Searchlight Digital

Twitter Hacked Again…

Pete — Tue, 14 Apr 2009 23:50:34 +0000

Well this is getting old. Twitter seems to be undergoing another spate of account hacks, this time using dating and NSFW style messages.

Seriously guys, I love your service as much as anyone, but get your acts together. You really need to get this sorted and fast.

kthnxbai

Why Google’s Ranking Factors Have Changed Over Time

Pete — Thu, 09 Apr 2009 09:41:16 +0000

Thought I’d leap on Rand’s recent post on SEOmoz about Google’s ranking factors, and how they’ve changed over time. However, I’m going to look less at the what (which I agree with), but the why.

The problem at its core is that whilst UGC is all well and good, if I want to read a review of say, a hotel, I’d much sooner trust one review from someone from The Times or BBC than a review from TripAdviser. As a general rule, people don’t have the depth of experience to give me a truly authoritative answer to the questions I have.

As such, Google is now presented with working out what will be the digital equivalent to our newspapers. What should be the sites that get massive trust and are seen as authorities in their areas.

In the past, I think Google did this more on a page by page basis, but now with the greater prominence being given to domain-wide authority, I think it’s going to bias further and further towards fewer sites.

This then leads to an obvious question, which is how do you continue to complete in the search space in five years time, assuming that domain trust (or something like it) continues to increase in prominence?

My belief is that if you want to rank for something, you need to become the online authority in whatever it is you do. Otherwise all you’ll rank for is your business name (because your site is the authority on your business, and nothing else).
This in turn creates an obvious change in how we currently build websites – most sites we work on nowadays have a blog of some form, which is where we can place great content with the aim of attracting links and traffic, and building the client’s profile online. However, if that is the end goal, then the blog part really should be what the site is about.

As such, I’d propose that any site you’re building, or going to build in the future, should focus on the content and authority building elements first, and the company should play a secondary role. I’d actually run two sites – one corporate one, which is more like a brochure for your company, and a second one, with a small section for your company, and the rest dedicated to being the online authority in whatever it is that you do.

For instance, if you’re a company that sells beds online, build your site with the intention of making it the authority online in beds, so that if I want to know about how to sleep better, or cure insomnia, or anything else bed-related, your site is the one I end up on. Then, both the search engines and your visitors will trust you more, and that will lead to more sales.

The Coming Revolution in Internet Service Revenues

Pete — Tue, 07 Apr 2009 13:41:57 +0000

Much has been written about the problems services like Facebook, Twitter, YouTube, Digg and most other large, productless websites have in breaking even, let alone actually making some money. Essentially, the problem comes down to two things:

Companies believing that advertising is a business model, and
The “success” of the ad-supported model thusfar

There are two staggering and obvious flaws in these ideas. Firstly, focus. People are not, repeat not on your site to view adverts. They’re there to use your service. If that’s MySpace, it’s to listen to tracks from their favourite artists, without paying for them. If it’s YouTube, it’s to listen to music or watch videos. If it’s Twitter, it’s to watch conversations and interact with others.

Note, in none of these places does “view or click on an advert” appear. Ads work on Google’s search product because a link is a link, and as long as the end result is still a good match for what the user searches for, they really couldn’t care less.

As a result, as a marketer you need a lot of eyeballs to make ad impressions work, even if your ads are highly targeted to the demographic viewing them. Therefore cpm costs aren’t that high, and never will be. In fact, they’re likely to get lower and lower over time. This is bad for publishers.

The second massive flaw is that if, for instance, the bottom fell out of the ad market, all of a sudden the ad supported business model dies. So here’s a radical idea for you, that you may not have thought of, if you’re either starting, or already running a service-based site:

Have a product.

Seriously. If what you’re doing is worth my time, then charge me for it. If it’s useful, and I’ve no other choice, I’ll pay. Why do companies still buy MS Office rather than using Google Docs? Because it’s worth it. Why do people spend hundreds of thousands on a house rather than living in a tent? Because it’s worth paying for.

Offer me something that satisfies a need, and ask me to pay for it. You’d be amazed at what’ll happen…

Python and the is.gd API

Teifion — Thu, 19 Mar 2009 11:19:23 +0000

Owing to the fact that zi.ma is down I have been using is.gd for my URL shortening needs. I have thus created a python script to send URLs to is.gd and then print out the new shorter URL. It’s better coded than my zi.ma tool and can be imported and used like that or run itself with no modification needed.

#!/usr/bin/env python
"""
Created by Teifion Jordan (@teifion)
 
Feel free to redistribute this script though I'd appriciate credit being given. I take no liability for anything that may go wrong with this script.
"""
 
import urllib2, sys
 
def add_url(url):
	"""
	Send it the URL you want shortened and it'll send back the new URL string
	"""
 
	# At this point I either can't code or it's all worked, lets just hope it's the latter
	return urllib2.urlopen('http://is.gd/api.php?longurl=%s' % url).read()
 
 
if __name__ == '__main__':	
	# Get URL
	if len(sys.argv) > 1:
		url = sys.argv[1]
	else:
		url = raw_input('Enter the URL you wish to use: ')
 
	if url == '':
		print "Script cancelled"
		exit()
 
	print add_url(url)

How Microsoft Could Win the Search War

Pete — Mon, 16 Mar 2009 11:32:06 +0000

First up, a little bit of a disclaimer: I don’t like Google being a monopoly. I don’t mind Microsoft’s monopoly on the OS market, as it means you only have to write a program for one OS. However, I don’t like one company holding sway over the public’s access to information. That’s more troubling. That aside, I recognise that you’re always going to have one company that’s a monopoly in these kinds of arenas. Catch 22.

Now, moving on, for a few years I’ve been arguing that the next big search engine won’t be one that’s more relevant. For all our hyperbole and high-mindedness, actually we don’t really care about relevancy. We use Google because it’s a habit. With that in mind, what will replace Google won’t be something with better results, it’ll be something that changes the game. On that front, people have been suggesting that maybe it’ll be Twitter, or some form of social search. And whilst that’s certainly an option, I don’t think it’s quite right. What I think it’ll be is a different form of UI for search.

As a for instance, I don’t use Flickr anymore to find images. Nor do I use Google, Yahoo or anything else. I use Cooliris. Why? Because the interface is SO much better for the task at hand. In a similar way, I think the next evolution of search will happen by moving search outside of the SERP.

In a similar way, I think the next big search engine will look something like this:

(click for larger view)

Now, will this be the exact look? Of course not. But what you’ve got here is the ability to search without actually going anywhere. You can search and get your results without leaving the page you’re on. Each result would then open in a new tab.

Google didn’t change the game because it’s results were better. That was just something that helped. Google changed the game because it took search from being an add-on to a portal, to being the raison detre of the site. Yahoo et al were mammoth sites that happened to have search engines. Google was a search engine and nothing else. When you went to the homepage, that was all that was there.

In the same way, something like this would allow for less disruption to the searcher, both mentally and practically. As such, I believe this will be the next evolution of search.

What do you think?