Jonathan Mendez's Blog

Afterthoughts on SES San Jose

SES San Jose 2007 started for me during the one-hour delay waiting for the gate to be fixed so I could board the Sunday night flight from Newark Airport. With all the Search knowledge waiting to get on board it was easy to kill the one-hour delay talking with Kevin Lee, Barry Schwartz and Tamar Weinberg.

Things really kicked off for me the next morning on the Ads in a Quality Score World session. This was the third time for this panel and I’ve been on all three. However, as much as one would think I would getting better at presenting on this subject the more I do it, I’m afraid the opposite is true. Quality Score is such a difficult subject to shed light on for many reasons. The factors in QS are changing, the algo is opaque and it is calculated differently for Google/Yahoo/Live. SEMs struggle with wanting to understand the details of something that if they saw the details would probably make them only more confused.

The mood of attendees seemed tense (consistently no one laughs at jokes in this session) and I’m sure attendees left dissatisfied coming to a session hoping to take something actionable back with them and leaving without answers from panelists (Google) or panelists telling them to “ignore the score” (me). In a strange way I think everyone was glad when this session was over and not for the reasons you would expect.

After lunch it was time for some redemption with Personalization, User Data and Search. I was honored that Chris Sherman invited me to speak on this panel because in my mind it was the most important panel at the show. It didn’t disappoint the SRO crowd in the big room.

On the front end Gord Hotchkiss and Sep Kamvar of Google spoke about the symbiotic and incredibly important relationship that personalization of the SERP has with widget, application and mash-up development and optimization. This is in my mind the most important sea change to impact digital marketing since AdSense. (For more of my thoughts on this check out this post on Optimizing Widgets & Applications this one on Social Media and Search and this one on Brand, Product & Community)

My presentation was about creating dynamic site-side or landing page personalization using parameters from the referring URL string (or cookies) to create more targeted and relevant user experiences. I received a tremendous amount of positive feedback from this presentation so thanks to all of you who took the time to let me know. It’s possible that I will redo it again adding a few more things that I didn’t have time for in the form of a webinar.

After the Monday sessions I had pleasure of having drinks at the speaker reception with Avinash Kaushik who introduced me to his team at Market Motive. I also had the pleasure of meeting Dave McClure who’s taken a keen interest in analytics of late.

Tuesday morning kicked off with Ad Testing Research and Findings. This was my third time on this panel too and unlike Quality Score it is consistently a great session. All the presenters provided some real actionable recommendations. Of course Quality Score reared its ugly head here too and poor Clay Bavor of Google. He had the kamikaze mission of sitting in for Nick Fox on both this session and the Quality Score session. A funny aside and proof that multivariate testing is getting traction - Gord joked after my presentation that I would be holding a discussion group over lunch about Taguchi method orthogonal arrays. After the session four folks actually came up and asked where the lunch was taking place. Next SES we will really make that happen.

Tuesday afternoon I sat in on Offermatica President and OTTO Digital CEO Jamie Roche’s presentation on Landing Page Testing. Jamie steered away from MVT and similar to my earlier presentation focused on what we’re doing with clients through personalization and targeting showing some more case studies and the great results. At OTTO we’re all about results.

Wednesday was a mix of hanging at the Offermatica booth, hitting the session on CSS, AJAX, Web 2.0 & Search Engines, networking, meetings and catching up with email. I took the red-eye back from SJO Wednesday night and found it hard to believe there was still another entire day of program content ahead at SES.

This was my 11th SES show as a speaker. It will be interesting to see what happens with SES now that Danny Sullivan and Chris Sherman are no longer involved in programming it. My gut feeling from looking at the new SES advisory board is that there will be less emphasis on industry “insiders” and more emphasis on client side speakers sharing their experiences. Whatever SES morphs into, it has played a crucial role in building recognition for search engine marketing and optimization.

Bravo to Danny and Chris and also to Rory, Stewart, Karen and everyone else whose hard work over the years went into allowing SES to help legitimize Search. It was only a few years ago that Search was a marketing afterthought and regarded with skepticism. Now it is a multi-billion dollar a year industry and fundamental marketing practice for businesses of all sizes. If there's one thing that everyone was saying at SES it was that "this thing of ours" is still only a baby. Search in 2012 will look little like it does in 2007. The learning never stops...

