Moderator: Chris Winfield
Dave McClure, Angel Investor
Neil Patel, Co-Founder, Crazy Egg, KISSmetrics
Alright – first session of the conference. The only question is whether Dave’s gonna find my liveblogging so awesome that he offers to invest in my SEO audit company. Though my business plan is “as little client work and as much blogging as possible”
Dave’s up first.
LOL he’s got a slide up “Startup Marketing Metrics” (for pirates)
Talking to a few customers who you think are in your audience.
Large business – biggest challenge is beurocracy
Startups – nobody knows about your product. Rules – don’t market your product until people care about it. If your product sucks, you might generate viral marketing that hurts you. Your job isn’t to do a lot of marketing, its to understand your customers.
He says the concept of “you have to get it right the first time” is BS. – if you screw up, so few people will know about it it won’t matter in the long run.
He recommends reading “The Mating Mind”, and “Spent”, both by Geoffrey Miller. Sex + evolution + consumer marketing = awesome sauce.
Also Influence, the Psychology of Persuasion, and Understanding Comics, the invisible art.
The most important job in marketing – discover meaning.
Optimize for happiness – between your user/customer and your business. Look for a happy medium.
Move users from a lover value to a higher value through your product offerings and marketing.
How to discover meaning
Talk with people – observe their reactions. On the web it happens through text and images. Landing page tests and click-through. Experiment with keywords, images and calls to action. Look for the combination that leads to the best outcome.
Remember that you have different customer types.
The Lean Startup
Talk to customers, discover problems. Progress is not based on features, its about learning about customers.
Focus on fast and frequent iteration – try something out, see customer initial reaction. The faster I can get through that loop, the less I have to guess. A/B testing may work if you’re on the right track but you may find that you need to jump to something else altogether.
Focus on Product/Market fit.
Keep it simple & actionable – KISS (keep it simple stupid) ( don’t get lost in analytics).
The search for a business – customer discovery/ customer validation. The growth of a business – customer creation / scale company.
He says Sean Ellis promotes the idea of “if I take this away, would you scream?”
Don’t launch, spend a lot of money until you’ve discovered, validated.
Examples of Metrics to care about:
Best subject lines on emails
Email deliverability rates
Acquisition, Activation, Retention, Referral, Revenue.
(okay at this point, Dave’s just ripping through slides, concepts, and the slide up now looks like an information flow chart on too many RedBulls…)
What’s my business model? Get users, drive usage, or make money. He suggests getting usage is more important than getting users – whether they care – do this on 5 to 100 users. If and when you get to an 8 out of 10 or at least a 7 out of 10, you’re ready to move forward.
Understanding on a small scale is much more important than growth before that.
What do you choose to measure? Focus on a critical few actionable metrics. If you don’t use the metric to make a decision, it’s not actionable. What are the most important metrics for your business?
What features do you build, and why? when are you done? It’s when, you are relative to competition, you’re at least 2 points better (on a scale of 1 to 10). Base it on the alternative that customers are aware of.
Stop building, make what you have better.
You’re done when you’ve got something easy to find, fun / useful unique features that increase conversion.
Marketing what matters: Volume, cost, conversion.
Design and test multiple channels of marketing. Here are my channels – here are my costs to be in those, here are my customers in those, here’s the conversion rate for each…
Defining customer value over what length? Single session, an entire year? There’s a timing cost as well. I might pay for them initially and make money over time. Try to understand when you make money from that.
Up next is Neil
Track People, not Numbers
How unique is a visitor? People have multiple computers, smart phones… Someone could come to your site 20 times and appear as unique visitors.
Why not track people? One way to track people is to tie IP addresses to login credentials – so multiple IPs can be tied to one visitor.
How deceiving are metrics?
Very. You have to dig deeper.
He got 1 conversion from mobile phones – so he figures he’s not going to get mobile compliant. It turned out that they get lots of mobile visits, but from existing members – people really like it.
Some people need time before they buy.
They go to your web site a few times before they buy. Maybe they come from Google, come back direct, then maybe they see something on TechCrunch and then buy. You need to look at that.
Better tracking means higher ROI. Analyze the first entry source.
He’s a big advocate of ReTargeter. Someone goes to your site, then they leave, go somewhere else – and on that other site, your ad shows up.
Get Targeted Feedback.
If you get it from all the people who visit, feedback is useless. Lots of people who visit your site are not your target market so you get useless feedback.
By getting feedback from returning visitors, you’re getting better information.
With those responses, I then match them to IP addresses and see what their path was – to find my site.
Never stop asking questions.
People change their minds, so sometimes you have to ask the same questions again.
Not everyone engages with your site the same. When he first used twitter, he hated it, stopped using it. Discovered that twitter users were some of the most engaged. So even though he hated Twitter, he realized it’s an area he needs to make use of.
Helps you determine how loyal people are. (Events are different actionable situations on your site/ features). You can use event tracking with Google Analytics. Lets you know how many people (not unique visits, but IP tracked people who might use different computers / connections to your site).
Real Conversion Rate
Know your real conversion rate. Conversion tracking doesn’t end when they sign up. Example – if they sign up for the free version, then later on they upgrade, you need to track that along with where they first came from. (Alan’s note – It’s a lifetime value tracking thing so you see the bigger picture). Also know how long before they upgraded.
Q – About Google Analytics – the accuracy is okay, do you also recommend looking at log files?
Niel: They’re great at page views, which content is looked at. Supplement your info with other solutions as long as it doesn’t impact your load time too much.
Dave: Looking at GA distracts from looking at the most important metrics. In general, it gives you a nice pretty report, it doesn’t tell you the information you need to take action.
Q – I have a few products, not one clear path. It seems fuzzy.
Dave – follow them from first contact through your online presence over time. Also, you can ask them when they convert.
Q What type of tracking methodologies do you find good or bad?
Neil – There are some privacy issues.
Questioner – let’s say you’ve got visitor approval so privacy isn’t an issue…
Neil – you can use Flash to pull browser history. but I’m not going to get into some of the things you can do…
Q – You use the free sign-ups as the population to refine your marketing – is that what you’re doing – and how long can you live off the free sign-ups?
Neil – if you can’t get your free solutions to convert, you’re offering too much for free, or you’re not doing a good job marketing the upsell.
Dave – Some great info from Evernote, Dropbox – there’s some great discussion on “fremium” – there probably is having some value of having some free users, but its up to you to determine that value.
And that’s it – this first session is done!