Blog

We are beaming!  No, Jeff Bezos didn’t mention us by name.  But this is just as good.  In his annual letter to shareholders, Bezos devotes an entire section to the success Amazon has had with their Community Q&A feature:

One recent success is our new feature called “Ask an owner”. It was many years ago that we pioneered the idea of online customer reviews – customers sharing their opinion on a product to help other customers make an informed purchase decision. “Ask” is in that same tradition. From a product page, customers can ask any question related to the product. Is the product compatible with my TV/Stereo/PC? Is it easy to assemble? How long does the battery last? We then route these questions to owners of the product. As is the case with reviews, customers are happy to share their knowledge to directly help other customers. Millions of questions have already been asked and answered.

As the pioneers of the Active Outreach(TM) mechanism for getting fast answers to shopper questions from real product owners, we were flattered a year ago when Amazon first rolled out this feature to see them following our model so closely.  Now, we’ve received the only bigger compliment we could have wished for: this approach has been the same smashing success for Amazon that it has been for the >100 stores that use TurnTo for Community Q&A.

Of course, we haven’t been sitting around waiting for Amazon to catch up.  Our newest version introduces great capabilities Amazon hasn’t got to yet: instant answers, FAQs, category and topic Q&A, a magnificent new user experience, mobile capabilities designed for the omni-channel world, and lots more.  So, Amazon shoppers, keep your eyes open; we’ve got a pretty good idea what you’ll be seeing next, and it’s fabulous!

And if you sell online and are tired of eating Amazon dust, give us a shout.  We’ll show you what the future looks like, too.

Scott Anderson of Iterate Studio sent me an internal memo he wrote last week on the implications of Google’s movement towards semantic search.  It’s interesting and important, and he offered that I could share it.  So here it is:

I pulled the attached article from my favorite SEO/SEM site. It gets into “semantic search” which is the big new thing at Google as evidenced by Knowledge Graph, which is a meager step 1 down a path to answering complex questions for searchers.

Since Google wants to be the place that dishes up answers to questions, the clear SEO implication for ecommerce sites (well, any site for that matter) is to dish up more and more quality answers to relevant questions.

This frankly makes TurnTo an even more strategic solution provider. Using customers to ask and answer questions in their own words for SEO is actually a main reason we adopted it at Vitamin Shoppe.

Of particular interest given Google’s increasing focus on complex questions is the product’s support for category-level questions, which are more likely to be asked on Google than the very detailed product questions.  Again, it’s user generated content so there isn’t a burden on the retailer’s overworked staff.

While TurnTo’s mission is to lower customer support costs and humanize the user experience, the content getting generated is right in the bullseye of what Google wants to see.

Traditional SEO practices will remain essential, but the future is already here.

Promoting the idea of Q&A on eCommerce sites at the category (or “topic”) level the last couple of years has felt a lot like pushing a rock uphill.  On the whole, our customers have been focused on the traffic and conversion benefits to the product detail page.  So at first I thought it was coincidence that we’ve recently had a number of our customers come to us to begin implementation of category-level Q&A.  But what’s really happening has become clear:  businesses are figuring out that more general topic discussion and Q&A content is increasingly important to their organic traffic strategy.  And they are realizing that hosting this sort of discussion on their category pages is a great way to generate it.

Automated personalization and recommendation tools are great at helping shoppers and increasing sales, up to a point.  These tools can make it easy for shoppers to find alternative items that may fit their needs a bit better.  They can propose complementary products that help raise average order value.  They can adjust the selection of items from a catalog that are presented to each shopper to highlight those most likely to be of interest.

But often, shoppers have needs or preferences that can’t be inferred from their browsing history or the profile data you can collect on them.  There’s no way a personalization engine can know that, this time, I’m shopping for a present for my mom, not for me.  And if I was shopping for a present for mom last time I was here, the engine may easily think I’m an 85 year old grandmother rather than a 47 year old guy.  But even if I’m just shopping for me, how is a personalization engine going to guess that our coffee machine is dying and it’s time for a new one?  Or that I’ve just become interested in sous vide cooking?  Or that I had a bad experience with customer service from a particular brand a while back and I’d rather not give them my business?  How do you personalize the shopping experience for these visitors?

That’s where humans come in.  There’s still no substitute for the dialog that happens between a shopper and a great sales associate.  The shopper articulates her needs, and the associate suggests a targeted, creative selection of products to solve them.   That’s personalization!  Sure, it’s old-school, but it’s still the gold standard.

However, this sort of human-powered personalization is expensive to provide – online as well as in stores.  Further, not every sales associate has deep product knowledge or the gift of making that knowledge really useful to shoppers.  The things you can do to improve the performance of associates – higher pay to reduce turn-over and training to increase knowledge – make the cost problem worse.  Provide fewer associates and the customer experience declines – whether it’s live chat or in a store, shoppers don’t like to wait.  And then there are the challenges of addressing spikes in demand, like the holidays.

The solution lies in a hybrid approach to recommendations that combines the ability of humans to come up with creative suggestions with the ability of technology to re-use that expertise and deliver it economically:

• First let your shoppers express their needs by submitting questions on your site.  “I’m looking for a birthday gift for my 85 year old mom.  Here’s some info about her.  What would you suggest?”  “We need a new coffee machine.  Here are some things we want from it, and here are some things we want to avoid.  Which ones should we consider?”  “I’ve narrowed down to 3 sous vide cookers.  Which is going to work best for me?”  (This means your system needs to support multi-item comparison questions!)
• Then get those questions answered from the most appropriate sources for the question –  past customers with relevant experience, your in-house experts, manufacturer reps, and independent experts.  Provide the broadest possible range of answers and opinions, and make it clear what the perspective and background is of each person answering.  And be sure you deliver those answers fast – that keeps the person who asked happy, and it also makes other visitors a lot more likely to ask their own questions.
• Finally – and here’s the key ingredient – put that Q&A dialog in a knowledge base and connect that knowledge base to your question submission form so that the next time a shopper has a similar question, they will immediately see if their question has already been asked and answered.  Those future shoppers will receive INSTANT answers to their questions, and that is accomplished with zero additional work for your staff.

With this approach, you can quickly build up a knowledge base that incorporates the combined wisdom of your customer community and your own product experts, delivering the benefits of true human-powered recommendations with the scalability and cost-effectiveness of automated systems.  You’ll also find that you have  created a valuable resource for your internal customer support team.  And you’ll have the foundation for providing automated, self-service customer support on mobile devices and kiosks.

