500 Startups Growth Handbook

Conversion Rate Optimization?

What is CRO?

Conversion rate optimization is about identifying & eliminating friction points that prevent visitors/users/customers to move from one step to next - usually from one page to the next page, or one step in the app to next the next step, or from one step in the buying process to the next.

How?

You start by identifying the step that’s least efficient (lower conversion rate than expected in that step), identify why (observe customer behaviour, understand their thought process, ask direct/indirect questions), and test to validate the different ideas that will eliminate the friction until you increase the conversion rate to a satisfactory level.

If there is more than one conversion rate you can optimize, pick the one with highest upside, or biggest impact on bottom line but require least amount of resources (time, $$$). The pages or features with most traffic are quicker to validate/invalidate hypotheses (less time). If the test is time consuming, brainstorm what could be a minimum viable test alternative that can directionally prove the hypothesis. If conversion rates start to flatten out, figure out if it’s because you’re hitting local maxima by brainstorming out of the box ideas.

Common Mistakes Founders Make

Copying others/not testing. Just because it worked for someone else doesn’t mean it will work for your business model, your audience, your geography, your product etc.
Too many changes in one go. “Oh! We’re launching a whole new version of our website because our earlier version was ugly/not converting/not convincing…” Hard to identify cause and effect pairs.
Testing small/changes that won’t produce significant impact even if it wins. Especially if you’re small startup with a not a lot of traffic, it’s a waste of resource to be testing incremental changes, test hypotheses that if validated will have a massive impact on your business. Plus, best case outcome for optimization of known variable is a local maximum, innovation/thinking outside the box is the way to hit the global maximum - the earlier you are, the less you have to lose by innovating, and the more you become prosaic by optimization.
Running split tests with too little traffic - start with pages with most traffic
Traffic Drivers (Adwords, FB ads, influencers, etc.) don’t match the site experience / driving the wrong type of traffic

Fictitious Example

Let’s say your data shows you that activation or onboarding is your biggest bottleneck - i.e, once a user starts using/buying your product, they remain engaged and make repeat purchases or use the app repeatedly, but the biggest problem is getting them to try, then observe user sessions, talk to them etc. to understand “why” and remove the friction. Example: with self-serv B2B products, where they need to integrate/upload data to see the full potential of the app, give them an option to play around with dummy data you’ve created and have it ready to go. Another frequent example is having users create login or enter any critical information you need (shipping, name etc.) AFTER they try out the app, or after they enter their credit card information. Another example would be mobile apps asking for information that the user may not have handy when on the move.

Real World Example

Venture Harbor found that if they increased urgency in their landing pages, they would convert at a higher rate. Turns out, they were right! Read more about it here.

Related Keywords

Activation, onboarding, funnel optimization, conversion rate, landing page, split testing, analytics