MN Search – Mobile

Excellent event! ¬†Three superb presentations; Aaron Weiche’s presentation is probably the best one on mobile marketing topic I encountered so far ūüôā

The most interesting concept from Aaron’s presentation is complexity of technical issues in our world, and rather limited impact mobile strategy can create (at this point in history ūüėČ ¬†).


Mobile User Experience starts from user’s objective. ¬†Why would you access from:

  • targethome?
  • road?
  • inside the store?

According to where the person is the need for the information and user experience varies.  Questions need to be addressed:

  • Who?
  • Where?
  • Which device?
  • Goal?

After this exercise all user goals and business goals are listed and prioritized.


Prioritization allows company to agree on what is more important (and what content is more important) before any content has been created. ¬†Then, the content prioritization is applied to the site’s architecture. ¬† All of these activities happen before design phase starts.


Basic principles of mobile:

  • Keep it simple in terms of design
  • Keep it simple in terms of content
  • Reduce “digging”
  • Take advantage of native capabilities

Know thy user, and YOU are not thy user

Proliferation of mobile is happening fast, Even Google thinks that mobile is faster than they anticipated. ¬†However, mobile has significant challenges and marketers seem to be struggling with these challenges. ¬†Ha! ¬†It feels identical to our struggle with first generation of web sites ūüėČ

almost literal translation of the web site into mobile site does not result in a good site from user’s or marketing perspective.


How navigation is handled in mobile? ¬†Use “hamburger” icon. ¬†Even Microsoft is using it ūüėČ


Mobile channel can be served differently, and companies have different approach to their mobile customers.


Responsive design – the site design is “fluid” – one content that is presented differently on different devices; content is the same, some pieces of content may disappear depending on the browser size.

Dynamic (or adaptive) design – the site design is specifically adjusted for certain type of devices; content is the same, but it is positioned differently based on the specific device where the site is viewed.

Parallel design – the mobile site is completely different from the main company site.

Wonderful example of mobile specifics – mobile situation suggests that most likely the user would find phone number that can be tapped is more important than other information normally presented on the main site.


Interesting points:

  • Responsiveness of the site is not a ranking factor
  • Google bot and Google Mobile bot are different
  • Mobile SERP is changing more often (Google)

Need for testing of major changes:  one organization was going through brand change and took advantage of the moment with major home page upgrade, including HTML5, parallax, etc.  Unfortunately, the conversion dropped 20% and nobody noticed the drop for a few months.  Now the company has a rule of testing all changes for the home page.

Screen sizes – a list of popular devices with corresponding screen sizes (and popularity!)


Aaron’s complete presentation:

Eloqua Users Group – EE13

EE13Everybody seem to be preparing for the Eloqua Experience – what is very exciting. ¬†Topliners boasts a group dedicated to the event… ¬†this is probably where we can learn about any additional parties. ¬†ūüėČ

Tip from the Eloqua Users Group – there will be no printed materials this year – bring iPad or download an app. ¬†My strategy? ¬†Print the schedule in advance!! ¬†ūüôā ¬† the most popular object during the last Minneapolis Mobile March event was a black and white printout of agenda… ¬†even if the event produced the app and each room sported a QR code ūüôā

3rd party cookies:  browsers are becoming less hungry for 3rd party cookies, which are used by Eloqua application now.  Eloqua is in the process of transitioning to a 1st party cookies for this purpose.

Another interesting development – Google is developing an AdID to replace 3rd party cookies completely – this development is interesting to watch. ¬†Though we, marketers, worry…

BMA – Marketing and IT

EduardoEduardo Conrado, Senior Vice President РMarketing & IT, Motorola Solutions,  shared his perspective on evolution of relationships between marketing and IT in his presentation to BMA audience.

Eduardo describe organizational evolution from products to solutions, to trusted adviser to the customers of the business. ¬†All parts of the organization work on “solutions,” but not nervelessly IT. ¬†IT can report to CEO, or IT can report to a different parts of the organization. ¬†If IT reports to different parts of the organization, it is either Operations or Finance. ¬†If IT reports to Operations, it transforms itself to support the supply chain. ¬†If IT reports to Finance, it is concentrating on optimizing it own processes and operations. ¬†However, if the organization as a whole wants to be organized around customer solutions, IT needs to report to Marketing.