So don’t limit your personalization and recommendations strategy to automated systems, and don’t give up on human-powered recommendations just because it’s expensive.  Take a hybrid, Q&A-based approach to deliver the best of both worlds.

While at IRCE 2013, George Eberstadt, CEO and Founder of TurnTo, was interviewed by Website Magazine for their Video Spotlight about our social question-and-answer utility for online retailers (Social Q&A). George also shared his thoughts on what e-commerce changes we can expect in the future. Watch the interview below:

I was talking recently to a customer frustrated by his inability to drive meaningful revenue out of Facebook, despite substantial efforts.  The conversation got me to revisit this by now well-discussed topic.

The conventional wisdom is that selling on Facebook doesn’t work because it’s like trying to sell to people drinking at a bar.  (Ref 1.  Ref 2.)  Certainly context matters, but I think this metaphor is misleading.  The same logic could be used to argue that advertising in a newspaper shouldn’t work because it’s like selling to people reading at a library.  Or that advertising on TV shouldn’t work because it’s like selling to people watching a movie.  Or that shops at airports shouldn’t work because they’re like selling to people, um, in the middle of a painful surgical procedure (an experiment on the effects of stress on rats, maybe?).

I’d propose that a lot of the frustration about the difficulty stores have in monetizing Facebook comes from misunderstanding where Facebook fits into the purchase intent cycle.

Purchase intent comes from many sources.  A lot of intent comes from non-commercial sources.  If someone has a baby, they’re going to need a crib.  Trying to create purchase intent for a crib absent a baby is pretty tough.  For discretionary purchases, intent is often sparked by direct experience.  You see something that looks good on someone walking by and you decided you want one, too.  And, of course, purchase intent is sparked by friends.  (One among a gadzillion studies showing this.)  To the degree that purchase intent can be engineered, that’s generally the job of brands, rather than stores.  If a brand can get a consumer to want the thing they – and only they – make, then they are going to make a sale, regardless of the channel in which the consumer chooses to purchase.  Stores generally don’t invest much in intent-generation because of the risk that the money they spend on intent generation comes to nothing when the customer goes elsewhere to buy.  It’s much safer for stores to invest at the intent fulfillment stage.  Thus the success of search engine marketing, which is all about ensuring shoppers who already have purchase intent chose your store for the purchase.  This is also why Sunday circular inserts from merchants generally contain offers on such a large number of products.  The store isn’t trying to convince the reader to want something they don’t already want; they are fishing for intent that already exists – hoping that one of those many special offers will match up with an existing customer need.

So back to Facebook.  I’d argue that Facebook is a pretty good environment for intent generation, and a pretty bad one for intent fulfillment.  This applies both to the organic side of Facebook (friend sharing) as well as the paid side.  This means that Facebook is going to more easily deliver value to brands than stores, and that that value is going to be best measured through traditional advertising metrics like awareness rates and control population studies rather than through online metrics like click-throughs.  It’s possible that with Facebook’s continually improving tools for micro-targeting, they will get to the point of being an economical vehicle for intent-harvesters , but this capability will arrive for some product categories long before others, so sellers should be careful to evaluate the relevance to their particular category before plunging in.

Because Facebook is primarily an intent generation environment, higher-level brand messaging will succeed better than the sort of late-funnel, fulfillment-oriented messaging that works for intent harvesting.  For example, one pet supplies store is having success by sharing cute animal photos that get widely distributed.  The click-throughs are low, but they view this campaign as awareness-building and see the results in site visitor counts and conversion rates over the long run.  Pinboards filled with positive-sentiment checkout comments are also appropriate on Facebook as a tool for intent generation.  Here’s a nice example.  On the other hand, some types of content that are very important for conversion late in the purchase funnel – like customer reviews – are totally out of place in Facebook’s early-funnel, intent generation environment; a product review is just spam to someone who has no interest in the product.

Viewing Facebook through this intent-cycle lens doesn’t produce any magical solutions for merchants to monetize Facebook.  But it does suggest what sort of investments are likely to pay off, and what sort of expectations are realistic. Facebook is going to continue to be a tough environment for merchants because of their focus on intent harvesting rather than intent generation.  And strategies aimed at clicks leading directly to purchases are likely to disappoint.  Stores which take that approach will indeed find themselves in the position of selling in a bar.  But those who find value in developing their store brand and are able to look at brand advertising metrics rather than direct web conversion metrics may yet find water in the Facebook well.

While at eTail West 2013, TurnTo’s CEO and Founder, George Eberstadt, was interviewed by Tim Parry of Multichannel Merchant.  They discuss trends in social tools and different ways to leverage them.  Watch the full interview below:

This post was first published as a guest article on Multichannel Merchant.

Where Amazon leads others follow. No one else has the resources or data that Amazon has to figure out what really works. So it’s a good idea to take notes when they introduce a major new element to the shopping experience.

Rolled out over the last few weeks, Amazon now offers true Social Q&A on most of their product pages. And it’s great. (Disclosure: I’m biased. It works just like the Social Q&A system that my company, TurnTo, provides. Hmm…) Amazon is not the first to introduce this, like they were with customer reviews. But they have leapfrogged the competition that uses customer-service-oriented Q&A with a beautiful execution of the Social Q&A concept. It is designed top-to-bottom for shopper-customer dialog about products; they say that they remove questions that are about shipping, availability, orders, and customer service. And they have built a powerful engine for ensuring that questions reliably get answered by past buyers, providing a great experience for shoppers without creating a massive support burden for Amazon.

The key to making Social Q&A work for eCommerce is speed, and Amazon has done all the right things to make their model fast:

• The questions are accepted and appear immediately on the page when submitted. In the age of Facebook, this is what people expect from a social experience. Not a message that says “We’ll alert you if we decide to accept your question. It may take hours or days…”
• The question is immediately emailed to a large selection of people who actually bought the product, not just people who reviewed it. I just used the system to ask questions about 2 items, one of which had only one review, the other had no reviews. And yet within 2 hours I received 4 answers to one and 5 to the other! There’s no way to do that if you don’t email the question to past customers, or limit the recipient list to reviewers.
• The answers get sent immediately back to the asker. That provides fast reminders about the purchase the shopper was considering – while the shopper is still in the buying moment – and a smooth path back to the product detail page complete it. Plus, the answers appear immediately on the site for future shoppers to use and for the asker to review.
• Askers can easily submit follow-up questions back to the answerers, or even just send thanks. That, too, is email enabled, so that it’s easy to have rapid, back-and-forth dialog about products that one person knows about the other needs to learn about.
• To make the whole question-sending-answer-delivering cycle work as fast as this while still protecting their reputation and their customers, Amazon must be automating the moderation. “Optimistic moderation”, where content is moderated after posting, works fine for reviews, but not for Q&A where posts are emailed to real customers. And manual moderation doesn’t even come close to the speed needed, not to mention that it is too much work at any sort of scale.

