Hi again folks! This presentation was a very exciting one moderated by Yahoo!’s Jon Myers and featuring Dave Coplin from the Bing/Microsoft team, Paddy Moogan of my alma mater agency Distilled and Aaron Khalow of online marketing connect.
The panel was really well rounded and we got perspective from a search engine (Bing) that has, to its credit really been pushing things forward in terms of social and real time. We also heard from Paddy who had done a lot of tests around optimising for Real Time and had some great ideas. Finally we heard from Aaron who comes from a more traditional marketing background and asked some tough questions that all big companies will be asking: is it really worth it?
Dave gave a presentation that really looked out where we are heading with “real time”, the power and aims of Bing as the “decision engine” and why we are so far away from where we want to be with real time and trusted social signals.
What is search and what could it mean to us?
First important thing is to understand search and to treat it as a revenue stream. Search is a way to make money for a great deal of people, however, Dave focused primarily on the consumer and what they are looking for (something we marketers need to be in touch with in order to produce solutions and make our money). From a consumer’s perspective, search is a window into the internet. It is the way in which a modern society accesses all that the internet has to offer: rich multimedia, queries, concerns, social, whatever.
Where are we on the search journey?
We are still very much in the early stages of where we could be:
1 in 4 searches is successful (25% success rate as an industry). How many airlines do you know that operate with a 25% success rate?
At the moment people really need to refine and re-search before finding what they are looking for.
50% of time searching is spent refining queries.
30% go of these searches go on for more than 30 mins.
And worse still, far too often this still ends up taking us somewhere we don’t really want to go.
Real Time Decisions Versus Real Time Search
The internet began as a text based system meant to mimic a library in terms of accessing and calling information.
We have clearly evolved from this – how much of the internet is now text based? The original plan for a library-like system is no longer relevant.
We may now want to blend in the social signal, we may want geo-spatial information to find where we are going (particularly important real time), applications that bring search to us without sending us to a search page through mobile apps.
We need to consider all this and to be successful we need to access all types of media to help us make decisions more easily.
Intent is soooo 2007
We need to move beyond intent. Navigational, informational, transactional are the traditional realms and a general path that many people take to reaching their end goal.
However, we don’t work through each intent getting to the final intent (transactional). We want a “super intent” that brings all of these things together and that’s the big step forward.
How can this happen?
To get there we require more information about the searchers and privacy is a real issue and this will be our battleground for the next 10 years? How do Bing and others help us make better decisions about what to do with privacy? In David’s eyes, Mark Zuckerberg has made this completely transparent but there are too many choices and most people don’t know which of the 500 things they care about.
If the outcome is to serve better results we can think about this a different way: we need to talk about trust and if Bing can show us results and help us relate it to something that is what will help us understand. As a result of this sort of thinking Bing has been proactive about blending signals that we trust (i.e. social signals) into the landscape.
Recommendations are an element of trust and are really powerful when we are talking about results.
Friday Night Test
Bing tests most everything this way to decide whether they are accomplishing what they aim to.
Example: on Friday night I’d like to go see a film (true grit), I want a Mexican meal and I want to be home by 10 in bed. At the moment you can’t find all of these things at once – you find yourself going point to point to point with a number of searches to find film times, how to get there, where the restaurants are, how to get home etc. And how will you know if the food is any good, the cinema of any quality, etc.
Uses the train apps as a perfect example of these issues (it may be real time, but it is ultimately fairly useless). David points out “I don’t want to go there just to find a tube. I want it to tell me how to do the above three things and I want credibility.” That is the answer, that is the future and where we want to be.
How might this Work?
Looking for a film but I don’t know if this is a nice theatre. My friend Mike is a big film buff and I trust his recommendations.
We Want Decisions, Not Results
When you say “I want it now” do you really want the next time the Victoria line or do I want to know how to do everything that I just said I want to do in an evening and trust the results.
Paddy spoke to us about Real Time Search Optimisation and where is it going from an SEO standppoint?
How has it changed?
Relevance is becoming more important. When it was first launched Google were just showing off and they put it on pretty much every keyword out there. Now they are getting better at displaying it when needed.
Google are looking at CTR for these massive SERPs rather than just showing off: is anyone clicking on these results? Is the user looking for the latest results?
Example: ac milan vs tottenham – probably want real time results
“rand fishkin” – do you really want real time results or just a blog post he’s written, etc.