Microsoft says Brand Search Ads a Waste of Money - Show Me the ROI!

I have to chime on this since the timing of this Microsoft release and SES is not coincidence. It's certainly attention grabbing but I think it's almost irresponsible to make these claims without the right data to support them.

In an article backed with little ROI data Microsoft/ Atlas is saying that it is a waste of money to put ads on your branded queries. This runs counter to my own extensive research that I'll be presenting at SES and I've posted before. My research is backed up with conversion data and RPV. Nowhere is this mentioned in the AdWeek article leading to believe it was either not measured, does not support their story or AdWeek missed the most important aspect of the report. I've yet to read the report but I hope to get more color on it tomorrow as I'll be on a panel with Nico Brooks from Atlas.

Ad Testing

Ad testing can lead to huge increases in conversion rates. I don’t care if it is text based search ads, rich display and banner ads or video ads - all benefit site-side from advanced testing techniques done on the ads leading to these landings.

We know marketing messages have a tremendous influence to persuade action. This effect is magnified the more relevant the ad is with present or emerging goals and intentions. Yet testing and optimization of ads does not get the respect or attention it deserves.

Current display is testing is stuck in the dark ages. Most search ad testing focuses more on CTR than conversion. In both channels creative is not tested as strategically as it should be considering the large amounts of spend and the impact that optimized ad creative can have to a campaign’s ROI.

• How many ad tests that your agency or business is running are validating segments?

• Are you using multivariate arrays for testing performance?

• Can you track campaign performance experience in a holistic and transparent way?

If the answer to any of these questions is no (or don’t know) than be prepared for a competitor to eat your ad ROI lunch over the next 6-12 months. Ad testing is just scratching the surface. In search we’re getting into advanced testing on match types and targeting (I’ll talk about testing display in an upcoming future post).

I’m at SES San Jose this week evangelizing and presenting case studies on testing and targeting at three sessions - Ads in a Quality Score World, Ad Testing Research & Findings and Personalization, User Data & Search. I hope to have these posted here by the end of the week. In the meantime Offermatica will have booth on the expo floor so stop by and say hello if you are at SES.

A Digital Marketing Lesson from Wall Street’s Woes

As the New York Times points out in Sunday’s Business section the recent collapse of the stock and bond markets has much to do with the automated black-box systems of hedge funds whose algorithms are seeing behavior that despite all their statistical modeling has no precedent. The automated systems have failed and it’s now people that need to come to the rescue.

From the article The Week When Risk Came Home to Roost by Gretchen Morgenson:

None of this would be a problem of course if fund managers were not relying so heavily on precedent to make their decisions. Computer models seem so perfect, so scientific, so flawless and they are advertised in precisely that fashion. Ingenious models lull investors into a dangerous complacency…

Just like in the investment markets black box algos in the marketing world will lull marketers into complacency and will never improve upon human results. This should be a reality check in our industry around the current buzz and euphoria for the automated behavioral targeting and "set-it and forget it" suggestion tools getting loads of hype and capital.

Why aren't these automated systems always going to work?

Having done hundreds of tests I can tell you from personal experience that just like the investment markets so much of what is seen in consumer behavior is counterintuitive and could not have been predicted.

Financial markets do not always trade in a way that is typical or predictable. And when they deviate from the norm, all the wonderful and smart trades stop behaving according to plan.

C-level execs and other decision makers investing in these platforms should understand there will be risks risks associated with them. The risk will be magnified if they are depending on these automated technologies to replace other tools and technologies controlled by the human marketer.

Every marketer would love to just sit back and collect great results - but just like Wall Street - in the markets where we play risk is never gone because while behavior is measurable and quantifiable, it is never predictable.

That's why they call it marketing.

