They *Finally* Developed the “Mind-Reading Module”: Baynote Collective Intelligence Platform

*DISCLAIMER* If I sound overly-exuberant about Baynote, don’t be surprised. As my excitement for this technology has grown, it has spilled over to the rest of Avalon management, and we have already signed an alliance with Baynote. So I suppose I’m a bit biased. . .

A friend and I used to joke that what our end users really wanted wasn’t a better way to use their keyboard and mouse, but simply a “mind-reading module” so they wouldn’t need to provide any hints, and yet our software would know exactly what they wanted. Since then, every once in a while, I’ve run across rare user experiences that felt like the software read my mind, and knew exactly what I was looking for with little or no information from me. Baynote falls into this category and, based on what I have seen, is likely to end up being not-so-rare on account of the degree to which it can enhance a well-designed Search implementation.

Take, for example, a Google search for “viking stove”:

Above the search results, the second sponsored link points to www.US-Appliance.com/Viking. It’s a helpful link, no big surprise yet–Google Adwords not only finds applicable ads, but allows them to customize the title and URL to my search terms.

Next, I click that link and see a Viking product page. Here there is some surprise, since that page interestingly shows no refrigerators, dishwashers, etc. as I would see if visiting their normal Viking products page. Instead, the only Viking products shown are ranges and cooktops, with a side-bar showing “Shopper’s Picks” — and in that sidebar, again, all ranges and cooktops:

That’s getting closer to reading my mind–either through Adwords or by using the HTTP “Referer” us-appliance.com was able to see the search terms I entered into Google, and recognize that I’m looking specifically for stoves.

But now comes a couple of real mind-benders: I hit CTRL-F and search for “stove” and no product on the page contains the word stove, which makes me wonder “How did this page know what I was really looking for?” Let’s take it one step farther and search on us-appliance.com for “viking stove”. At first glance the results make sense, there are plenty of cooktops and ranges listed:

But at closer look, again the word “stove” is nowhere to be found in any of the search results.
Being a search professional, I know this magic feat could easily be accomplished using synonyms, but according to Baynote, the technology provider that showed me this example from their customer, no synonyms were configured or used for this example. In fact, before they started using Baynote’s Social Search technology, when us-appliance.com customers searched for “stove” the search results would show no cooktops or ranges, just a couple “stove-top” tea kettles, since it turns out the appliance vendors don’t call them stoves, they call them cooktops or ranges, and no synonyms were configured. In fact Us-Appliance administrators probably didn’t even know their users were using the wrong search terms.

But customers searching for “stove” don’t want stove-top tea kettles, whether those search results contain the word “stove” or not. And Baynote’s software watches what users do as they use the site, detects the patterns, and determines what it is that users really want when they search for “stove”. Then, Baynote can offer the “Related Products” widgets and enhance search results based on the wisdom obtained from tracking user behaviors.

Is that what you thought your search engine already did? I can tell you that many of our clients bought a leading search engine because they thought it would track user behavior and improve search relevancy as it “learned”. I’ve had to carry the message that the search engines generally just “learn” from new content, not from common user behaviors.

I’ve also had to carry the message that search relevancy in Enterprise Search is a tough issue, for many reasons you’ve probably heard before. While we’ve helped our customers get better control over relevancy and search user satisfaction, I’ve longed to be able to tell customers “Yes, there is a simple, reliable, low effort way to have search relevancy ‘learn’ from your users’ patterns,” yet that’s a message that had to wait until now. While it’s not actually a “mind-reading module”, as my title suggests, I must admit that Baynote comes pretty close. What’s better is Baynote is nicely packaged to make it easy for me to use it to enhance the search relevancy for any of our clients.

For those of you waiting to hear the rest of my search security thoughts, don’t worry, I’ll write more about that next, but for today I couldn’t pass up the chance to share my excitement about Baynote. If this gets you even remotely as excited as it does me, contact me and I would be happy to show you a live demonstration and assess with you whether Baynote can benefit your user base.

Avalon Consulting, LLC About Avalon Consulting, LLC

Avalon Consulting, LLC implements Big Data, Web Presence, Content Publishing, and Enterprise Search solutions. We are the trusted partner to over one hundred clients, primarily Global 2000 companies, public agencies, and institutions of higher learning.

Headquartered in Plano, Texas, Avalon also maintains offices in Austin, Texas, Chicago, Illinois, and Washington, DC.

Leave a Comment

*