the answer to this seemingly simple question.  The good news is that there is a lot going on here.  As I’ve become involved in metrics and have spoken around the world on this topic I’ve learned that there are no easy answers.  There is however a framework with which to develop an answer.  (If you’d like, please visit www.marketing-calculator.com for information on a generalized marketing effectiveness framework.)

At one level the number of visitors to a social site can be one measure of success.  At the next level we have to measure the level of engagement.  Time spent on site, the number of comments, videos or other content submitted can be an indicator of success in the social media space.  Unfortunately, these measures don’t determine how successful these social marketing activities are at delivering sales in the stores.  Just like there is an indirect link between TV and in-store sales, so too, is there an indirect link between a consumers’ social media activities and sales in the store.

So we need a measure of transforming this emotional engagement into brand awareness/consideration and preference/purchase intent.  Once we’ve done this, there are a number of tools that we can use to actually calculate the effectiveness of a social marketing activity.

But this is only backward looking.  As marketers, we want to look forward into the future.  How well will this next social marketing activity drive awareness, engagement and sales in the store?  For this we need to use a number of different tools (not complex) and then put the pieces together.  However I will address this in my next post.  In the meantime if you’d like more information or discussion on this topic, feel free to reach out to me on LinkedIn (www.linkedin.com/in/guyrpowell) or at www.DemandROMI.com.

 

New marketing effectiveness blog in Singapore

Marketing Institute of Singapore

I wanted to let you know that I have been invited to blog on the Marketing Institute of Singapore’s site (Marketing Institute of Singapore ).  If you’re interested in learning about general marketing conversations and about marketing in Asia, take a look.

Marketing Calculator

With my new book out (Marketing Calculator ) and the general level of discussion in this marketing ROI area, the discussion should be pretty lively.

I do have a question for you and that is, what do you see your marketing investments to be for the next two quarters (Q408 and Q109)?  Will spending be made to gain share in a down market, or will the finance team win and marketing will be cut to maintain margins?

I look forward to your views.

Last Touch Attribution

DemandGen Report just published an article of mine on Last Touch Attribution (The first word on Last Touch Attribution, DemandGen Article).  This term clarifies one of the key areas of marketing measurement and accountability.  A related term has to do with marketing contributions from an Integrated Marketing Campaign which I will talk about in my next post.  Last Touch Attribution was coined a year or so ago to help marketers communicate some simple steps relating to marketing ROI. It refers to attributing all of the weight of the results to a single marketing activity.  It disregards all other touches, whether they are prior or parallel touches using other media.  Synergies, halo and cannibalization are ignored.  So if a single direct mail piece is dropped all of the responses to that piece are attributed directly to that direct marketing activity.  All brand effects are ignored as well as all other components of other marketing campaigns.

Last Touch Attribution

I just spent a day at a ROMI (Return on Marketing Investment) session with a major high tech company that is actually pretty advanced for their ability to track marketing costs and results.  Marketers have been struggling for a long time to finally differentiate and communicate this type of measurement with the measurement of response to an integrated marketing campaign, where all of the direct results can’t be directly measured.  In my example above, response codes clearly delineate where the specific attribution came from (ignoring all other unmeasured effects.)   In an integrated marketing campaign, each of the campaign elements contribute to the overall response, both in the short term and in the long term (e.g., brand value). 

Integrated Marketing Campaign Contribution

So the IMC Contributions represent the value that each of the campaign media elements contribute to the final result.  There are a few other terms that must be covered to deliver the final picture.  But those, I will talk about in my next blog.

If you have any comments, please let me know.

Sneak preview of my new book

My new book, Marketing Calculator:  Measuring and managing return on marketing investment is about to be published and should be available shortly.  It has been exciting but also time consuming to actually go through the publishing process.  Nevertheless it will be available in the next few weeks, initially in Asia and then in September in the US.

New marketing effectiveness concepts

There are a number of new concepts that I think you will find valuable, namely:

• The Marketing Effectiveness Framework
• The Marketing Effectiveness Continuum
• The Marketing Effectiveness Culture
• Last Touch Attribution(TM)
• 4P3C1E Framework

Drive Marketing ROI

Each of these will help marketers to prove and improve their marketing effectiveness to increase ROI and drive revenue profit and market share.

Stay tuned.  I will have a bunch more to say in the coming weeks.

