It may be tempting to provide a code sample to be implemented on all product detail pages or send a link to Google’s product structured data implementation guide. While this may technically be enough for your developer to get the job done (although sending them a link might be a bit of a pain – your developer probably doesn’t want to have to become a structured data expert before they can start), it doesn’t do a good job of demonstrating why you’re making the recommendation or why it should be a priority. So how could you communicate this request in a way that really sells its value?
Presenting your idea in a hypothesis framework makes it much easier to ground your recommendation in the context of how it will improve the site, and the tangible effect it may have. A hypothesis framework for a product structured data request could look like:
- We know that product structured data makes our product pages eligible for rich results.
- We believe that winning these rich results could increase our click through rates by X%.
- And will result in an Organic traffic increase of Y%.
- Which translates into a revenue increase of Z%, based on our current conversion rate and AOV.
This framework explicitly shows what you hope to implement, why you want to, and a tangible measurement of the benefit to the business.
However, it’s not always easy to estimate what the performance increase of an implementation like this could be until it’s actually complete. Luckily for many technical SEO tasks (canonicals, structured data, hreflang etc.) it’s straightforward to run small scale tests using Google Tag Manager. Test the recommendation on a handful of URLs, monitor the performance of these URLs against the rest of the site and use this as the basis of your estimation.
Sometimes it’s not possible to estimate the potential impact of the recommendations; maybe you can’t use GTM to test, or you may be preparing for a Google Core Update (for example, it’s a good idea to start addressing Core Web Vitals now, ahead of them coming into play next year, but there is no way of telling how much they’ll impact rankings until then).
In this case, it can be useful to present support from Google’s (or another Search Engine’s) documentation, or a piece of third-party research that shows the task’s importance.
Now you’ve presented the recommendations in a way that should help it become a priority, the next step is to ensure you’re making the work as appealing and straightforward as possible, so that developers can easily pick it up and complete it without having to do huge amounts of background research, without the risk that’ll you’ll need to make amendments or revisions once it’s complete.
Going back to the product structured data for our Halloween Crocs. Rather than sending a code example and asking our developer to implement it on every page of the site, there are a few things we might want to consider.
For example, are there any products that you don’t want this implemented for? Many sites include product widgets on content and product detail pages (under a ‘Related Products’ block or similar). It’s likely you only want to implement structured data for the main product on a page, so you need to make sure you’ve identified these caveats and pointed them out.
Or perhaps the developer is particularly busy at the moment, and won’t have time to dynamically create structured data on each page for some time. If you only presented your solution (to have this dynamically created for each page), you might have to wait a long time before it would be implemented. However, if you explain what you are trying to achieve and work with the developer to find a solution, it may turn out that they could create a free form meta field in the CMS that allows you to paste in your own JSON-LD in much less time.
If your recommendation lends itself, it can be helpful to provide code samples for your developer to work from. Canonicals, structured data and hreflang tags are all great examples of implementations where you could provide your developer a code template as a starting point.
Other tasks may be trickier to put together a code sample for, so try to point out some other sites or pages that are good examples of what you’re aiming for.
Finally, make sure your recommendations are realistic. Take some time to understand how complex your request is going to be to implement, and work with your developer to see if you can come up with a sensible compromise or middle ground. You may have the ‘perfect solution’ in mind, but often sites don’t need highly complex ‘perfect solutions’ to see results (‘good enough’ is often good enough!). Consider whether the time and complexity of your task will be outweighed by the benefits or whether there are more straightforward recommendations you could focus on.