Eduardo mentioned “Start with Why” and understanding what makes the company unique. ¬†Purpose driven companies outperform competition.

Motorola combined IT support for marketing and digital marketing into one group that is reporting to marketing.

Interesting: responsibility for data resides in the front office not in the back office.

Working with vendors – vendors educated the company on what is possible in technology.

Ideal marketing hire is both a quant and creative.

MIMA – Facebook for Business

phoneInteresting – this was the first event in my memory where conversation about prevalence of mobile was illustrated not by an image of an iPhone ūüėČ

As Facebook considers itself a publication and a platform at the same time, it concentrates on mobile first.

facebook-studioDuring popular vacation times general advertisement engagement activity decreases, but not on Facebook – it remains the same.

48% of people use Facebook while exercising in a gym ūüėČ

A note was repeated several times during the event: any social marketing initiative needs to start from business objectives and not “social metrics.” ¬†Generating followers or likes is not a business objective ūüėČ

Facebook maintains Facebook Studio, which works as an award-based recognition forum for marketing initiatives. ¬†Though my attempt to find b-to-b examples lead only to one – Jive – a promotion of social business software (what somewhat makes sense for Facebook environment). ¬†Interesting point of the campaign is a superior performance of Facebook comparing to email and even Wall Street Journal on the main objective – download free trial of social business software ūüėČ


Coursera – A Beginner’s Guide to Irrational Behavior

DukeFantastic class! ¬†Interesting, insightful, and very entertaining ūüôā ¬†Though I read about some of the referenced research before, the entire collection of the class is a spectacular testament to our irrationality as species. ¬†As the class states, this is completely OK, as we adjust our physical environment for our bodies, we can adjust our mental environment to suit our minds.

A very insightful point: subliminal priming. ¬†Subliminally exposing people to different logos (IBM and Apple) influences their behavior. ¬†People primed with Apple logo were measurably more creative in subsequent task and people primed with IBM logo, were slightly less creative than average (per Duke University research). ¬†Daily exposures to brand logos seem to be quite valuable ¬†ūüėČ

Dan presented wonderful research on the size of the incentive (aka “bonus”) in physical and mental tasks. ¬†Though the experiments in this field clearly show the lack of any positive effect of the bonus on performance (and a significant negative effect), it is surprising how often the bonus structure is used in the society.


A slide below shows an influence of a “not so good” offer on a choice. ¬†Given two offers: ¬†an online magazine for less money and online plus printed magazine for more money, people generally select a less expensive offer, ¬†However, if the third choice is added – an offer of printed magazine only for the exactly the same amount as online plus printed, most people chose online plus printed. ¬†Suddenly, the other offer becomes more attractive.


How to sell something using the power of “free”: ¬†more people will buy a desired object, if the object can look inexpensive or free. ¬†Stating that the book costs $10 and the shipping costs $10 is much less effective than stating that the book costs 0, but the shipping cost $20.


Using of the social proof: more people will find the offer or activity attractive if they are informed that other people do exactly the same.  Plus, if the other people can be narrowly associated with the individual, the offer or activity became even more attractive.

A statement in the hotel encouraging people to reuse towels with a statement that 75% of other hotel guests reuse towels is more effective than just a request to help the environment. ¬†The same statement indicating that 75% of the people who stayed in this room reused towels is more effective than, general hotel statement. ¬†It made the “herd” closer to the person who made the decision.


yesAlso interesting: asking people if they plan to do something (such as vote) increases probability that they would do it.  However, asking people to set a plan when they will vote, increases the probability event more.

Also interesting, telling people that few citizens vote and election is close is less effective in generating voter turnout than stating that many people vote. ¬†People seem to like to behave in the same way as “many other people like me.” ¬†ūüėČ

A very important point for me personally to remember is relationship between perception of data and emotion РI love data!  However, it is not recently what can persuade and convince.