Multiple airings, interesting statistics and marketing ROI

I read with interest the article on Nielsen’s new measurement statistics (MediaDailyNews, Nielsen Sings New Ratings Tune: 'We Can Be Heroes, Not Just For One Day', September 26th, 2007) and I was surprised to think that the new numbers to be provided will be the unduplicated reach for the new show “Heroes” on NBC.  This will certainly be very important but some additional research will also be enlightening:

It would be interesting to know:

• How many viewers will watch it twice
• How many households will be counted only once, even though different family members will view it during the first airing and others the second airing

Since we know that every impression received by a viewer adds additional value and marketing ROI to the brand and their advertisers, just providing the unduplicated reach won’t tell the whole story.  Secondly it will be interesting to see how this will work when we finally get commercial viewership which is what the advertisers really care about.

Surround-Sound Marketing

Surround-sound, or 360 degree or integrated marketing campaigns have become hot topics in the last few years and have enabled marketers to design and deliver very effective marketing campaigns.

Surround-sound marketing and psychologics

Rex Briggs and Greg Stuart describe the process in detail in their book ‘What Sticks,’ but let me give a brief overview here.  At a high level the interaction between media is more than just providing multiple touches about a specific message.  The idea is that the messages are reinforced between each other.  In particular, the concept of ‘psychologics’ means that, for example, a print ad actually helps the mind conjure up the imagery of a corresponding, previously viewed TV ad.  Because of this the value of the print ad is actually higher than the value of a print ad where there was no corresponding TV imagery.  This is probably the reason why many non-TV ads are improperly attributed to TV because they trigger the mind to conjure up the images from the TV commercial.

If, for the moment, we leave aside the active or passive exposure to a media type and if we assume that the viewing of a commercial on television delivers the most imagery, because it has both video and audio components, then it should be that a single viewing has the largest influence.  Radio has only an audio component.  Print has a visual and written component (and sometimes odor), but does require more active participation to read and absorb the message in the ad.  In this case in order to get the biggest bang for the buck out of a surround sound marketing campaign, then the most imagery oriented ad should be run first followed by the other media.

Traditional media analytics may be wrong

This has a lot of implications to help marketers improve media buying and flighting.  It also has implications of how marketers measure the results of their surround-sound marketing campaigns.  Simply correlating impressions of the various media elements to bumps and wiggles in the outputs may provide incorrect conclusions as to what is working and what isn’t.

We now need to know which consumers actually saw one ad only, two ads from different media and so on.  Only with this detailed viewership model can we truly see how effective one media channel is versus another.

If you have any comments on this I would love to hear them.  Please let me know.

Product Sampling and Marketing Analytics

I was recently involved in an interesting discussion on the use of sampling which I will attempt to repeat (in a disguised fashion) here.  Certainly sampling can be a great way to seed the market with to induce trial and hopefully repeat consumption.  But the question is how do we account for the results from the sampling?

Multi-step marketing

Let’s say that we offer samples of a new product at some major event.  Let’s assume it is a food product costing less than a dollar per serving.  The assumption is that after the sampling event, the consumer will go home and then purchase it again (and again) in the traditional outlets of food, drug or mass.  As opposed to a typical television commercial, this is a two (or more) step process begun with the initial sampling event leading to further purchase actions.

Sampling and marketing mix modeling

The traditional marketing analysis technique, marketing mix modeling (MMM) has difficulty capturing the results from a sampling event, because it doesn’t capture the true purchase dynamics.  MMMs try to correlate the bumps and wiggles in the marketing inputs with the bumps and wiggles in the outputs, sales to determine the relative impact of each of the marketing activities.  In the case of sampling, a multi-step marketing tool, the follow-on purchases can’t easily be determined because the increases in the output bumps and wiggles don’t easily correlate back to the inputs, although some synergy elements can be tried.

Because of this, attribution back to the sampling activity is potentially misrepresented and in the usual case under-attributed.  Smart marketers know this and continue to use sampling knowing that the traditional analytics don’t correctly account for these types of multi-step marketing tools.

Agent-based modeling and multi-step marketing

In order to measure these results or those of other multi-step marketing activities, we need to use other analysis techniques.  Survey techniques can determine the propensity to repeat immediately following the initial trial and deliver reasonable results, although they may not be able to include all of the nuances of synergy and other surround-sound marketing effects.

In order to properly account for sampling amidst all of the other marketing activities, a consumer-centric approach is required and agent-based modeling can be one of those methods.  Because agent-based modeling can accurately represent the impact of this multi-step trial and repeat process it can easily deliver the necessary analytics to determine the true impact of sampling on the marketing mix.

There are many other examples of multi-step marketing.  Certainly in the business-to-business marketing environment where marketing delivers leads to a sales force is one.  Catalytic marketing also has elements of a multi-step approach.

If you have other examples of multi-step marketing that fall into this discussion please let me know.