FYI: I'm also here to help answer questions as the co-author of Traction. In my other life I'm the
CEO & Founder of DuckDuckGo.
Also, for everyone stopping by, you can get the first three chapters for free at http://tractionbook.com/
@caseya in the book we present a framework called Bullseye to figure out what traction channel (out of 19) to focus on to move the needle for your traction goal (as I painted a bit more in this comment: http://www.producthunt.com/books...). When you reach diminishing returns in that effort, you have to restart the process, either because that channel isn't going to get you to that goal, or your goal was attained and now you have a new one, and the channel won't get you to that goal.
At DuckDuckGo, we've gone through that process 6 times, including going through it again right now! In fact, we've even run it more than that because we have a separate traction goal for DuckDuckHack (our open source instant answer platform) where the goal is developer engagement (as opposed to consumer user acquisition).
The channels we've ended up focusing on for various steps from 10,000 searches a day to 10,000,000: SEO (search engine optimization), content marketing (my blog, and then micro-sites), social ads (mainly reddit), PR (online and then TV), and Business Development (integrations into Safari & Firefox). All of these have reached diminishing returns and we're essentially back to the traction drawing board, running Bullseye again internally.
@eliotpeper in doing the second edition of Traction we really focused on these mistakes were seeing, especially for people who read the first version, and we saw four broad non-intuitive mistakes:
1) Not starting traction efforts early enough. In the book we offer the 50% rule where you should spend 50% of your time on Traction. Even if people buy into this, however, they generally aren't starting early enough, namely right when they start product development. This is non-intuitive because people feel they're wasting time and money on a leaky bucket product. But the reality is that a steady stream of cold customers really help you figure out where the holes are, what niche to focus on, messaging to use, etc., information you don't get fully from product development and that if you don't get from this early traction effort, usually leads to further rounds of product development after launch.
2) Not considering all channels. People are always asking us what is good for B2B or B2C or SaaS or what have you. All channels have been useful for all types, and in fact, the most useful channels are often the most under-utilized ones in a given scenario, simply because they are less saturated, and less saturated efforts usually lead to better conversions.
3) Not scoping tests well. In the testing phase of our Bullseye framework you are testing 3 of the 19 channels, trying to find one to focus on. People mess this up by trying to do too much with these tests, and moving into more what we'd call optimization. You want to do the minimal amount to validate your assumptions and move on.
4) Not doubling down on one channel. When a test works, our framework calls for solely focusing on it. People mess this up by not dropping the other channels that had moderate outcomes, and this makes intuitive sense because they had some outcomes. However, those outcomes are less promising than the main channel, where this extra effort could be spent to make it even better.
Report
@yegg@eliotpeper Thank you for a thoughtful question and answer!
@bbalfour Good question. I actually don't find many good books around traction/growth/marketing, which is why blogs (like yours) are so useful to stay current. And why we wanted to write this book.
Scientific Advertising by Claude Hopkins was excellent, as was Breakthrough Advertising by Eugene Schwartz. I reference those books quite a bit. Thinking Fast and Slow by Kahneman. For whatever reason, I had a ton of marketing thoughts and ideas when reading 33 Strategies of War by Robert Greene.
Otherwise, I don't think most growth/marketing books have led to fundamental shifts in my thinking about marketing. I think a lot of marketing and growth is following a few principles, understanding a few things about human psychology and rigorously applying them. So that said, I've found psychology, strategy and history books more helpful in terms of rounding out my thinking about marketing.
@bbalfour the one piece that comes to mind that **fundamentally** changed the way I think about it was The Psychology of Human Misjudgement, a speech Charlie Munger (who works with Warren Buffet) gave at Harvard in 1995. The full text is at http://www.rbcpa.com/Mungerspeec... and there is audio at
He basically walks through biases, now made more popular by behavioral economics in books like Predictably Irrational and others, but also ties it to marketing. He makes the point that the best marketing efforts generally work by relating emotionally to more than one of these triggers like curiosity, reciprocity, etc.
@rrhoover I ran marketing at Airbrake, which was a developer tool for software developers. I was always surprised at how well a $5 t-shirt worked as an incentive for our customers who made $100k+ per year.
Though to be fair, they were awesome t-shirts :)
@rrhoover we just started sending traction tactics like these to our list (sign up at http://tractionbook.com) and here was mine:
"On my phone I set up a funny caller tune — the Count from Sesame Street counting my rings — that delights almost everyone that calls me. I get so many comments it is kind of ridiculous. The Count does an amazing job of putting people in a good mood right before I talk to them, such as press and potential partners. I made a recording of what it sounds like here: https://soundcloud.com/yegg13/th.... I think part of what makes it work so well as there is no message or ring before the Count starts counting; this is done through T-mobile caller tunes, and I'm not sure if there is an equivalent in the other carriers."
