A fair point, although I'd say it's highly dependent on the kind of product you're building, too, and who your audience is.
For instance, if you're selling enterprise products rather than consumer products, and you're selling them to a less communicative industry (like the construction industry, for example), you may find that there's a lot less "word of mouth" going on, so the fact that it's not just spreading completely by itself shouldn't necessarily discourage you...
Another point - you might need to reach a critical mass of users in order to be able to effectively improve the product using A/B testing and other techniques that rely on larger numbers... if so, it may be worth investing in marketing the product to build those numbers and then be able to make those improvements measurable and rapid.
I agree with you that some models and markets have retention and referral as a much bigger priority than others.
Did you cite the construction industry as something that was a best guess, or do you have particular insight into the insularity of companies in that field?
As far as needing "larger numbers" for split testing, you might be surprised at how few samples you need to achieve an acceptable confidence level for most effect sizes. It is less than most people think (of course I can't give numbers because I don't know the effect size or confidence level that you are looking for.)
My company deals with the construction industry, and it is not a very tech-enabled industry (despite the fact that they use many very advanced piece of software to design buildings).
We're still figuring that out - we're at the beginning of our sales/marketing process, really. The application has spread virally between users through the invitation process, but we're finding that we get good leads from, for example, turning up to conferences and telling people about it in person.
Basically, traditional sales/marketing... find the potential customers and talk to them directly.
Yes. As you mention in your other comment, trade shows and magazines are generally well attended and read (respectively).
So far the most effective we've found in terms of scale has been a kind of twisted version of word of mouth... involuntary word of mouth, sort of. Since our app is a collaboration tool, people invite their collaborators, both within their own companies and within the other companies they work with. And so the application spreads, fairly consistently.
Sounds like a perfect viral approach to me. So you find that once one of these called-in collaborators uses the product, they then instigate use on a separate independent project? That is, a viral/chain-reaction, not just one user inviting their immediate collaborators and it stops there?
EDIT You're probably already on top of this, but (just thinking) if the collaborator-->instigator step is the problem, you could offer a discount for that specific conversion; the idea being not to make it more affordable, but to buy the bargain-seeking mindspace of "I wonder if there's a way I could grab this bargain?" (totally opposite of the article's "improve the app" approach).
It's just that wom is the best marketing. It's free. It's credible. If you can make it work, do it.
"Crossing the Chasm" claims wom works better in smaller, specialized niche markets, because everyone in it has similar needs. Though it takes your point that they do need to actually be talking to each other for wom to work. There are trade-shows and trade-magazones for every industry you can imagine (not saying you should necessarily use them, but it indicates communicativeness), and I believe (without proof) that when people of related industries bump into each other, they'll talk shop, sharing their news and common concerns. hey, what do you think of this new thing? Isn't it terrible what's happening with that new issue?
In practice, I've seen wom for my developer software product on mailing lists, websites and blogs. Someone even targetted it as an AdWords keyword (cool). My product doesn't target a tightly communicative market - it's not specific to an industry, but cuts across them; and it's not a whole solution to a problem, but a tool that's part of a solution. So I think its wom is fairly diluted. And yet... there is wom.
EDIT hmmm, perhaps I'm really targeting the "IT" industry (even though its members work in other industries.) And, I guess, it is by far the most "tech-enabled". I still like "Crossing the Chasms" idea of crafting your product in a way that fits in with those shop-talk conversation: the shared concern +cool solution, in a short, memorable, interesting soundbite (like a joke that goes viral.)
I see now: you're interpreting the article as saying wom is a litmus test of how good your product is: if wom isn't working, your product is rubbish, and you're better improve it. You're saying that the ease of wom varies with industry, so that as a metric, it's not a universal indicator of the quality of your product. I think you mean "discouraging", in that one thinks one's product isn't good enough. But maybe you mean "discouraging" in that one thinks one's marketing isn't working.
I interpret it a strategy: make your product so extraordinary that wom comes into play. Wom is so great, that it's worth the effort, even in an industry where wom is harder.
But I see your point: maybe the wom-bar is so high, that you're better off just marketing in traditional ways. I still think wom is great, and worth trying for, but I acknowledge it's a tradeoff, not a simple "if no wom, do nothing but improve your product".
Yes and yes. It's not to be taken as a general bible, but contains some insightful truths drawn from hugely successful startups. Very well written (guy has a PhD in literature.)
See also "The Innovator's Dilemma" for an insightful academic treatment (compellingly researched, but also less practical) of the same idea of disruption, though mainly for large corporations.
For most social apps/websites, critical mass is important. Nobody is going to recommend your site to their friends until you get at least a few thousand users. "Social proof" and all that.
I don't think is argument is predicated as much on spreading through word of mouth as it is on the efficacy of marketing something people don't care about.
So what does that say about people who put a lot of effort into promoting? Does it mean their customers aren't telling their friends? Are the customers just not internet users, or is the product bad?
What does that say about companies (newspapers, Google) whose income stream is from people pouring money into promoting their products? Does it mean they'll be continually in search of new advertisers as the old ones go broke, unless they can fool their eyeballs into confusing ads with honest recommendations?
I have been thinking this from a theoretical point of view for years. But it doesn't seem to be happening. GM is still in business, the New York Times isn't going anywhere, AdSense payments-per-click aren't dropping, and Anheuser-Busch still seems to be spending plenty on ads — right?
I think you're absolutely right. This idea sticks like the emperor's new clothes. Nobody wants to point out it's not true for fear of looking bad. Because supposedly the people who believe it's true will be producing exciting, awesome products.
I'm still trying to work out what amount of organic growth one should see before doing advertising. It certainly isn't every user getting ten other users.
I think there are two pieces to it - "making things exciting for the user" and "making it easier to share the excitement". A "tell your friend" link here, "post to facebook" there - this sort of thing.
I don't know, I posted this to generate a discussion.
Thinking logicaly there has to be something to it - if a given person is exceited to level X he will call his friends and tell them about it. If he excited to level of 0.75 X he may send them an email. What if he is excited to 0.1 X? How can we make a good use of that person's modest excitement?
If you writing a twitter app this is relatively easy - nudge the user to twit to all their friends about that particular app. The barrier is very low in this case.
If you are writing a facebok app you can also use pre-canned social graph.
But what if e.g. you are writing an iPhone app? What can you do to convince people on a given level of excitement to talk about your app with others?
This is quite useful, but I challenge your assertion that nothing can be done for an app that is not inherently viral.
I am not looking for a miracle here. Since I already have a stream of prospects I don't really need a VC>1, I'll be happy if each normal sale ended up bringing in one extra viral sale - this would double my revenue.
They don't entice me to share. OTOH, when I have the impulse to share something, I'm much more likely to do it if it is easy to do, especially via apps I already use regularly.
Yes. YouTube only had 150 videos uploaded after their first 1.5 years of existence. But once they added the click to share buttons and allowed users to embed their content they went crazy viral.
On Domino's pizza tracker for online ordering, you can now "share it on facebook" with a button. I'm surprised at how popular that thing is.. and it definitely social proofs domino's and leads to viral behavior patterning
For instance, if you're selling enterprise products rather than consumer products, and you're selling them to a less communicative industry (like the construction industry, for example), you may find that there's a lot less "word of mouth" going on, so the fact that it's not just spreading completely by itself shouldn't necessarily discourage you...
Another point - you might need to reach a critical mass of users in order to be able to effectively improve the product using A/B testing and other techniques that rely on larger numbers... if so, it may be worth investing in marketing the product to build those numbers and then be able to make those improvements measurable and rapid.