This is part 2 in a series detailing the wonderful and overly full seminar, Converting Visitors to Buyers at Search Engine Strategies New York. (Read part One).
The third speaker was Michael Sack, Director, SEM technology & Development, Idearc Media Corp. I feel personally involved with this presentation because I “fixed” it for him on my computer when he brought his thumb drive to me in desperation. 😉
Michael thinks that we as online marketers do a bad job at orchestrating the experience online. He actually thinks that the way homepages are implemented are the worst thing to happen to retailing, but he wants to say it in a nice way.
Too many websites have too much clutter on their home page – ridiculous amounts of links, too many pictures, not a clear path on what to do or even where to start. This is not what happens in the retail world. What if you walked into a grocery store and there was just a giant full shelf in front of you – and someone said, “Now SHOP!”
Let’s get some milk
We have all shopped for milk at the store (unless you are lactose intolerant, but even then the soy milk is still back there too). Why is the milk always in the back of the supermarket? Because the store wants to orchestrate your experience in the store. Everyone needs milk, it has a short lifespan (so you have to buy it often), and they want you to see everything else in the store before you get there (and get you on the way out).
Retail experience:
- it’s orchestrated
- flow of store is scientific
- placement of product is being tested
- what people see and when they see it affects buying patterns (and vice versa)
Virtual Doorways
This is a marketers dream: Mike is coming, we know he is most likely to buy milk from isle 6. So we move milk to isle 6.
But: online you don’t know it’s Mike.
But, as it turns out, online we don’t need to know the who, we just need to know what someone is searching for to show them the right door. On the web, we can put the milk anywhere.
A virtual doorway works like this: when someone clicks on a search phrase (chocolate milk), you should deposit them on your site exactly where they can find the chocolate milk.
Michael wrote a white paper about such web aisles in 2002, but it still holds true.
Michael’s suggestions:
- connect specific searchs to specidifci landing pages
- expand keywords into the tail,
- use longer keyword phrases with 3-5 keywords – they may have a lower search rate, but when clicked, have a higher rate of conversion
- test where your visitors land and what they see
- use analysis to adjust and improve conversions
- if you only have 50 links in google, you have too few (only true for stores with many products)
Howard Kaplan, Chief Operating Officer of my favorite blog right now, (Future Now, Inc.) was the last speaker of the session. In contrast to many of the other very detail-oriented conversion authorities around, Howard seems to have a casual attitude towards the conversion process. He made a comment that made many in the room hold their breath:The secret to conversion is: it’s not about you.Howard says that conversion is a choice by your visitors. Clicks are people, links are decisions. All you can do is understand who your visitors are and what they need. He had the room stand up (god bless him, because I was getting a bit stiff), then down in sections to represent customers who won’t convert.
- The first section you lost didn’t belong on your site in the first place – they were looking for Rolling Stones the band and found Rolling Stones the magazine (the favorite analogy of the conference it seems for misplaced site visitors).
- The second section of visitors you lost were the right customer, but you had the wrong price for them.
- The third section you lost were close but no cigar: they were the right customer, you had the right price, but it was too early in the buying process – or you didn’t answer all their questions right.
- Hurray! The last 25% decided to convert!
So as you can see, it wasn’t your fault! Well, except for those middle two, which were a little your fault. Howard reminded: you can’t reach your goals until customers reach theirs. Here come those archetypes again…
Howard brought up the four archetypes that are used heavily by Future Now, Inc., but in an easy to understand analogy: Seinfeld characters.
Spontaneous Kramer. Competitive Kostanza. Methodical Elaine. Humanistic Seinfeld.
But he added that you don’t need to be a behavioralist to optimize your site, and you don’t need to think in absolutes. You can bring it down to just two:
- quick vs. deliberate (pace)
- logical vs. emotional (bias)
A lot of web optimizing is geared towards competitive user (all above the fold, nav and links need to be perfect).
Roadmap for testing
- Define conversion goals: what action, what page, what’s success look like?
- Define your profiles: how many profiles? What types? Where are they in the buying cycle?
- Do the creative: create the driving point (ppc, email, etc.) Does the message match the motivation, type and buying process?
Where should I test first?
Test where customers are leaving from – high bounce rate pages, high exit pages.
5 testing tips to help increase conversion
- Product images tell a story: make sure your pictures aren’t just perfunctory. Would you rather buy a pear just shown sitting on a table, or a cut pear on a beautiful plate, with juice running down it? Also include multiple images, zoom, and video if helpful. Even things like white papers can be merchandised to be more appealing.
- Test your headlines and copy: do tests with slightly varied headlines, call to action button changes, benefit text changes. For example, Dell increased conversion when they changed one link from “learn more” to “help me choose.”
- Forms & point of action assurances: reassure the customers they are making the right decisions by putting it out there, “we value your privacy,” or “100% satisfaction guarantee”
- Calls to action – get them to click: make it the easiest thing in the world for them to trust you – given them shipping and delivery information early in the cart process,
- Don’t make them wait: remove gratuitious flash or heavy images
Conclusion: Goal in testing for improved conversion is to just move the needle, you don’t need to hit a home run.