Here are a few others:
1) As I mentioned in this other comment (http://www.producthunt.com/books...) I was one of the first advertisers on the reddit ad platform. They soft-launched asking for beta tests on Hacker News (then still Startup News I think) and I was lucky and watching then. It had such good conversions early on.
2) In my last company, which grew mainly through viral marketing, you try to eek out less-than-one-percent improvements, because they compound as the viral loops keep turning. As a result, you A/B test *everything*, and you get kind of creative with it. I eventually hit upon an email template with HTML pastel boxes across the top of the email that consistently worked better than any other template. At the time images in email was still problematic so this essentially and the effect of getting color without the problems of images, and it worked really well.
3) Early on at DuckDuckGo (2009) I built a Karma widget that you could embed on your site and it would display your "karma" across social media, showcasing your number of followers, etc. At the bottom it would have a link to "new search engine" and this tactic immediately had us rank first on Google for new search engine, bringing in a steady stream of users. That said, and as I recount in the new preface to the book, this was an overall mistake because SEO wasn't going to drive enough users to really move the needle. Nevertheless, I was proud of the execution :)
Hi @jwmares. Thanks for doing the AMA! Few related questions for you:
1. How do you decide features to prioritize for a product - ones that would help you bank users, or features that existing customers want to have?
2. Which gets you more traction in different phases of the business?
@kunalslab Great questions.
1. This is a tough one. A lot of it is product-dependent, but one way I think about it is, if a bug or lack of a feature is hurting your conversion rate and adoption in your market more than a new feature that'd bank users, focus on building the feature. For example, at Airbrake (company I ran growth at previously), we had 1 feature that every other competitor had. And, when I asked customers what they wanted, they all wanted this feature. So we built that first. Later, when we heard some customers wanted feature Y but it wasn't a sticking point for new customers interested in the product, we focused on growth-oriented features.
Does that make sense?
2. Could you clarify? I think there's a real balance here: adding growth-oriented features can get you a ton of traction, but you likely only want to focus on this once you have a core product that people are responding positively to.
@jwmares Makes a lot of sense. Weigh the growth you could get from a new feature vs. the retention you could if you gave exiting customers a feature they want. For question 2, I was asking what changes your decision making in different stages of the business (0-100 users, 100+, 1000+, etc.)?
Amazing book absolutely loved it! Such a great read for a podcaster and has really helped me build The Twenty Minute VC! So Justin, if you had to choose one interview which was your favourite to do, which one was it and why? Can't wait to hear the answer to this one!
@harrystebbings Thanks Harry! How did you use it to grow the Twenty Minute VC?
Favorite interview... that's a hard one. However, I will say I was blown away by Alex Pachikov's interview about how they leveraged partnerships and new channels to grow. They've done a ton of really, really smart stuff behind the scenes to grow.
@mtufekyapan in the book we enumerate and give primers on all 19 traction channels, and also provide a 3-step framework (called Bullseye) to figure out which channel to focus on. It's called Bullseye because you're aiming for the center of a target, the center representing the channel that will move you toward your chosen traction goal. Roughly here is the basic approach:
Step 0) Get a traction goal. This goal needs to be quantitative and meaningful to your business, often at the beginning around raising funding or getting to profitability, though could be something else. How many customers do you really need to move the needle? Knowing this informs the rest of the framework.
Step 1) Brainstorming, or "The Outer Ring: What's Possible". The first real step is to go through all 19 channels and really effectively brainstorm how they could be used for your project. The idea here is to really try to come up with tests of you could use each and every channel to answer how many customers the channel could reasonably bring, at what cost, and if they are the right type of customers? Where this gets messed up is not really focusing on all channels, since we are biased towards some based on our experience, but often the under-utilized channels, those ones we know least about, are the best.
Step 2) Testing, or "The Middle Ring: What's Probable". After the first step we encourage looking through the 19 tests you outlined and picking the three that seem the most promising and then literally going out and running those tests. Good middle ring tests should take no longer than 1 month and $1K each; this is testing, not optimization. The goal is to validate your assumptions and really get a sense whether the channel could really move the needle for the traction goal or not.
Step 3) Focusing, or "The Inner Ring: What's Working." Assuming one of the middle ring tests was very promising, you drop everything else and really focus on that channel, now really trying to move it to get to your traction goal. This focusing step involves a lot more testing, but within that one channel to uncover strategies and tactics that work best.
DuckDuckGo
Swaypay
DuckDuckGo
Neon Fever Dream
DuckDuckGo
Reforge Extension
Give Local
DuckDuckGo
Product Hunt
Give Local
DuckDuckGo
SlidesUp API
Give Local
SlidesUp API
Hyper
Give Local
Bearupper
DuckDuckGo