The result of all these pieces working together is a system that finally realizes the promise of “social commerce”. In the lingo, it’s leveraging the “interest graph” rather than the “social graph”. Which means that the system enables total strangers to actually talk to each other about products in which they share an interest or experience. Customer reviews are great, of course. But they are not interactive. The 2-way dialog that Social Q&A enables – when it’s done right like this – delivers a level of user engagement far deeper than what reviews can provide.

Online shoppers are going to find this system incredibly useful. For example, those questions I asked this afternoon weren’t tests. I was buying a whiteboard and I needed to know whether the model I was considering erased cleanly or left a ghost image. I wasn’t going to trust the manufacturer’s claims, and the information I needed wasn’t in the reviews. Here’s one of the pages – check it out. The information I got back is far more informative for my question than what’s in the manufacturer’s description or what a customer service rep (who would not have had personal experience with the product) could have provided. And since I now know that answers come back fast, the next time I’ve got a question standing between me and a purchase, I won’t hesitate to ask.

The business significance of this utility is huge. It’s not just that shoppers are more likely to buy when they get the information they need. With a tool this powerful, Amazon has now given shoppers yet another reason to go straight to Amazon next time they are considering a purchase.

If you run an online store, you need this functionality. It provides significant conversion lift, produces a mountain of user-generated content (which search engines love), and off-loads work from your customer service team. Plus, it builds loyalty! (Here’s some data we’ve collected on all these points.) You don’t need to let Amazon run away with yet one more reason for shoppers to buy there rather than at your store.

The best way to understand how Amazon’s Social Q&A works is to go there and try it out. But, for a shortcut, here are screenshots of the main elements.

A teaser link near the top of the page:

The main Q&A area embedded in the page:

The question email:

 And the answer email:

The essence of social media is that the content comes from users.  A social graph is important for some types of social sites, like Facebook, where posts tend to be of interest only to people who have a connection to the poster.  But it’s not essential.  I can lose myself quite happily in Pinterest without following or being followed by anyone.  It’s the UGC (user-generated content) that’s the key.

So by that definition, should eCommerce sites be considered social media?  Emphatically YES.  On many eCommerce sites, most of the content is user-generated.  On this page on Backcountry.com (I don’t know if it’s representative – it’s the first one I clicked on), the word count for reviews and Q&A is 1,125, while the combined Description and Technical Specs word count is 179.  On this page on Adorama.com, the word count for Social Q&A is 6,116.  The word count for customer reviews is 1,302.  And the combined word count for Overview, Features, and Tech Specs is 478.

And yet eCommerce sites rarely think of themselves as social media sites.  Most of the larger brands and stores we work with have separate teams for “site experience” and “social media marketing”.  That makes sense.  You want to organize your teams around the 80% of things they focus on uniquely, not the 20% of things where responsibilities overlap.  But a consequence of this way of organizing is that the social aspects of the site experience often get too little attention.  The site experience team needs to focus on page design and navigation and check-out and cross-sell/recommendation and branding and loads more; social interaction is just a small part of their mandate.  The social media team, on the other hand, has become the center for expertise on how the store interacts with its customers, and how to encourage customers to interact with each other and spread the good word.  But the social media team’s domain is everywhere on the web except the store site; that belongs to the site experience team.

With the social mojo focused off the brand/store website, and the store site team spread thin, it’s not surprising that the user experience on most store sites is not very social.  But just ’cause that’s how it is doesn’t mean that’s how it should be.  In fact, by ceding the social arena to the social media sites, most stores are missing huge opportunities to create value.  A different approach, which recognizes that eCommerce IS social media and makes social a high-priority responsibility of the site experience team, can address many of the toughest challenges that online stores face.

Challenge #1: Differentiation.

If other stores also sell the same products you sell, then your product detail pages probably look a lot like theirs.  Likely, you both get the same product descriptions from the manufacturer and use the same images.  Not only does this leave you competing solely on price (yuck), it means you have little chance of generating search engine traffic organically.  Whatever margin you have left is going out the SEM window.

But social content is unique.  Build social engagement on your storefront and you can generate content no other site has, increasing the value you bring to your shoppers as well as your performance with search engines.  (Jack Kiefer, CEO of BabyAge.com, has a great discussion of this point in this recent webinar.)

Challenge #2: Customer Support.

Pre- and post- sales, customers have questions.  Sometimes these questions get posted on social media sites.  But  more often, those customers come to your site, and one way or another (email, phone, livechat) they end up in your call center.  That costs you $, and it doesn’t always make your customer happy.  While many inquiries need your staff (e.g. “where’s my order?”), many others can be handled at least as well socially.  Past customers are often more accurate, faster, and more persuasive than your own team.  Really.  Here’s some hard data.

And here’s a little illustration:  I stumbled on a customer question recently at Overstock.com about a chair I had bought from them.  This person couldn’t figure out how to make it recline.  Since I sit in it all day long, I had a pretty good idea what the problem was.  AFTER I sent in my answer, customer service posted a vague “We want to help you…” non-answer.  Then, to my gratification, the asker wrote back that my post indeed solved the issue.  (See it here.)  Social (1), Customer Support (0).

Challenge #3: Loyalty

Shoppers who engage deeply with your site are more likely to direct-navigate back to your site the next time they need to buy something, rather than just typing the thing into Google and going where ever that leads. So what opportunities for deep engagement do you provide?  Social interaction is the most powerful tool you’ve got in the engagement tool kit.  In fact, in a recent study, we found that first-time buyers who interact with Social Q&A while shopping are 15-40% more likely to make a repeat purchase within a year than first-time buyers who don’t.  (Blog post on that coming up.)

