The Secret of Viral Marketing

The secret of Viral Marketing (VM) is that it’s all about popularity. Which is to say … sooner or later, it’s all about sex.

When we made Perplex City, the Alternate Reality Game, I learnt a lot of basics about VM just from going out there and trying so much new stuff on a week-to-week basis. It was a small startup, so (at least at first) everyone got involved in everything, and you got to see the things that crashed and burned in as much detail as the ones that shone like gold (this was before the company grew enough that people started covering-up the failures).

Viral Marketing is so much abused both as a phrase (used meaninglessly) and in production (used to spam people’s addressbooks) that it’s become a dirty expression these days.

But it’s an essential part of the New Wave of strategies for designing, making, and selling computer games profitably. Anyone hoping to be successful at challenging the industry’s Stone Age reliance on “big budget, bigger teams, binary hit-or-miss, massive risk” approach needs to understand and adopt VM.

The Science of Marketing

Marketing is a science, plain and simple. It’s more scientific than almost any other part of modern business. Even accounting is less science-based than Marketing (did you know that most Accounting isn’t based on laws and mathematics but on hand-waving, “do whatever you want so long as you can invent a jusification for it”, and arbitrary principles that get changed frequently by the governments?).

If you haven’t already, get yourself a copy of Sergio Zyman’s “The End of Marketing as We Know It”. It’s aimed at busy marketing people; it’s a short, easy to read, book. There’s no excuse for anyone not to finish it. He explains neatly the “marketing IS a science, it’s just that lots of people don’t realise this” concept. The “end” he refers to is the end of people treating it as an art-form – because anyone approaching it scientifically is already wiping the floor with those who don’t.

(Sergio was the Marketing Director for Coca Cola. Right or wrong, it’s an interesting read)

Myths of Viral Marketing

• Usability / Low barriers to entry of the product are essential
• It has to be easy to spread virally
• There has to be benefit to each user the more people they share the product with

None of that is essential. So, arguably, none of it is important. It’s very similar to making a hit AAA video game, because if you screw up everything else, it doesn’t matter, so long as:

you provide something that the individuals perceive to be of a value that will make them look good to their friends

If you like, think of it as the “critical decision point” within the Viral Loop concept. That single value-judgement, in the mind of the consumer, is what drives your Loop into a virtuous circle, or into a vicious one. It provides a base percentage chance of the product spreading; everything else at best modifies that percentage up or down, but cannot overwrite it completely.

Perception is EVERYTHING

Let me repeat that statement, with some hilights:

you provide something that the individuals perceive to be of a value that will make them look good to their friends

In case that’s *still* not obvious enough (shrug; there’s so much misinformation and prejudice around VM that I’ve found its often hard to get through to people, past their cynicism, skepticism, and revulsion…):

1. it DOES NOT MATTER what *you* think; it only matters what goes on in the CONSUMER’S mind
2. …it doesn’t matter what ANYONE thinks; this revolves around “what WILL my friends think of X” – and that is OFTEN misaligned with “what the world thinks about X”, “what *I* think about X”, or even “what I would like my friends to think about X”
3. Even that is just a stepping-stone to, or a vector for: “what will my friends THINK OF ME”
4. VM is not about your product
5. VM is not about your brand
6. VM is not about “a need, satisfied” (although it can be, coincidentally)
7. VM is about a personal conversation between two people you’ve never met (and never will) in which your product just happens to be a convenient bargaining token to help one of them get more sex (or popularity; which can ultimately be converted back into: more sex)

Great; but how does this theory help *in practice*?

Need to Know: Viral Loop

If you are not intimately familiar with the VL, and understand how to use it, and how to improve it, etc … go read Andrew Chen’s introduction to Viral Loops.

Some things you need to internalize for yourself about VLs:

1. You determine the VL through *analysis*, not wishful thinking: you make your product FIRST, you make the GUI FIRST, you make the website FIRST … and then you sit down and look at it and ask “So … what’s the VL here? And how could it be better?”
2. The VL is a list of percentages, and spread-factors: each one is the “probability that someone will progress to the next point” and “the number of people it will spread to, IF it spreads”.
3. Maths can predict the future, with VLs: you multiply all the bits together, and run a recursive mathematical analysis … and Maths can accurately predict the “maximum number of people this will spread to”, and the minimum, etc etc. This is awesomely powerful. It’s also quite tricky Maths, but a good spreadsheet can do surprisingly large chunks of it for you. (hint: it’s the same basic problem as working out “with the current rate of interest, how much money will I have in X years?”; solutions for this kind of thing are built-in to most spreadsheet software)
4. …especially when you monitor the “actual, observed, percentages” and put them back into your analysis. This is not guesswork: it’s a science. And the science of recursive growth has been studied for centuries, and it’s extremely well known and understood by now.
5. The VL tells you “how much things will grow, given the virality”; they tell you nothing about how things will grow in other ways (e.g. marketing campaigns, good PR, etc) – you still have to do ALL the traditional marketing analysis and planning, and merge it in to your VL as the “total number of current consumers” (which will change each month as your traditional marketing brings in big chunks of new people on its own)

