Back in April I wrote about the two things you can do to improve your site search. Those are two things among many options you have available to you as you seek to keep visitors on your website and help them achieve the task at hand. Of course, one thing you can consider is a search engine replacement. Better technology has an allure. However, it shouldn’t be the place you start.
This morning I had to log on to United Airlines’ website to request a refund for accommodations from a recent overnight flight delay. Surprisingly there is no form specifically for this type of request on the site. I struggled with a bit of cognitive dissonance on how to fit my request in the standard fields where one might complain about rude service or a poorly maintained restroom.
Needless to say, I didn’t come away from the experience with a favorable opinion of United or its process. This at a time when they should be trying to take a bad situation (my original overnight delay) and turn it into something awesome. It didn’t help that there were errors in their login process as well as an inexplicable refusal to load a 900KB JPG file that was both less than 1MB size limit and one of the approved file types.
Am I less likely to return to united.com because of this bad experience? No, I’ll be back. Fortunately for United, oligopolists can get away with poor service. Can you?
Can you imagine not caring about how you rank in Google? Maybe someday something will replace Google as what is arguably “the” most important source of top-of-funnel traffic but today much of your marketing activity is rightly focused on SEO and SEM. Having a strong Google game is critical to many companies success. But is Google your friend? I’d argue that Google as Frenemy is the proper way to look at the relationship; Google is critically important at some points but dangerous to rely upon later. Let me explain.
Total Cost of Value
Last week I was working on a proposal for a client. We understand the value we need to deliver and we’re going to bring in some external tech to deliver a complete solution. There are several options for the client to chose. What’s interesting about this part of the process is that my role has gone from seller to buyer. Any good seller looks at the deal from buyer’s perspective. But when you actually become the buyer, your vision is narrowed even further. You focus not only on what it’s going to cost but what is it going to give. It strikes me that when looking at total cost, what’s really important is understanding the total cost of value. What does it take not just to get and operate the tech, but to get the total value that can be extracted?
One of the things our clients have been asking for is automation tools that help make search better. This is a great idea. For example, there’s no reason that search success metrics can’t be embedded in algorithms that A/B test improvements and automatically kick in if the testing demonstrates improvements. But automation alone isn’t the answer. Humans need to be in the loop a critical parts of the process to ensure that the automation is achieving the right goals. The greatest myth of search automation is that is eliminates the need for people to be involved.
Search is one of the most critical customer experiences on your website. Search is also the customer experience that likely gets little attention in your management system and at budget time. Sometimes this is because site search improvement seems like a dark art. However, if you’re just getting started with search improvement (and many of you are, you just may not know it) there are two things you can do now to make your customer experience better.
I’m just back from a week in southern Georgia (the state, not the country) fishing with a few buddies. For the past few years, I’ve had a side hustle writing for fly fishing magazines about fly fishing in places where fly anglers don’t normally lurk — most fly anglers go after trout — and fish for species that are not normally pursued. This past week we fished the Okefenokee National Wildlife Refuge and pursued Bowfin (called Mudfish locally) and we caught a bunch. The weather was perfect.
It was a great week. While on vacation it always takes a while to shut off my work brain so I do have two lessons from the road.
When the VRBO listing says “Only a short walk to the train viewing platform” we mused about the hobby that is Trainspotting. Apparently, trainspotters come to Folkston and stay at the charming little house we were renting. We should have looked into this further. We didn’t.
When we arrived in town, we crossed the two tracks that run the length of downtown Folkston. Our house was about a hundred yards from the tracks. Taped to the refrigerator was a sample list of the trains that come through Folkston daily. The list contained 40+ trains. We then Googled “Folkston Trains” and learned that Folkston is not just a train viewing locale. It is one of the best train viewing locales in the US. Almost all the trains that go in and out of Florida must go through Folkston. They even refer to this area as the Folkston Funnel.
We didn’t count the trains. There were a lot of them. Running 24 hours a day. It seems like a particularly busy time of day is between 4 a.m. and 6 a.m. and then again in the evening. And then again in the early morning. Really, all day long, trains, trains, trains.
If we had gone deeper upon discovering that there was a train viewing platform we might have chosen a different location for our lodgings and would have slept better. We didn’t.
Fun Fact: At every railroad crossing a train must blow its horn. The pattern is: two longs blasts — one short blast — one long blast.
Fun Fact: Regulations say that a train horn must be 110 decibels. That’s rock concert loud.
Fun Fact: At 160 decibels your eardrums burst. The threshold for being driven insane by train horns is much lower.
Be where your customers are
This is a town that has little industry outside logging and only a handful of trainspotters making up the tourist base so there aren’t a lot of dining choices for tourists. When you look on Yelp, you find a handful of fast food restaurants mentioned and a few local eateries. We worked our way through these and found that Jalen’s BBQ, a dodgy looking place, served excellent barbecue and were mystified why the top rated restaurant in town was, in fact, the top rated restaurant in town. It wasn’t very good.
Tuesday evening when we stopped by the grocery store to replenish our supplies, we noticed the Brickhouse Restaurant a few doors down. It wasn’t on Yelp (it is now, I added it). It’s a breakfast/lunch place so we stopped there the next morning. This restaurant turns out to be the best one in town. Great staff. Excellent cooking.
While the Brickhouse is well known to locals, you would pass it by in this electronic age when increasingly travelers are using apps to find dining options in unfamiliar places. Is your business listed where your customers are looking? If not, get going. You need to be where your customers are.