Further, social gives you an opportunity to reach out to your past customers and invite them back to your site that is completely different from the usual promotional material you send.  For example, past customers click through on shopper question emails and return to the store site to answer at a 10% rate.  And the unsubscribe rate on these emails is typically ~ 0.2%.  Most stores using the TurnTo Social Q&A system tell us that, by these measures, this question email is one of the best performing marketing emails they send.  Period.  Not to mention that the purchase conversion rate for these past customers who come back to answer is 2-4X higher than that of normal shoppers.

With benefits like these, it’s time for site experience teams to recognize that eCommerce IS social media and start prioritizing projects that socialize the on-site experience.  Leaving social to the social media team is leaving money on the table.

Guest post by: Tim Kilroy

Tim Kilroy is a consultant who focuses on helping companies achieve dramatic growth through marketing and business development efforts. He has worked with high-growth companies like Wayfair, Karmaloop, The North Face and many, many more. You can learn more at www.timkilroy.com

I spend a lot of time thinking about growth. I help companies generate more traffic, generate more leads. I spend my days thinking about scale, share of voice, acceleration. I have worked in search, in display, in mobile, brand advertising, direct response advertising. I’ve been around the block a couple of dozen times. Someone that I was working with looked at the plans that we were putting together, and he said, “OK, so what do we do with all the traffic we are going to get?”

Nobody ever asks that question.

For many companies, how to treat your visitors is an entirely separate exercise from getting visitors. There are “Acquisition Managers” and “Site Experience Engineers” and “Director of Low Intent Consumer Engagement” (that is a real title, by the way). Each of these chops up the consumer experience into silos. But customers don’t perceive your efforts that way. Acquisition and engagement are part of a continuum for consumers.

So, as my client and I started to talk about ways to engage the new visitors that we were starting to drive. He immediately jumped to reviews as a method of engagement (and I can also share that they have a measurable impact on search engine visibility). And I think when I was an SEO guy, I would have stopped the conversation right there and dialed up PowerReviews. But my experience as a CMO of a $200mm fashion retailer tells me a different story. We had product level reviews on many of our 20,000 products. The SEO boost from those reviews was 1-2%. But we were redesigning our pages and were trying for simpler. So we started doing some multi-variate testing of the new page design. Amongst other things, on the new page, there were no reviews. At the end of the testing, we saw a meaningful difference in conversion on the cleaner page. And when we did all the math, removing the reviews resulted in about a 9% increase in conversion.

I, to put it mildly, was shocked. How could this be? Reviews are supposed to add the texture and nuance that drives consumer trust and conversion. In our case, it was the exact opposite. Why?

Our brand already had a high degree of trust with our consumers. We were the authority in our space. This is where math fails a little bit, because we couldn’t do a sophisticated sentiment analysis, but it was our opinion that reviews are too subjective. Products got bad reviews because they didn’t fit, or the color was not as expected, or the shipping was too expensive. Reviews tended to be experientially-based rather than product-based regardless of how they are written. And reviews are written by two kinds of folks – those with an axe to grind or those who are cheerleaders. And to be honest, neither is unbiased.

So, I started to reflect on my other experiences in online retail over the last two decades, and at the largest online furniture retailer, we tried to improve the customer experience by adding all of the product specific Q&A from our customer service department to the product page. We had questions like “How heavy is this item?”, “Will it fit through my door?”, “What is the seat depth?”. These aren’t sexy questions. But we saw real conversion lift with this tactic. But my client doesn’t have 200 call center employees with reams of data at their fingertips. What he does have is 7 years of customers who have experience with his product. They know the facts about his products. And by shaping the information that is asked for from those consumers by making them specific questions about the product (“Will it fit in a space X by Y?”, “Does it work in Europe?”, “How long is it from shoulder to cuff?”) it removes the subjective nature of experience into objective fact distillation.

It is the answers to questions that make you want to buy. It is the product as solution that justifies the purchase (honestly, shopping online isn’t all that much fun, so if you end up on a product page, the impulse to buy is probably pretty high). UGC does make me want to buy the product. It is the information that removes the reservation. But why doesn’t that work in the review space? Frankly, the review justifies what you want to hear. If you have enough reservations, inevitably, you will gravitate towards those reviews that make it easy to walk away. If you want to make the purchase, you will, of course, focus on the positive reviews. Product reviews can be self-reinforcing events. The cheerleaders hang out with the cheerleaders, and the axe-grinders hang out with the axe-grinders. Reviews are a self-reflexive lens.

I’ll share a personal holiday shopping anecdote – I bought a Chromebook for my kids for Christmas (shhhh…don’t tell them). I read a million reviews and they were all positive…but too high-level. I don’t actually care about processor power or inane memory specs, those things that professional reviewers care about. And the product reviews definitely fell into the “GOOGLE ROCKS – The Chromebook is the end of Apple!” or the “It isn’t as good as Windows because you can’t install a faster graphics chip, who would buy this….” And, truthfully, those reviews were just noise. I had one simple question – is there something like Skype for the Chromebook? That is what I needed to know. That was my point of reservation. I travel a lot, and I love to video chat with my kids while I am away. It is meaningful to me and them. So I asked that question on a product forum (NOT on the product page, mind you, but rather a forum that I had to search for…). As it turns out, it doesn’t do Skype, but someone was nice enough to tell me that you can do a private Google Hangout, and someone else told me that Skype was actively working on a browser only plug-in, so there would no need to download a client. These were objective answers to my objective question. And that single bit of information that was relevant to me, when answered, made the purchase decision simple.

Consumers are looking for information, not opinion. Decision purchases are made on rationalized facts, not influence. Passion and emotion run high on the engagement end of things (Do I like the site? Is this a good price? Is my credit card safe? Will my friends laugh at me for buying this?) All of that happens before the decision to buy. What drives the “Add to Cart” button click is information that helps me make the decision.

And just like Detective Joe Friday, we all really just want the facts, ma’am.

Your shoppers have questions that stand between them and a purchase. Doubt this?  In an article in Internet Retailer just a couple days ago, Stephen Gillett, President of Digital at Best Buy cites 3 top reasons why customers who come to their site don’t purchase, and the first of these was “needing more product information”.  (To put in context how important this is, the other two were “product wasn’t available” and “the price was too high”!)

So how do you ensure shopper questions get answered?  One remarkably simple strategy: let them ask!  Well, maybe it’s not so simple, because if you are going to invite your shoppers to submit questions, you have to be sure someone replies.