Adding statistics into equation dampens our emotional response

More fantastic resources can be found at the center of advanced hindsight


UXPA – Contextual Task Analysis for an Enterprise Mobile App

traneThis event worth the drive to the opposite side of the city in traffic ūüėČ ¬†Very insightful presentation – a wonderful overview of creation a usable app and considering customer and business needs.

A few points from the presentation:

  • trane-app-2One of the reasons for the creation of the app is pressure from the competition. ¬†As apps became more common, system specifications included an app as a requirement and a company would not be considered in the selection process if this requirement would not be satisfied. ¬†Interesting: the app is becoming a part of the product itself – an extension of the functionality rather than a promotional element. ¬†The app increases the attractiveness of the product itself.
  • To be useful, the app requires a specific piece of equipment; so far there were more app downloads than equipment sold – it indicates the app popularity.
  • The company decided not to charge anything for the app – as it was technically the extension of the product and the objective was to sell the product itself.
  • trane-app-3The objective of the app was to make “something useful” so this element of the product will be in the specifications of the requirements during next purchase cycle.
  • Two opposite options existed – to create “alarm log” – a minimalist app from the feature perspective or to include into the app “everything the PC has” – the app already had a PC version.
  • The team used contextual task analysis: ¬†observation of the users in the context of their work (the product already existed online so it was possible to see how users interacted with the existing product.
  • Mobile contexts (from Tapworthy):
    • tapworthyMicrotasking (became for focus)
    • I am local
    • I am bored

Enterprise microtasking:

    • frequently performed
    • time sensitive
    • come up when the operator is away from his desk

Task analysis:

    • Which tasks to support
    • What are desired steps involved
    • What is the environmental context
  • The researches wanted to observe an average customer – not the most proficient, not the worst, but an average. ¬†Several buildings and customers were observed.
  • The researchers did not ask about mobile; the objective was to observe how the current online application was used and what the target audience did during their work day. ¬†Only at the end of the observation the question about the app was asked – in some cases, a particular type of user noted that no app was needed.


  • Two types of system users were identified, which became micro-personas for the effort:
    • Perry – a person who wanted to automate everything , a person comfortable using all features of the system
    • Stan – a person who tend to control equipment manually. ¬†From the perspective of the engineers, this approach might be considered “wrong,” ¬†but it gave the use needed tools for the specific job. ¬†For example, a concerned employee comes to Stan’s cube and asks to increase temperature in his/her location – Stan opens the application and increases the temperature by half of a degree. ¬†Though there will be no effect on the temperature, the person who stopped at Stan’s desk will happily leave and assume that his/her surroundings became more comfortable.
  • Only building operators were targeted for the app, but if the app is easy to use, it would find other users. ¬†At the end, sales people (an unintended audience) liked the app too and used it during the sales demonstrations.
  • Evaluation of the existing online app showed that not everything was understood and used on the main interface. ¬†Equipment graphics were linked and used.
  • The app adapts to a specific building; graphic appears only in the landscape mode


  • Removing labels: ¬†as much as possible labels were removed. ¬†For example, just a word “occupied” was used and it was understandable what it was. ¬†Different colors of font was used for current and desired temperature, and anything adjustable had obvious buttons to adjust the setting. ¬†Removing labels allowed to remove clutter and add more useful data to the screen.
  • Unexpected: from the beginning of the project, adding schedules into the app was considered important. ¬†However, because of scope creep of other essential features, the schedules were not added. ¬†If the question about the schedules arise, the fact that schedule overrides are available eliminates any concerns. This is considered a right decision.

It was a wonderful event, and free coffee from Trane made my online marketing moment a little sweeter ūüėČ

Book – You Should Test That!

bookThis book exceeded my expectations!  It is a wonderful resource with many examples to refer and an elegant methodology that should be easy to follow.

The book is not limited to the theory of marketing testing; it includes a wonderful overview of the relevant aspects of the entire marketing discipline.

A great quote from the book attributed to scientist Peter Medawar

The intensity of conviction that the hypothesis is true has not bearing over where it is true or not.