Paddy implied that they are trying to figure out what exactly people are looking for and what sorts of results people are clicking on when real time results are shown.
When Would Google Show Real Time?
*Look at social mentions (Facebook, Twitter, Friendfeed) – lots going on?
*Indexing of new content via trusted sources (speculative, Paddy has not been able to test this one yet but this is what he thinks is going on).
Paddy also pointed out that you don’t always need to be a trusted source. This can be gamed by just firing off lots of content around a certain event (Twitter bots, etc) though he offered the usual disclaimers that people should not abuse this.
Google suggest getting a lot smarter (Rooney overhead kick got picked up as a suggest item really quickly) and there is some indication that Google testing these sorts of things can help them with these sorts of queries going forward.
Image courtesy of Paddy Moogan
Where is it now?
Being an influencer is becoming more important. Google aren’t very good at this yet but they are certainly trying to get a bit better about finding who to index. Paddy points out that live/real-time results from Twitter cannot be taken directly or it would be too open to abuse. People do abuse this but if they took it straight from Twitter it would be even worse.
How to Improve CTR
Paddy suggests that you include an image in your tweet where possible (10-20 seconds when going on you’ve got to get in there) because when it shows up as an image in real time people are more likely to click through to your site. Paddy and I discussed the fact that this may not get you a TON of traffic after the presentation but it will certainly make the most of your time in the spotlight on Real Time
Problems with This
Problem 1: It can take Google a while to show the image 4-10 minutes
What is the solution? Tweet an image prior to the main launch to get it showing, then when you launch properly your image will show instantly.
Problem 2: How to get value?
If the user clicks on an image that just takes them to yfrog.com (not going to go to your site).
If you tweet the exact image source they just see a jpg, no call to action, unlikely to stay on your site.
So what’s the solution here? 301 redirect once Google has indexed the image directly from your site.
You’ve got to get the right order though to make this work – let Google index the image first, put the 301 on the image file. If you do it backwards the image won’t get taken and the page title instead – nowhere near as likely to click through on real time.
Quick Get More Traffic Tips:
Get into Google News – news search box given preference for breaking news. Distilled got this for “SEO” after one of their stories went big and saw a big spike in traffic. So, how do you get included?
*Need multiple authors (or you can just pretend)
*Use Numbers in the URL (Add to the end)
*News Sitemap (less important now seemingly)
*Keep asking. Eventually they will likely let you in.
Blog on breaking news – hard to do without someone watching the breaking news. But if you have resource to do this or get lucky looking at the news at the right time you may as well jump on it.
To do this effectively you either need to be a bit lucky (and stumble upon something new before others talk about it) or be ready to react very quickly to breaking news signals in social media. For there it’s simples!
*Write a story
*Get people to retweet for you.
Last up on this panel we heard from Aaron Kahlow who spoke to us in a slightly different way. Aaron was looking at making sure the ends justify the means and pointed out both we and the Search Engines still have a lot to learn.
Aaron’s first and most important question was: is real time worth learning? IS this a priority or not? Ultimately, he was talking about the fact that he and Rand Fishkin were discussing this at a conference and although there was some debate the end result was that no one knew if it was worth it and could prove to what extent. The search engines still don’t know what they’re doing really and it’s not REALLY real-time, as they cannot crawl and index things right away.
We are very early days in this.
What should we do?
Figure out if it is a priority. Aaron’s view is that this is really more of a “nice to have” because there are so many other important things that will impact site performance ongoing and even the best sites still always could use improvement in other areas that might have a bigger impact.
His view was basically that, unless you’re in the news industry, this is just a blip in time. Is it going to be worth it? Probably not for a lot of sites!
How does this effect the user? If your site is found in organic AND backed up in real time you are really likely to convert. However, this is a rather big “if” and probably not going to occur regularly. His take seemed to be that, if you’re not a news site and you’re not selling ads on a CPM basis this isn’t likely to generate any significant traffic.
Final Tips from Aaron
1. Watch and analyse first- monitor what’s going on before you invest budget!
2. Look at the content you have. Start syndicating that stuff. The more it is found, the more comments and likes you get the more likely you are to get recognition and sell your products. In my view this one depends a bit on what your product offering is, how you deal with duplicate content etc., but for the right client this would definitely be a good idea.
3. Reach out to your community- what do they want from you?! If they don’t want information from Twitter then leave it be.
That’s it for that panel – I’ll be covering one more session today on “How Come Search Gets all the Credit?!”