Need to Know: Exploitation is a gimmick

VM got a bad name very early on when it was used to spread utter crap “non-products” to huge numbers of people. It got a reputation for being vapid and meaningless and a cynical abuse of the consumer.

Well … wake up and smell the roses. The same thing happens *every* time someone discovers a new vector for marketing. It happened recently with the iPhone platform, with a wealth of mediocre rubbish making its authors millions of dollars of revenue. The goldrush is always bright and alluring, but brief and ultimately empty, and when it goes, it implodes in an instant.

If you think VM will “save” your crappy product, forget it. You’re about 10 years too late. Make a decent product, and try again.

Myths … revisited

OK, I lied: those myths aren’t “irrelevant”. At least … they *are* irrelevant up until the moment you get your core VL working, and growing successfully.

But … once you hit that point, they suddenly become very important, as they provide low cost ways of making big positive changes in the overall success of your VL.

Just don’t forget that they are not the VM itself: e.g. don’t get so obsessed with “ease of use” that you suck out the part of the product that was making the consumer look good to their friends.

(I’ve seen that happen. More than once)

Putting VM into Practice

The best starting point I can see for a VM campaign is to look at the conversations. This is nothing new; Marketing in general is often described as being all about the relationship of the consumer with the brand – and specificially about the conversations the consumer has with other people … the ones that mention the brand.

To shape your VM strategy, think about the conversations people might have about your product.

Then throw them away, and remember the core point in this post; think of the conversations people might have that would make them seem cool, clever, funny, interesting, and attractive to their friends. And work out ways that *the idea of* your product could help those conversations.

Remember: the idea of the product may be more important *to the conversation* than the utility of the product itself. Not to the actual sales that you make – they still need *the product* to be the most important. So, be wary … this is where a lot of the abuse starts – the realisation that you can spread a product without it actually even doing anything. But what’s the point of spreading a useless product? Everyone may know your name, but if they hate you, and won’t buy it, you’re no better off.

Once you have your conversations, look at how – physically – someone could do virtual product placement of your product within the conversation.

For instance, if I’ve got an iPhone app I’m telling my friends about, and they think it sounds interesting, I need to pull out my phone and *instantly* show them the cool stuff. Not fiddle around with menus and crap – my friends getting bored and just nodding with fixed smiles to humour me.

But then again … if its not an app I would show you that flippantly, or if it’s something that only one friend would want to see but would bore the rest of them, and I’d want to carry on talkin to everyone else about something else … maybe I’d rather hand you the phone and let you see for yourself how it works, and I want there to be lots of menus, and patient explanations, so that I don’t have to stop talking to re-engage with you.

And think carefully about the sorts of things that make anyone more popular, more often:

• Impressing a group of people, all at once (huge social value right there), rather than just an individual
• Making people laugh
• Telling people about themselves
• Making one member of a group look good / clever / attractive to everyone else
• Being generous; giving people something of value, for free
• Improving people’s lives in small but memorable ways (making things easier, the little things more than the big ones)

And things not to do:

• Bore people
• Exclude people
• Devalue people (spamming addressbooks? BAD IDEA. Funny example recently: Ed Castranova got de-friended for viralling his contacts with a mail that said they weren’t important enough for him to read their emails … unless they paid him, per email, in Serios. FAIL!)

And … well, I’m sure you get the idea. Really, this is all about helping your consumers to be the most popular that they can be. You can say that’s a cynical, and “bad” thing, but at the end of the day … what’s so bad about making people happy?

PS

This post is all Andre’s fault. We were talking about Social Games, and some experiments I’ve been doing recently with iPhone development. Particularly about how I’m about to quietly launch something fun but very simple, with some basic viral hooks, and see what happens. We got to talking about how and why it would / would not work, and what ideas I had for tweaking it in EITHER outcome.

If you have an iPhone, and want to be on the initial invite list, email me. Even if I know you’ve got one (its hard to remember who has and hasn’t; and you might get an invite anyway if I misremember you as having one. Oops).