That’s where Social Q&A comes in.  By enabling your past customers to answer questions from your current shoppers, you can provide faster, more credible answers at lower cost than by relying on your customer service team.  To many organizations, this isn’t obvious; customer service teams are viewed as more expert than customers, faster to respond, and more likely to provide the positive sentiment that will close a sale.  But if Social Q&A is done right, the reality is often just the opposite.

To see why this is so, it’s helpful to group the sort of questions your shoppers have on a spectrum from most social to least social, where “social” means suitable for answering by your customers rather than your staff.  Here’s an illustration of this sociability spectrum:

On one end of the spectrum are questions related to peer opinions, tastes, or real-world experience with the product.  These questions really need a social answer; the asker expects an answer from a peer, and staff answers are viewed as useless at best and can even be seen as inappropriate. Without a system to reliably generate social answers, this entire category of shopper question will go unanswered.  In response, many stores just provide no way for shoppers to ask this sort of question.  But that doesn’t make the questions go away, it just means they go unasked instead of unanswered.

In the middle of the spectrum are fact-based questions about the product – specs, compatibility, intended usage.  These questions can be answered by either store (or brand) staff or by past customers.  Stores that lack a system to deliver social answers will, of course, go the staff answer route for this type of question. However, we have seen that when these questions are sent to past customers, the answers are often better than what the staff provides in 2 important ways:

1. They arrive much faster.  While, in theory, store and brand staff could be standing by to answer the moment a question is submitted, in practice, most have their staff answer in batches on a schedule – typically once or twice a day.  As a result, staff response times rarely average under 2 hours, and for most stores the avg staff response time is significantly longer.  In contrast, with social Q&A, shopper questions are emailed simultaneously to a group of past customers who have bought the product.  Some of these people happen to be doing their email at just that moment.  The results: on average, across all the stores in the TurnTo network, the first social answer arrives in under 1 hour.  You can see examples in this study or this one or this one.
2. They are often far more informative and contain far more positive sentiment.  Stores may be afraid that customer answers will be inaccurate or negative, but these fears are unfounded.  For one thing, most questions receive multiple answers from past customers, so shoppers can easily see if one stands out.  But more than that, we’ve found that while staff generally provide factual answers to  just the exact question asked with neutral sentiment, fellow customers tend to answer the question asked, provide additional information that they feel will help the shopper, and wrap it all in positive sentiment.  Here’s an example of a fact question about the brim on a fedora from a great hat shop called Hats in the Belfry.  The simple answer is “yes”, but these customers have added color to their answers that goes beyond what a staff member typically would (or could without sounding “salesy”).

And when customers can be counted on to answer shopper questions, the staff doesn’t have to, which reduces load on the call center.  That applies not only for the shopper who asked the question but for all future shoppers with the same question who can review the Q&A dialog posted on the product detail page.

Finally, there’s the 3rd category on the spectrum.  These are questions that past customers can’t answer.  They typically relate to policies and terms like shipping, availability, pricing, or returns, or, infrequently, support needs for individual orders.  Some stores in the TurnTo network allow shoppers to submit these questions and route them to their staff.  Others prefer to direct shoppers to their traditional support channels (live chat, phone, or email) or to online resources like FAQ or policy pages for these questions.

Now that we’ve divided up shopper questions into these 3 types, we can answer another important question: what % of shopper questions can be handled socially?

• For stores that handle most fact-questions (the middle group) socially and deflect non-social questions (the 3rd group), our data show that over 80% of all the shopper questions asked can be handled by past customers!

• For stores that handle most fact-questions socially and accept any type of question, including the non-social ones, over 60% of all questions asked can be handled by past customers.

So, in short, this is why Social Q&A is such an effective  customer support tool:

• It enables shoppers to get credible answers to the whole category of peer-oriented questions where staff answers just won’t do
• It delivers faster, more persuasive answers to product fact questions than staff can
• And it handles most of the questions that shoppers ask automatically, which reduces call center load

We don’t usually link to our competitors’ marketing collateral; they don’t usually link to ours.  But when we saw PowerReviews’ latest case study of their “Social Answers” product running on Skechers.com, we thought it would be interesting to check out just how “Social” it really is.  We figured: if this is the site they are featuring in their newest case study, it has got to be their newest product and a best-practice implementation.

So we did a simple test.  We asked a normal shopper question on each of 8 products on Skechers.com through the PowerReviews tool and we asked the identical question about the identical product on Shoes.com, which also sells Skechers shoes and uses TurnTo for Social Q&A.  Like usual, we kept track of when answers were posted, and if the answers were emailed before they were posted we counted the earlier email time.  Half the questions were asked on both sites at around 9:00am eastern, and half were asked at around 1:00pm.

The bottom line: TurnTo generated >3X more answers total, >90% were truly “Social” (from actual past customers), most arrived in 1/3 the time, with 1/4 the staff workload vs PowerReviews.

• Just one hour after the questions were submitted, the 8 questions asked through the TurnTo system had generated 7 social answers.  There were no answers to the questions asked through the PowerReviews system.
• At 8 hours, the TurnTo system had delivered 16 social answers and 2 staff answers.  Still none through PowerReviews.
• At 24 hours, TurnTo was up to 20 social and 2 staff answers.  PowerReviews delivered 8 staff answers – one to each question – but no social answers.  In fact, even after 2 weeks, PowerReviews never produced a social answer.
• At 24 hours, 7 of the 8 questions asked through TurnTo had received at least 2 answers total, including at least one social answer.  One of the TurnTo questions was unanswered.  The 8 questions asked through PowerReviews received one staff answer each.

Skechers says they’ve seen a 30% increase in sales when Q&A is installed on their product pages.  We believe it, and we congratulate them!  Social Q&A is a powerful way to increase conversion and to create the sort of user-generated content that search engines increasingly favor.

But we also wonder: imagine how much better their results would have been if their Social Q&A system delivered 3X more answers with most answers coming from actual past customers in under 8 hours (many in under 1 hour), the way TurnTo’s does?

Appendix: here are the specific questions asked on each shoe type.

For the last few months, Shoes.com has been evaluating TurnTo Social Q&A vs. a competing Q&A tool by running them side-by-side on their product detail pages. We took the opportunity to run a comparison of the social answer rates the two systems deliver.

