If you start a website from scratch, what's your first step?
Usually when I begin a project I try to put up some content on the site blog, and then look to provide value to people on Quora and Reddit and find a way to bring in a polite link. Sometimes I even make somewhat relevant content as answers.
How about you? I know each type of site is different, would love to hear different inputs. Some people I know run $100 adwords to see if theres any traction.
Find customers in real life and show them the website/product to get their reactions/feedback. Ask them to speak their thoughts out loud as they process what you show them.
This provides you with far more useful gradient of insights than the binary conversion rate.
I think this is almost universally great advice. At minimum, you're guaranteed to miss a few UX issue, but you also stand to learn a lot about your customer. Bringing your internal model of who the customer is and what they want into alignment with reality is at least 75% of the startup game!
Agreed, this first step should always be validating your product/service! It doesn't matter how pretty it is or any fancy features it might have, all that matters is if it solves their problem.
If yes, keep talking to more potential customers.
If no, adjust and keep talking to more potential customers.
I haven't launched enough new websites to have any kind of established pattern or trend, but here are some of my thoughts based on the experiences I do have, the people I've interviewed, and the books I've read:
First, every business has its own unique advantages and challenges when it comes to growth. Thus, there's no one-size-fits-all for how to get started. Things that work well for one founder might be a total waste of time for another.
So the TL;DR is that your first steps should depend entirely on who you envision your target customers being. As you learn more about your customers and their needs, you should tweak your product and your approach accordingly!
Something I sometime do
Look for specific 'news' sites. By this I mean, a submission site that will accept your title, summary and article link. These will be primarily community sites and not real news or article submission sites.
I did this recently for a programming language submit site and drove about 5k views to blog post just from that one submit (for 100% free). csallen has about 25x (and more) traffic from Hacker News, but that's a great example too.
I look for these kinds of sites now based on the vertical because they work.
I usually send it to a few people who are not tech-savvy to see if they understand the concept and the website. Even if the product is solving my own problem, I'm not the average visitor because I've build it myself and because I like Product Hunt, BetaList etc., I am good at understanding a landing page.
For example: I'm now building a community for social entrepreneurs and will start with Slack (and a separate question board). People that don't know Slack have a hard time to understand what it is and why it can be better than a LinkedIn group or Facebook group.