Jay-Z & Rocawear Lay Down the Viral Video Marketing Beat

I sell ice in the winter, I sell fire in hell
I am a hustler baby, I'll sell water to a well
- Jay – Z from “U Don’t Know”

In the lyrics to “U Don’t Know” from Jay-Z’s monumental September 11, 2001 release The Blueprint (amazingly one of the other top three albums from 2001, Dylan’s Love and Theft was also released that day) the Jiggaman closes his verses with “I will not lose.” Well Jay (Shawn Carter) has taken the refrain and is using it as the basis for a cross media marketing campaign for his Roccawear clothing brand featuring celebrities and individuals that have overcome trials and tribulations. The “I Will Not Lose” campaign launched last week with viral video and it too is a blueprint.

And if somebody woulda told 'em that Hov' would sell clothin
Heh, not in this lifetime, wasn't in my right mind

While last year Budweiser foolishly thought Jay-Z was going to sell them more beer (Bud sales slipped over 7% last year), Jay does know how to sell lifestyle better than just about anyone...and in any case, what lifestyle does “Bud” represent? (ok, please don’t answer that)

This time success is the aspiration Jay is selling. He pulls it off remarkably well. The campaign is the best execution I’ve seen since H&R Block's “Me & My Super Sweet Refund.” That contest last year delivered to the most linked to comedy video in the history of YouTube.

If we are to believe the I Will Not Lose website it was Hova himself that thought up this idea up and forked out a measly $450k to produce it and get it out. The fans seem to like it as well. Meaning it is already a huge success. Throughout his career Mr. Carter has understood rule 1 of brand marketing; a brand can never lose the respect of its fan base.

The vids are shot beautifully by former chief photographer for Rolling Stone Mark Seliger and are the perfect mix of subject, story and product being used to help define a brand. The Jonathan Vilma clip is over 100k views on You Tube since it was uploaded two weeks ago with over 50,000 views the past two days alone. That’s “viralocity.”

Not all the videos are first rate but most provide a great mix of branding, promotion and production that epitomize how the internet video medium should be used by brand marketers. In addition to a contest to help fuel the viral element a feel good factor is in play as proceeds from the campaign and I Will not Lose-branded Rocawear merchandise will be donated to an I Will not Lose Foundation, which will distribute funds to charities selected by campaign participants.

Now I'm going to step to Hova with a little advice for the 100 million man. Get a paid search campaign around the names of the artists in the campaign and the name of the campaign itself. Also, you should be tracking and measuring which videos are drawing the most views on his site, not just on YouTube. That way you can optimize the presentation of the videos can get more engagement with your visitors. Also some of the videos seemed a bit long. Test editing them and see if this drives more engagement (views) and more time on site.

I'm smarten up, open the market up
One million, two million, three million, four
In eighteen months, eighty million more
Now add that number up with the one I said before
I, will, not, lose

Testing & Personalization Kill Best Practices

Back in October of last year I wrote about lazy marketing or deciding to implement creative and design strategies based on so called best practices. Last week MarketingShepa published a case study of work we did with a client on button testing that once again proved that even for buttons, best practices need to give way to intelligent inconsistency as predicated by testing at each step of the funnel.

This makes sense when you understand that while there is an overall goal of your media or website each step of a user flow has a specific goal, so the messages, images presentations and yes, even buttons, are likely not optimized if they are the same across this flow. As mentioned in the original post this gets magnified as we start to look at segmentation to understand and address the different needs of users.

I’d like to reiterate an important piece from the original post:

Optimal is changing all the time. With the spread of Rich Internet Applications the web becomes a much more dynamic and engaging place. So too is the way people interact with it changing. This is the natural evolution of user interaction and experience. Those that study HCI observe continued change and can see distinct differences in behavior among segments. Bottom line is that the web is just too dynamic and evolving too fast as a medium to say what is good today will be good tomorrow. Or what worked for you will work me.

This gives birth to a new best practice.

ABT...always be testing.

The absolute results of this will be the great cycle that delivers LIFT:

Learning
Improvement
Failure (faster failure, a good thing!)
Thinking

(Note to self...no more acronyms for a few months)

It’s also worth noting that since the multivariate test results of our initial work with DYMO (that in addition to the Sherpa case study can be seen in more detail here) they have progressed into behavioral targeting. We’re proud that they will be at eTail 2007, (along with another OTTO client, Fingerhut) to deliver a case study on a personalization campaign to owners vs. non-owners that resulted in a huge lift in conversion rate.