We asked typical shopper questions through both the TurnTo and competing systems about 6 different products; for each product, we asked the identical question through both systems. And we kept track of exactly when the answers came in. As with past tests we’ve run, we monitored both the time when we received an answer email and the time the answer appeared on the Shoes.com website, and we credited the earlier time if there was any difference. Since the TurnTo system had just been installed, Shoes.com had not yet begun providing staff answers for the TurnTo questions, but we kept track of the staff answers from the competing system anyway. You can see all the questions we asked at the bottom of this post.

Overall, the TurnTo system generated 27 social answers in the first 24 hours. The competing system generated no social answers in that period and only 4 social answers total after 4 weeks. In total, the TurnTo system generated over 7x the amount of social answers as the competing system.

The next charts show the answers received question-by-question at both the 24 hour mark and at 4 weeks.  After just 24 hours, 100% of the questions asked through the TurnTo system received at least 1 social answer.   On the other hand, 2 of the questions asked through the competing system never received any social answers, even after 4 weeks.

TurnTo Social Q&A once again generated dramatically more and faster interaction between shoppers and past customers.  If you are interested in upgrading to results like these, give us a call.

These are the questions we asked about each product:

If you know customer reviews, you know that half of the value – maybe more – is in the insights you can extract.  So you might think the same is true for Social Q&A, since these are the two main sources of user-generated content on product detail pages.  But you’d be mistaken.  For Social Q&A, engagement is the key, which means that if your Social Q&A system isn’t delivering massive customer interaction, it’s falling short.

In a recent talk I gave to a gathering of e-commerce execs from major brands and retailers, I asked the audience for a show of hands on this: if they were forced to turn off part of their customer review system, which part would they chose?  The options were:

• Turn off the back end.  Visitors to their sites and storefronts could see all the reviews, as could search engines, but all the analytics would be gone.
• Turn off the front end.  All the analytics would be available, but none of the content would be visible to shoppers or search engines.

The room split exactly in half.

At the Shop.org Summit last week in Denver, the CMO of a fashion brand told me he had just run a rigorous A/B test on their customer reviews.  He was new to the brand, and even though they’d had reviews for a while, he didn’t want to just assume it was working.  He tested the overall, site-wide effect on conversion (not just whether items with reviews did better than items without, or whether high-scoring items sold better than low scoring items).  His discovery: negative lift!  Overall, sales dropped a bit when reviews were turned on.  So I asked if he was going to turn reviews off.  He said that hadn’t been decided; the insight value they got from reviews was important enough that they would probably keep them after all.  (There’s neat recent story on how stores are using the insights from customer reviews to steer their businesses in the Wall Street Journal.) n.b. Fashion brands seem to have a stormier relationship with customer reviews than many other retail segments.  Your mileage may differ…

If you have had this sort of experience with customer reviews, you might  think that the value equation is about the same for Social Q&A.  But it’s not.  While Social Q&A can also deliver valuable insights, it is first-and-foremost an engagement tool.  You are not going to make up for poor Q&A engagement with analytics.

To put it simply: an unanswered question is a real downer, whereas no one ever knows about the review that was never written.  Unanswered questions on your product detail page scream “nobody home”.  First, there are the disappointed shoppers who asked questions and never heard back.  Then there are the shoppers who come later and see all the unanswered questions stacked up.  Sure, you can hide unanswered questions, but that makes it even less likely they get answered, and it doesn’t help the person who asked.  You can have your staff answer all the questions, but then you’re probably better off with a live chat approach, and you’re missing out on all the benefits of getting your real customers to interact with your shoppers.  In short, if your Social Q&A system doesn’t quickly and reliably get lots of customer answers to shopper questions, you’re probably better off not inviting shoppers to ask.  It’s better not to create expectations if you’re not going to be able to fulfill them.

On the other hand, if you get Social Q&A right, the massive customer engagement it generates effectively drives top-line growth.  One fashion merchant that uses TurnTo for Social Q&A sees 1100% conversion lift from those who ask questions or read dialog from others.  And it’s not an isolated effect – about 25% of their orders come from shoppers who interact with Q&A before purchasing.

Further, there are the SEO benefits;  Social Q&A done right produces 2-4 times as much user-generated content (UGC) as customer reviews, which is great for driving organic search traffic.  If your Social Q&A system is not delivering enough customer engagement to produce UGC at scale, it’s under-performing.

So the next time someone tells you that engagement isn’t important for Social Q&A – that it’s the analytics that matter, just like for customer reviews – start by asking what sort of customer engagement their Q&A system produces.

September 10th marked the start of this year’s Shop.org Annual Summit in Denver, CO.

The Summit always seems to be the ideal place for online retailers and brands to close the conference season. In addition to the myriad of social activities, the Summit brings industry leaders and innovative players together and allows attendees to gather that one last nugget or exchange that one last idea with fellow retailers before they head for holiday lockdown.

Following is a summary provided by the Shop.org blog on five sessions:

Are you making one of these 4 online marketing mistakes?

With everything that goes on in retail/e-commerce, mistakes are bound to be made.  Luckily, some of the most common ones are easy to fix.  Read about the top 4 mistakes online marketers make, and how they can be remedied here.

Cost-effective tactics to optimize SEO performance

SEO may be one of the most valuable online strategies, right after email, but are you utilizing all the available resources to optimize?  Learn more about it here.

How to be a web analytics hero

Getting the most from your website requires more than just reporting.  You must analyze!  Landing pages, search functions and product details should all be examined.  Learn more here.

Why site speed matters for big time retailers

In life, most people don’t like waiting and the same goes for online shoppers.  Read an interview with Ted Middleton, Vice President of EdgeCast Network, for insights on why your page load times matter and how to keep yours fast here.

H&M: Lessons learned from David Beckham’s Super Bowl ad

I think the title is self-explanatory on this one!  Read about the tactics and strategy H&M used when planning to launch their Super Bowl ad surrounding David Beckham’s underwear line here.

For those of you not familiar with this event or have not attended in the past, this year’s Summit had one of the biggest turnouts recorded and with the Shop.org promise that ‘attendees will acquire valuable strategies and tactics to improve online and multichannel retail business’, you may want to check it out next year. This Summit is confirmed for September 30-October, 2013 in Chicago, IL. We hope to see you there!

David Streitfeld of the New York Times has been looking hard at the issue of fake customer reviews.  A year ago, he called out freelancers offering to write positive reviews for a few bucks.  In January, he wrote up a service called VIP Deals that offered rebates in return for positive reviews.  And a couple days ago he published a piece on Todd Rutherford, the founder of gettingbookreviews.com, who sold 4,531 book reviews in 2010 and 2011 at $20-99 each, before backlash from Google and Amazon forced him to shut down the service.