This morning I dig out of a stuffed inbox. The good news is that I slept most of the day yesterday recovering from the sleep deficit I gathered in Folkston. I’m energized to see what the week brings and marvel at the quiet.
If you close your eyes and imagine what a search engine results page looks like you probably see something that looks a lot like Google’s search results. You see a list of titles and text snippets that potentially describe the thing that you’re looking for. That’s what Google and Amazon and practically every other site has trained us to see. What few people see is a grid of tiles. This is probably a good thing as there are few use cases that tiled search results are effective.
Where do tiled search results work?
We don’t see search engine results presented as tiles often but we do see them from time to time. When you see them on a commerce site, tiled results can be quite effective. These results usually include a picture of the thing, a title, pricing, a call to action (usually a cart action), and perhaps a snapshot review or description. The picture is the key to knowing if you’ve found the right thing in a commerce setting and a tile is an effective way to deliver this content. But we also see this on sites that are definitely not B2C related.
So what’s the downside?
A large investment bank has made search it’s primary navigation method. When you load their homepage, smack dab in the middle of the page is a search box. They have a hamburger if you want to try navigation but the user experience clearly has doubled down on the effectiveness of search. It’s a bold move. It says “Our search is that good. Go ahead, we dare you to try and not find what you’re looking for.” So how do they do?
First let’s review the two things that all search results pages have to do really well:
- Present high quality search results that answer my question.
- Give me some indication of which of the results I should pick as the right answer.
Traditionally, those goals have been achieved in list form. Each entry in the list contains an informative title and a snippet that gives me more information about the content found on each result’s landing page. So why are search results always presented this way? Well there are a couple of reasons.
- It’s what we’ve been trained to expect after two decades of seeing search results. We don’t have to figure out how to use the search results, it’s our cognitive model.
- It is an excellent user experience for communicating this information. The title & snippet model contains most of the information you need to evaluate the response.
- See 1 above. It’s what we do.
Are tiled results good or bad?
I don’t know. In theory, tiles should be fine if, and that’s a big IF, your search results are awesome and the top three results are the right answer to every question associated with the keyword. The tile also has to tell me which of the three results I see above the fold are the right one. I haven’t seen the data to know whether the tiles are better or not and I suspect many of the tile adopters don’t really know that as well.
The investment bank example above, in my opinion, scores poorly on all accounts. First, I didn’t find many of the results above the fold to have a title that was descriptive enough for me to understand it. Second, there is rarely any additional information presented to help me gain additional insight and what little additional insight there is is behind another click. I might as well click on the result instead of wasting time on the “additional info” click . Third, I know this industry and the titles seem to be more marketing speak than information about the content. It feels like the exact opposite of what a search result should do.
My gut tells me, outside of commerce use cases, tiles search results are less effective than lists because I haven’t seen anyone crack the code on providing enough information on a tile to allow me to know which one I should click on.
So why change?
The list of search engine results is a highly effective method that has few challengers. So why do companies experiment with a change? There are a few good reasons. Most notably, some of our tribe are creative animals and they’re constantly seeking something more innovative, something better. Of course, a new design can’t just be an artistic design exercise, it has to be a usability exercise. Does this change make the experience better and deliver better results for the business?
There is no doubt that the emergence of mobile has changed the way we consume content and voice search will change it even more dramatically. When you think about voice search you can’t scroll through a list of things so perhaps getting our house in order where we can deliver results in terse, well constructed bit makes sense for both tiles and, eventually, voice results. But I think that’s a bit of a stretch. I think most companies do it because they think it looks cool.
I still want tiles.
Fine. But make sure it’s not just something you want, make sure it’s effective for your customers and prospects. If you’re going to give tiled results a try, I’d recommend the following:
- Make sure you work the design hard. Think about how tiled results fit into the rest of the site’s design and deliver the right information to make the search effective.
- Test the heck out of the tiled design. A/B test tiled against list. A/B test descriptive tiles against lean tiles.
- Measure success and whether search success for each of these design changes increases or decreases goal achievement.
The goal of search is to connect, as quickly as possible, your customers and prospects with the answers to their questions. Don’t lose site of that and you’ll be fine.
Autocomplete is the bane of any message sent from a mobile device. However, one place where autocomplete shines is on search engines. Autocomplete, also called autosuggest or incremental search, improves the user experience by making it easier to execute searches by suggesting words and phrases that a matching algorithm determines are appropriate based upon the characters entered into a search box. Google has had autocomplete in their search box since 2008. Following in their footsteps, any company that values its customer experience has implemented autosuggestion in their site search. But there’s a new trend in the type-ahead game that we’ve begun to see cropping up on more and more company websites: Instant Search.
A/B Testing Site Search
A few weeks ago I participated in a webinar over on Biznology. I shared five strategies for improving site search using A/B testing. Long a part of the digital marketer’s toolkit, A/B testing has relevance for search analysts as well. Using these methods you can test new settings against a portion of your live traffic without risking tanking things altogether. Once the new settings have proven themselves, you can deploy them to the primary search engine.
Most companies don’t measure site search well. They rely upon out of the box measurements that measure activity instead of outcomes. Nowhere is this problem acuter than when you move from one search engine to another.
- How do you exceed customer expectations if you don’t have an effective baseline?
- How do you know you’ve done the migration well if you don’t measure the gap before and after?
- How do you ensure improvements in content and algorithms are effective?
A/B testing of the search engines is the answer. Check out the webinar. You can also read a short version I did in a blog post a few weeks ago.
A full transcript of the conversation is below: