Having set out the outline of the experiment it's now time to get some web pages out there to be indexed. Whilst the time line will be determined by how Google reacts to changes in the sites, the narrative will follow the typical scenarios encountered by real companies.
The first event is the launch of the brand's website. As this is intended to provide a baseline search result for the brand name it's not optimised in any way, it's a single page with a hero image, a few paragraphs of meaningless content, an image carousel and some product modals - bang on trend for a "modern" website design.
Also at this point in the scenario the shops are starting their development process. Whilst none of them have yet worked out any of the details, the optimised site have put up a single page on their domain that says they are coming soon and will be selling the brand's products. They have also registered their site with Google Webmaster Tools and set up a sitemap that will let them track when pages are indexed.
By contrast, the control site don't have any web presence on their domain name. This isn't as uncommon a scenario as might be expected, building a new website takes considerable time and time and effort which is focussed on the full final result. When you're thinking of complex tech stacks and full page designs then putting content on a server can seem like a daunting prospect. Taking a step back and putting a single static HTML page up is actually something that can be done quickly and will start to get your site noticed. (If you are going for an ego driven "big bang" launch then this isn't going to fit with that - which is yet another reason to avoid that scenario.)
This is part of a series of posts that cover experiments in SEO.