4 ms·

Reverse Engineering Marketing For Startups

61 points by heapster 13y ago

ActVen 13y ago
There are some interesting tools mentioned here. But, remember that conversion to a customer is much more important than popularity. Conversion isn't something that can be looked at too well from the outside. Your best content is always going to be on topics that exist on the customer's pathway from the discovery of a need or problem to the selection of a solution. That type of content will almost always win out over what is popular. Example: "Differences between Contractor and Employees" -> "Salary for X position" -> "Payroll Taxes" -> "Payroll Software" -> Conversion to Customer of a Payroll Software Company
- JRFuentes7 13y ago
  I have to agree with you. We had a blog post, that percentage-wise, was by far the most popular in terms of traffic. But it also had the worst "path to conversion rate." We hypothesized that this was due to the post being interesting in search results but not necessarily attractive to our target customer.
carlospox 13y ago
Great collection of best practices. Thanks.
WhitneyLand 13y ago
This kind of thinking is pure gold for technical people. How do I hire or network with people like this? Figures she already is a founder...
- nana_gb 13y ago
  Check out www.inbound.org it's a bit like HN for marketers.
ericd 13y ago
For anyone looking at this and potentially taking it seriously, I can tell you from abundant first hand experience that Compete is garbage - not only is it frequently an order of magnitude off in its numbers, but it's very frequently directionally wrong from month to month, sometimes hilariously so. Edit: For example - http://techcrunch.com/2011/08/29/anti-web-analytics/ http://techcrunch.com/2011/08/29/anti-web-analytics/
- prakster 13y ago
  True, but the other info in that article is very useful and revealing.
- nana_gb 13y ago
  Your statement about the accuracy of Compete data is not entirely false, I have yet to see it match up closely with sites I have direct analytics data on. But if you are only using Compete you are doing it all wrong. Think of Compete, Alexa, Quantcast, Similarweb (A newer site with some very cool data http://www.similarweb.com/website/ycombinator.com http://www.similarweb.com/website/ycombinator.com), Adbeat.com (For display intelligence), NerdyData.com (Which is really good at letting you search for tracking tags that marketers use) and the dozens of other sites and services currently available on the market etc as different color paints on a canvas. Trying to create a photorealistic painting (accurate picture of what's working for your competitors) with just one color will yield poor results. Like the author in the post does, you need to combine and mix them for the the best outcome. Additionally, Michelangelo can use the same paint that you and I have access to, and end up with the Sistine Chapel ceiling. While I'd be proud to just be able to end up with a ceiling that's an even shade of whatever color I chose. A skilled marketer masters and understands the nuances of these tools and use them to work magic (At least in the eyes of someone without this knowledge). In the past, I use to jump from site to site collect data, but now when I need to research a site or market, I'll use the Follow browser extension http://www.follow.net http://www.follow.net since it aggregates all the information into one place. It's not perfect because I still have to collect data from other research sites, but only when when I see it's worth digging deeper. In my 13+ years of marketing online, I can say without a doubt that if you can figure out how your top 3-5 competitors are marketing online and how effective their marketing is, you can almost always out market them or at the very least have the data you need to be very competitive. Disclaimer: I'm good friends with both founders of Follow.net and Adbeat.com I did quite a bit of product dev work with the Follow team early on and have done customer acquisition for both companies.
spyder 13y ago
In my experience the services that trying to estimate traffic without tracking codes on the sites are very inaccurate (Compete, Alexa, SEMRush).
viennacoder 13y ago
Really useful post. Wish I had read this 4 years ago!
ycwes0 13y ago
If anyone wants a great free tool to see how websites acquire their traffic so you can reverse engineer it, check out http://follow.net/ http://follow.net/. It shows you how any site acquires traffic through SEO, SEM, Display, and Contextual Advertising, and Social Media channels.

61 points by heapster 13y ago

There are some interesting tools mentioned here. But, remember that conversion to a customer is much more important than popularity. Conversion isn't something that can be looked at too well from the outside. Your best content is always going to be on topics that exist on the customer's pathway from the discovery of a need or problem to the selection of a solution. That type of content will almost always win out over what is popular. Example: "Differences between Contractor and Employees" -> "Salary for X position" -> "Payroll Taxes" -> "Payroll Software" -> Conversion to Customer of a Payroll Software Company

JRFuentes7 13y ago
I have to agree with you. We had a blog post, that percentage-wise, was by far the most popular in terms of traffic. But it also had the worst "path to conversion rate." We hypothesized that this was due to the post being interesting in search results but not necessarily attractive to our target customer.

carlospox 13y ago

Great collection of best practices. Thanks.

WhitneyLand 13y ago

This kind of thinking is pure gold for technical people. How do I hire or network with people like this? Figures she already is a founder...

nana_gb 13y ago
Check out www.inbound.org it's a bit like HN for marketers.

ericd 13y ago

For anyone looking at this and potentially taking it seriously, I can tell you from abundant first hand experience that Compete is garbage - not only is it frequently an order of magnitude off in its numbers, but it's very frequently directionally wrong from month to month, sometimes hilariously so. Edit: For example - http://techcrunch.com/2011/08/29/anti-web-analytics/ http://techcrunch.com/2011/08/29/anti-web-analytics/

prakster 13y ago
True, but the other info in that article is very useful and revealing.
nana_gb 13y ago
Your statement about the accuracy of Compete data is not entirely false, I have yet to see it match up closely with sites I have direct analytics data on. But if you are only using Compete you are doing it all wrong. Think of Compete, Alexa, Quantcast, Similarweb (A newer site with some very cool data http://www.similarweb.com/website/ycombinator.com http://www.similarweb.com/website/ycombinator.com), Adbeat.com (For display intelligence), NerdyData.com (Which is really good at letting you search for tracking tags that marketers use) and the dozens of other sites and services currently available on the market etc as different color paints on a canvas. Trying to create a photorealistic painting (accurate picture of what's working for your competitors) with just one color will yield poor results. Like the author in the post does, you need to combine and mix them for the the best outcome. Additionally, Michelangelo can use the same paint that you and I have access to, and end up with the Sistine Chapel ceiling. While I'd be proud to just be able to end up with a ceiling that's an even shade of whatever color I chose. A skilled marketer masters and understands the nuances of these tools and use them to work magic (At least in the eyes of someone without this knowledge). In the past, I use to jump from site to site collect data, but now when I need to research a site or market, I'll use the Follow browser extension http://www.follow.net http://www.follow.net since it aggregates all the information into one place. It's not perfect because I still have to collect data from other research sites, but only when when I see it's worth digging deeper. In my 13+ years of marketing online, I can say without a doubt that if you can figure out how your top 3-5 competitors are marketing online and how effective their marketing is, you can almost always out market them or at the very least have the data you need to be very competitive. Disclaimer: I'm good friends with both founders of Follow.net and Adbeat.com I did quite a bit of product dev work with the Follow team early on and have done customer acquisition for both companies.

spyder 13y ago

In my experience the services that trying to estimate traffic without tracking codes on the sites are very inaccurate (Compete, Alexa, SEMRush).

viennacoder 13y ago

Really useful post. Wish I had read this 4 years ago!

ycwes0 13y ago

If anyone wants a great free tool to see how websites acquire their traffic so you can reverse engineer it, check out http://follow.net/ http://follow.net/. It shows you how any site acquires traffic through SEO, SEM, Display, and Contextual Advertising, and Social Media channels.