No doubt, there is fakery out there.  The question is: how widespread is it?  4,531 seems like a lot of reviews, but it’s a small part of the billions of customer reviews available on the web.  The most recent NY Times piece says Bing Liu, a computer scientist at the University of Illinois, Chicago specializing in automated text analysis, “estimates that about one-third of all consumer reviews on the Internet are fake.”

Really?

One of the reasons there is so much suspicion of reviews is that so many of them are positive.  Liu has estimated that 60% of the reviews on Amazon are 5 stars and another 20% are 4 stars, but, he says, “almost no one wants to write five-star reviews, so many of them have to be created.”

Here’s an alternative explanation: there’s a lot more customer satisfaction out there than you might guess.  That’s a conclusion you might come to from reading customer answers to shopper questions gathered through TurnTo.  With the TurnTo system, there is almost no possibility for fakery.  When a shopper asks a question, the system chooses a group of people who actually bought the item (based on the transaction records of the store) and emails the question to them.  While the system allows in-line answers on the product detail page, >90% of the answers come in reply to this question email.  So unless there’s a big population of people buying products they don’t need for the purpose of providing artificially positive answers to shopper questions they may never even receive, these answers are legit.

And one of the most striking aspects of the answers provided by these real product owners is how effusively positive they often are.  For example, here are some customer answers to a shopper question about the height of the drip spout on an espresso machine at SeattleCoffeeGear.com.  The question doesn’t ask for any sort of overall evaluation of the machine – it’s just looking for a measurement.  Yet many of the respondents (including me) spontaneously volunteered our enthusiasm for the product.  (You can find this page here.)

Now I don’t want to be Pollyanna about the problem of fake reviews.  I suspect they are much more common on destination review sites like Yelp and Trip Advisor where anyone can submit than on ecommerce sites where the ability to verify purchase is an easy and effective way to police.  It’s also harder to believe the uniformly high ratings sometimes found on products which are judged on subjective personal taste, like books and food; personal tastes differ too much.  As Streifeld points out, even The Great Gatsby (which, first published in 1925, is presumably not attracting many fake reviews) has plenty of neutral and negative reviews (>300 reviews are 1, 2, or 3 stars out of 1,400 total at Amazon).

But while the battle goes on between the fakers and those trying to root them out, it’s possible that in many cases when the reviews are positive, customers might just be happy.

Retailers and brands at this year’s Shop.org Merchandising Workshop, which took place in San Diego a couple of weeks ago, left with pages of notes on how to improve conversion, site design, customer engagement and much more.

If you were unable to attend, here are a few of our favorite sessions from the two days:
(These overviews are compliments of Shop.org’s blog posts)

Doug Mack, CEO of One Kings Lane shared his take on ‘5 current trends within the e-commerce space.’  Read the overview of his session.

Sarah Veit Wallis, VP of Global Ecommerce for Bare Escentuals, and Michael Burgess, President of Saks Direct, shared a few of their lessons learned with launching mobile sites and apps.  Read more.

Kate Spade’s Johanna Murphy, VP of e-commerce, and Kristina DiMatteo, Digital Marketing Manager, showed how content and commerce are the same to this New York based brand. Read about their recent campaign here.

In addition, while I was unable to attend Tim Ash’s presentation on ‘Are you committing one of the 5 deadly sins of landing page design?’ I did hear that his ‘sins’ unfortunately resonated with some of the attendees.  Read about it here.

For those not familiar with this event, Shop.org does a great job – not only is it always in a beautiful setting, but they bring together interesting speakers with relevant topics in an environment where one can hear and learn what fellow retailers are doing, learning and experiencing. My calendar is already marked for next year’s Merchandising Workshop, which will be held July 15-17 in Huntington Beach, CA.  Hope to see you there, if not before!

There was an interesting  piece in the New York Times over the weekend about the experience Urban Outfitters has had with personalization.  In brief: they tried altering the merchandise displayed on the site based on the gender of the visitor – it seemed like an obvious win to show women more women’s clothes and men more men’s clothes.  But it backfired.  Dmitri Siegel, then at Urban Outfitters and now at Patagonia, noted that shoppers “took offense at being subjected to gender-based marketing.”

And yet some aspects of personalization – in particular, recommendations – seem to perform well, and even better when aided by automation.  Last week I was speaking with an IR50 retailer who said that in a recent test their automated personalization engine outperformed their human personalizers by 3X!  (Recommendations proposed by the automated engine received 3X as many clicks as recommendations manually composed by the merchandising team…)

This contrast – between personalization that works and personalization that doesn’t – reminded me of the windshield wipers on our car.  OK, OK, bear with me here.  Up until the car we own now, the intermittent wiper always worked the obvious way: you turn it up, the delay between wipes gets shorter.  But on our current car, it’s different; the wiper setting determines the amount of rain that a sensor needs to detect before triggering a wipe.  That is, you are not setting the wipe frequency directly, you are setting it relative to the amount of rain that’s falling.  The idea is this: if you like a dry windshield, turn up the knob, and the system will keep your windshield dry for you by increasing the frequency of wipes as the rain gets heavier.  And if you prefer to minimize the distraction caused by wiping, turn down the knob, and the wiper will only trigger when it senses a very wet windshield.  In fact, you never even need to turn off the intermittent wiper, because on a sunny day it never triggers, even on the highest setting.

Like personalization, I suspect that reactions to this feature vary, and some folks probably like it.  But I’m not one of them.  I prefer to have direct control over the frequency of my wiper.  While the sensor is pretty good, it’s not perfect, and when I want a different frequency than what the sensor has chosen for me, it’s hard to get it right.  Further, changing the speed manually is not hard, so the intelligence built in to the system isn’t really saving me from some difficult burden.

You can see where this is going.  In an effort to “personalize”, the wiper system forces me to give up some control.  But the cost to me of the loss of control is greater than the value to me of the personalization.

I suspect that my experience with the wipers is similar to the experience of shoppers at Urban Outfitters.  By personalizing based on gender, the site removed a degree of control from the shopper.  After all, women shop for men’s clothes and (ahem) men shop for women’s.  So there’s a cost to the shopper when the personalization is off.  On the other hand, there’s very little burden for the shopper in controlling the system manually – ie clicking the tab that says Women’s when shopping for women’s clothes.

