A Geek’s Guide To Gaming The Algorithms
Ian Lurie Sep 29 2014
I talk about all that’s wrong in marketing, tactics vs. strategy, etc. etc. We’re ignoring strategy and sound marketing. We work too hard to game the algorithm at the expense of long-term business development. FYI: For me, gaming the algorithm means finding the fine edge between legitimate and spam, and skating that edge.
For me, gaming the algorithm means finding the fine edge between legitimate and spam, and skating that edge.
The high-minded ideas don’t spring forth from my buttocks, fully-formed. I don’t have some moral objection to testing the limits of Google, Facebook and others. This is about smart application of resources.
I’ve developed my own decision-making algorithm to answer “Should I try to game the algorithms?”:
- Research tricky tactic
- If brain effort to use tricky tactic >= effort to do something legitimate, do the legitimate thing
- If unsure, default to legitimate thing
That’s worked pretty well so far.
Have a look inside my geek brain. I feed these kinds of questions through my personal algo every day. Here are the answers I invariably get:
I just did a search for ‘beat Google’ and got 160,000 results. They’re not popular. You need to decide: From 9-5, are you a revolutionary, or a marketer?
Algorithm: Google Panda
Question: How much do I have to change this paragraph/sentence/page to make Google think it’s unique?
Answer: If you’re asking, you’re already in trouble. Google’s getting better and better at examining content structure and meaning. At this point, your only guarantee is to completely re-order and rewrite your content. It’s easier to write something new. And, in an infinitely-large universe, there are plenty of related articles you can write on a topic. So do that, instead.
Effort level to rewrite content just enough: 7/10
Effort level to just write new stuff: 5/10
Algorithm: Google Penguin
Question: All these other sites buy links. Why can’t I?
Answer: Google dedicates massive amounts of computing power to finding artificial link schemes. They miss a lot of them. For a while. But they’re using NSA-dwarfing data processing oomph to hunt ‘em down, and nothing is forever. Consider yourself a pebble under a steamroller. The odds are very, very much against you.
Effort level to acquire links and not get caught, forever: 10/10
Effort level to not get banned by Google: 0/10
Algorithm: Google Hummingbird
Question: How many times should I repeat the keyword?
Answer: Google’s getting better and better at tracking keyword co-occurrence and semantic matching. Use it to your advantage, and write decent copy.
Effort level to shoehorn keywords into place: 9/10
Effort level to write content that doesn’t make my eyes bleed: 7/10
What’s it all mean? Search engines are applying mind-boggling resources to close algorithmic loopholes. The border between legit and spam keeps moving to catch more spam. Faking it has a very short lifespan. Don’t do it.
A search for “Facebook rank” got me 155 million results. All of those people are spending time pursuing magic formulas. That gives you a lot of room to do it right, and kick their collective rears.
Algorithm: Facebook EdgeRank
(actually, not EdgeRank any more, but I have no other name)
Question: How many likes do I need?
Answer: It depends. On Facebook, there are now 100k or more factors that determine what appears and what doesn’t. Who the hell knows what those factors are. Relationships, velocity, time decay, interests, ad clicks, average freckle count…? I have no idea. If you want to game the algorithm, aim for lots of likes by real users with broad influence. But that sounds an awful lot like marketing to me.
Effort level to acquire likes that look legitimate: 7/10
Effort level to acquire legitimate likes: 7/10
Question: Can’t I just buy followers and views?
Answer: Of course. Fiverr is a great market for such things. But YouTube sure doesn’t like it. On Facebook, it’ll actually hurt you, since their algorithm includes user behavior. Users who loll about doing nothing show up as uninterested. You want 40,000 uninterested users feeding into your post rankings? Be my guest.
Effort level to acquire buy enough YouTube views to rank and never get busted: 13/10
Effort level to acquire legitimate views: 2/10
Algorithm: Readers’ brains
Question: Can’t I just pay people to re-tweet my articles, or somehow force them to like the article before they read? Won’t that help me look legit?
Answer: Yes, but is it sustainable? People get tired of ‘like gates’ pretty quickly. And getting a bajillion re-tweets on one article won’t help the next if the re-tweeters are sitting in a room somewhere, using 200 fake accounts.
Effort level to fake popularity, forever: 10/10
Effort level to build popularity, forever: 10/10
What’s it all mean? The algorithms behind social media are so complex you’re better off trying to decipher the human brain. Which, of course, is what social media companies are really trying to do. If you’ve figured it out, stop doing marketing, build a bunker, then build an army of intelligent machines and take over the world. Anything else is a waste of time.
Apply your nerd-fu wisely
Instead of gaming algorithms, apply scientific-ness to legitimate methods:
- Smart, targeted content promotion
- Maximizing crawl visibility and performance
- Testing the crap out of everything
- Keeping a log of everything
- Getting content performance down to a science
We all have a choice: Use our brains to game the system for short-term gain, or use our brains to measure and improve serious, long-term marketing strategies.
You want to game the system? Go for it. I will not judge. But there are better ways to use your time. Think about it.
Ian Lurie is CEO and founder of Portent Inc. He's recorded training for Lynda.com, writes regularly for the Portent Blog and has been published on AllThingsD, Forbes.com and TechCrunch. Ian speaks at conferences around the world, including SearchLove, MozCon, SIC and ad:Tech. Follow him on Twitter at portentint. He also just published a book about strategy for services businesses: One Trick Ponies Get Shot, available on Kindle. Read More