My notes (not a book summary; a few quick points that I think were explained exceptionally well or were new to me):

  • CRO (conversion rate optimization) and SEO are not mutually exclusive. ¬†“You will achieve better search engine rankings with pages that focus on a single topic or product. ¬†This will improve you conversion rate, too.”
  • bounce-rateReasons not to optimize for bounce rate
    • The page might be attracting unrelated visitors (organic ranking for unrelated items)
    • The page has all needed information and does not require browsing the site
    • Bounce rate can be lowered without any business benefit:
      • changing one page to a microsite won’t increase purchases, but will reduce bounce rate
      • moving key information (pricing) to another page won’t increase sales either
      • adding popular links can reduce bounce rate and also destruct/reduce conversions
      • replacing productive offers with free offers may reduce revenue
  • templatePrioritizing tests: though the home page might be perceived as the most reasonable place to test (traffic!) it might be politically difficult. ¬†Even if the home page is the most popular page of the site, the aggregate of all pages that use the same template can be higher. ¬†This template could be a primary candidate for testing. ¬†Also test most popular landing pages
  • A recommendation to use 5 seconds test (very good idea! ¬†I see it periodically)
  • 3 questions from 4Q (love 4Q!)
    • What is the purpose of your visit to our website today?
    • Were you able to complete your task today?
    • If you were not able to complete your task today, why not?
  • Test pages – 20,000 – 30,000 monthly unique visitors. ¬†Statistically significant result can be achieved within a few weeks. ¬†If traffic is less, the test would take longer.

If your value proposition is strong and the communication of it is weak, you can have large conversion-rate lift from improvements to your pages. ¬†In other words, if people want what you offer, but they can’t figure out what you are saying, simply saying it more clearly will motivate more of them to buy!

On the other hand, if your value proposition is weak, or weaker than that of competitors, you’ll have a difficult time improving your conversion rates, no matter how well designed your website and landing pages are.

  • Conversion rate elasticity – ability to improve your conversion rate
  • LIFT model
    • liftRelevance
    • Clarity
    • Anxiety
    • Distraction
    • Urgency


  • Use anchoring for prices presentation – high price on left (for left-to-right reading locations). ¬†Unbounce¬†has a good pricing example.


Wireframes should be in black and white to avoid preoccupation with design elements and are intentionally not pixel-precise.  At the wireframing stage, you should be thinking about eye-flow, the content hierarchy, and the visitor decision-making process.

  • In general, reducing number of columns improves conversion rate – particularly important for forms. ¬†One-column form is typically performs better.
  • Use graphic elements (direction) – even in tweets. ¬†In a tweet an arrow –> can point to the registration URL, etc.
  • If images on a web page use captions, they are most likely to be read. ¬†They can be called “action captions – and should include an offer statement.
  • softwareImages on the page
    • should support value proposition
    • should be directional leading to the call to action
    • should be high quality for retailers
    • consistent alignment improves clarity and reduces visual processing
    • images also important for companies with intangible products or services. ¬†The challenge is to make a product or service more tangible. ¬†(Software vendors show software boxes, or screenshot of product interface)
    • Do not use “clever” image concepts – they reduce clarity
  • Testing headlines – easiest way to find most powerful motivational trigger message
  • Form option – moving optional form fields to the thank you
  • Adding security items to the shopping cart needs to be tested. ¬†One company placed McAfee Secure symbol in the site-wide shopping cart area and saw reduction of e-commerce sales by nearly 2%

Especially above the fold, your paragraphs should be no more than three lines high.

…sales reps had a 100x greater chance of successfully contacting a lead if the first call was within 5 minutes after the lead form was filled out rather than 30 minutes. ¬†And if the sales rep did not call for 5 hours, their success rate decreased by 3000x

  • If manual entry of phone numbers would be required in order to track revenue, don’t track it. ¬†The inaccuracy of manual phone number entry is much greater than that of estimating phone lead value. ¬†If you can not track phone call revenue, track phone calls over a certain call length as conversions.

The book has a wonderful companion site with downloadable case studies and discussions associated for each chapter.