Automated WP site speed optimizations SaaS idea
Would IHs out in the world be interested in a turnkey, pre-configured web speed up solution that works out of the box for Wordpress? Not focused on internal WP optimizations such as DB caching etc but only on end-user perceived site speed using CDN, defered/async JS, critical CSS analysis, automatic cache purging and pre-fetching into CDN of not only static content but html and texts.
I use WordPress for various marketing sites and with my background as a developer I have done lots of optimizations for speed over the years. Not only on WP but with various other web technologies and platforms.
There is a list of stuff that always needs to be put in place and various plugins for Wordpress but all the plugins are lacking in some aspects in my opinion and running multiple plugins with overlapping features creates risk and increases complexity. Besides all optimizations cannot be focused on WP because you also need to put a CDN in place... I use Cloudflare which helps with some caching but you need to put custom page rules in place and really pay for Enterprise level to have an optimal caching setup for admins, comments and logged in users... But I think I can build a solution at $50-$100/month with everything solved for you.
With one of my current sites I have become very frustrated with the work required which takes focus away from the real marketing work and feels like it should just be ready out of the box. So in many respects this is also to scratch my own itch!
Thanks a bunch for any feedback you want to offer! Maybe this already exists and I haven't managed to find it.
imo an obstacle to the solution itself is that it will require some techy input and understanding of optimisations from the user anyway. You mentioned async for loading JavaScript assets, well that alone can break the entire site since each script may have a dependency on another.
I've used so called automated tactics that add async and perform other optimisations, in the end it just breaks as there is no reasonably solid way to figure out the dependency tree, which css styles are truely unused across the site etc.
E. G you may see no dependency on a script hence add async, you may see no use of some css styles and remove them, but the developer may have bundled those knowing that different dependencies exist on other pages, and decided to pack it all as a good compromise between saving a few kb in transfer and holding the page rather than having to worry about all the pages.
Performance optimisations is not easy, even with manual intervention it is quite easy to break some non obvious aspect of a web page. Solutions already exist and imo totally suck.
You may go for it and see if you can deliver on your promise, but here you go you have potentially a large set of your customer prospects already hit by experimenting with other (non working) products for their use cases.
Also, the monthly subscription model is hardly digestable since a lot of users would see the optimisation needed only once, since wp designs tend to remain as is for a year or longer, just updating of contents from publishers.
Look, even page performance insight tools are broken or at least have become broken as they highlight non sensical recommended optimisations, they don't take in consideration the site hollisticly, half the recommended changes wouldn't actually make a significant enough improvement to actually spend the time making such change. They are mere automated suggestions, quite useful on hundreds/thousands of assets and aspect of page loads. If even a few of those optimisations were blindly applied, I guarantee you something would break.
Yeah great input and as you mention automated js & css minification/bundling does have a tendency to break sites - been there myself many times... And most sites have already optimised it in various ways. Probably this is not the main selling point, and some sites will be broken with a no toggles js/css optimization. Still I think most wordpress sites would work with a generic optimisation and identifying if the js/css is already optimised. I guess the challenge here is to test how many sites would reasonably work with the standardised optimizations and how to handle onboarding for clients and verifying if the site actually works with the solution.
I also appreciate your take on a monthly fee potentially being challenging. The monthly fee is required to cover CDN, bandwidth and compute time. but yeah would have to be clear that the value is provided month on month and not being a one time thing.
Wpengine and a few other services already pack preconfigured nginx caching for full pages with cdns and internal optimisations and more without people needing all the details no? ..
Thanks for that input, I have looked at these services. Yes you are correct that they pre-configure wordpress and hosting but they focus on WP performance and caching on the hosting side but again have a bunch of toggles that a website admin needs to understand where in actuality all the dynamic content could be edge cached automatically and purged after an update. They don't really provide anything other than what regular wordpress performance plugins in combo with Cloudflare provide.
Wpengine describe their caching here https://wpengine.com/support/cache/ my idea goes much further than this and is something that I have created on a few sites which can result in sub 30 ms response times for wordpress pages.
I would need to clarify how this is different to other existing services... Thanks @hatkyinc always difficult to step outside your own bubble and see what people would compare your product to!
I haven't been in that niche for years, I just remember there was a need yet a lot of solutions.
From plugins you mentioned that would do static content CDNs
To plugins and systems that make wp a backend system that fully generates static pages that are used to serve pages from even a memory based store to these full service WP optimised hosts that might go as far as be your techies with an email getting you deep dive answered about what plugin is causing issues, fix suggestions on your side and extra optimisations on their side..
I think the niche was a bit saturated and you might need a lot to stand out.
Any ideas or input and tips on good ways to validate that there is a paying market for this would be welcome.