This doesn’t mean personalization is always a mistake.  There are some times when direct control is too hard and having a system help is very valuable.  One example: in modern fighter jets, direct control of the airfoils through wires and hydraulics – the way pilots used to control planes – is no longer possible.  So instead, the pilot tells a computer what she wants the plane to do, and the computer tells the control surfaces how to move to accomplish that.  Similarly, on an ecommerce site with many SKUs, smart, personalized recommendations can help me discover items that I would miss with less-smart recommendations.

The trick in getting the user experience right is to figure out when allowing direct user control is preferable and when automated personalization is the better choice.

I spent last week at IRCE 2012 in Chicago, and the #1 question I got asked was “what do you think of the acquisition of PowerReviews by Bazaarvoice?”  Now that I have a quiet moment, I thought I’d set down an answer.

For the merging companies: this is a smart move for Bazaarvoice, and while it’s the end of the road for the PowerReviews team and their products as we know them, it’s the best possible outcome for the PowerReviews shareholders.  The $152 million purchase amount, about 13 times PowerReviews 2011 revenues of $11.5m, is a steep price to pay for a unprofitable business that’s a distant second in their segment.  So why did Bazaarvoice pay up?  Because even though Bazaarvoice won most of the larger accounts in the market, PowerReviews was in every deal competing on price.  The downward pressure this put on Bazaarvoice pricing didn’t just kick in on new deals; every time a Bazaarvoice contract came up for renewal, the customer always had the credible threat of switching to PowerReviews to keep a lid on Bazaarvoice prices.  What’s the value to Bazaarvoice of eliminating price competition from PowerReviews?  Here’s a sample calculation.  Bazaarvoice is currently at a $120m revenue run rate.  If they are able to increase their price realization by 20%, that’s another $24m in revenue, with no associated cost – ie it’s all profit.  The present value of an incremental $24m a year in profit, at a nominal 10% discount rate, is $240m, making $155m a tolerable price to pay.  If the PowerReviews business turns out to be worth something, after their team has been shrunk and their prices raised, then that’s all gravy to Bazaarvoice.  So that’s what you call a win-win, right?

Well, not if you are a customer.  If you are already a Bazaarvoice customer, the effects will be straightforward enough: when your contract comes up for renewal, expect to pay what you’re quoted.  Your alternatives just got a lot narrower.

If you are a PowerReviews customer, it’s a bit more complicated.  Yes, you should expect to pay more when you renew.  But you should also be thinking about what this means to the support your product is going to be getting over time.  Here’s my bet on what’s going to happen.  Bazaarvoice will keep the PowerReviews Express product.  It won’t get a lot of new investment, but it wasn’t before either, since PowerReviews has been focused on their enterprise offering the last few years.  It costs little to sell and service this offering, so there’s no need for Bazaarvoice to cut back to maintain the business.  On the other hand, Bazaarvoice is likely to cut back heavily on investment in the PowerReviews Enterprise product.  This is the offering that’s driving the PowerReviews sales, marketing, and R&D costs, and it competes directly with the Bazaarvoice product line.  It’s hard to see Bazaarvoice carrying the R&D cost of two overlapping products or training their soon-to-be-integrated sales team to sell both.  On the other hand, they won’t want to alienate their newly acquired customers by forcing a rapid change-over.  So I expect they’ll keep the PowerReviews Enterprise product around with minimal investment, and they’ll convert the stores using it over to the Bazaarvoice platform little by little as the PowerReviews platform falls further and further behind.  For those who use the core PowerReviews reviews product, no immediate action will be needed, since that’s already a mature, full-featured platform.  But for those using or considering the new tools in PowerReviews’ recently-announced “Essential Social Suite” – analytics, loyalty, social sharing, Q&A – the picture is different.  All of these new tools are raw, and some are explicitly still in Beta.  Without the significant R&D investment needed to bring these products quickly from their current state to even a modest level of maturity, adopting them may cause real pain.

Yesterday George Eberstadt, our CEO and Founder, hosted our webinar “Beyond Customer Reviews – Meet Social Q&A.” George covered topics ranging from “what is Social Q&A?” through to the positive effects Social Q&A can have on a website’s SEO. Below are the slides from the webinar:

The new Nielsen trust report is out.  (Write up at Social Commerce Today.)  Duck – here come the inflated claims from social commerce companies about how this proves the effectiveness of what they do.  The problem is, it’s not that simple.  First, here’s the data:

The temptation is to conclude that because “recommendations from people I know” are most-trusted, a sure-fire strategy is to get your customers to recommend your stuff to their friends.  The problem is that the whole reason friends are trusted is because they WON’T recommend your stuff to their friends – not unless they sincerely and spontaneously feel it’s in the interest of their friends to do so.  And as soon as you throw in an incentive for your customers to tell their friends about you, you are going against the whole trust equation – no one wants to feel like a shill, and no one wants to be shilled to.

Also important: social trust affects the purchase cycle in different ways at different stages.  “Consumer opinions posted online” are effective at the evaluation stage, but not at the awareness stage.  No one reads customer reviews unless they already have purchase intent.  On the other hand, recommendations from friends tend to matter more at the awareness stage than at the evaluation stage, and this varies a lot by product category.  For example, I just got a new Dyson hand vac because I saw it at my friends’ house this weekend and they told me it was awesome, but I would never have called them to ask what brand of vac I should get at the moment I was about to buy one.   Of the last 40 items I bought at Amazon.com, 6 of them had some sort of social influence at the discovery stage, but only 1 had any social influence at the evaluation stage.  That was a ping pong table, and I didn’t seek out the advice about which to get; I just mentioned to a friend that I was thinking of getting one, and she said to get a Stiga because it’s a Swedish brand and she’s Swedish.  Similarly, for vacation plans, I sometimes call friends for advice during the evaluation phase, but since it can be hard to know which friends have relevant experience, my process tends to be haphazard.  Fashion is a tricky category – people ask friends for advice, but assessment of a friend’s expertise is often based on their taste rather than their direct product experience.  (You don’t have to own a particular brand to answer “does this look good on me?”)  How do you, as a brand/store, affect that conversation?!?

So, the next time someone makes a blanket statement to you about how you need to be doing social commerce because people “trust” their friends and fellow shoppers so much, say “whoa there!”  Trust is not the same as influence, and leveraging the trust that’s inherent in social relationships to promote your business is not that simple.