In many of the threads here at WebmasterWorld there is a continuing refrain that goes something like:
* most of my traffic is still desktop, I get very little mobile traffic
and/or
* my mobile traffic converts at a fraction of my desktop traffic
Then when it is suggested that the mobile context is different than desktop context and that perhaps mobile needs to be approached differently the refrain goes something like:
* my site's responsive!
and/or
* why would I want to chase such poor conversions?
and/or
* Google is mismatching their traffic and my site.

One can lead...etc. et al

Some very serious numbers to consider:
* desktop sales have basically plateaued. It is a mature market. As a consequence the number of desktop users is going to be relatively flat for the foreseeable future. As that pie is no longer getting bigger the only way of increasing one's share of the desktop pie is to take it from someone else.

* last (2016) Cyber Monday saw ~3.5 Billion USD in sales. ~30% was by mobile.
---last Single's Day (China's Cyber Monday) saw ~18 Billion USD in sales; ~85% was by mobile.
What this shows is two critical information points:
1. mobile is NOT a negligible traffic/revenue stream.
2. mobile in the US has an enormous potential upside (if certain legacy structural problems are run around or over).
---eg: last American Thanksgiving's twin eCommerce 'holidays' of Black Friday and Cyber Monday WalMart's traffic was 70% mobile; purchases were 60% mobile.
What this shows is that beating the current 30% average is very possible.

The desktop with it's large screen, keyboard, mouse in a fixed location is a very different critter than a small touchscreen on a street corner or doctor's office. Sometimes the information desired is similar but the presentation or interaction needs to be different, sometimes the information itself needs to be different.

Forms are a huge pita whether desktop or smartphone. One benefit of mobile is that forms generally are being revisited and unbelievably simplified. The typical desktop form has over 20 input boxes. The newest mobile versions (which perform as well on desktop) have ~5.
Plus they've inverted address inputs putting zip/postal code first. Why? because that allows the site to auto complete several other address requirements. Then as visitor inputs street address an autocomplete drop down based on the foregoing usually means that only 2-5 key inputs plus a click are needed.
Plus only one name input is needed - and that allows for everything from one to a dozen - let the software do the heavy lifting, don't offload onto the customer.
Plus inline verification.
Plus input mask or scan for credit card info.
Plus smart default masks.
Plus...

Now take similar thinking and apply it to every single interaction between your site and the visitor.
* can it be made easier? simpler? more obvious?
* what brought the browsing public? the buying public? can/should I provide more?
---how+w5
* what are the contexts of the browsing public? the buying pubic? can/should I (better) target specifics?
---how+w5
* most buyers use 2 to 5 different devices before buying. Can/should you track more/better? Can I short circuit 'normal'?
* etc.

The initial switch to mobile has been shifting from desktop but the benefits traction is actually both ways.

Backing up a bit to that 2 to 5 device usage. Many/most webdevs think of mobile and desktop as separate entities. Almost no one here (or elsewhere) seems to consider web enabled TVs or automobile screens or the increasing number of idiotic IoT appliances. The web is in yet another transition and if one can't get mobile right (after a decade) or even desktop (after a couple) future success has to be considered rather bleak.

The 'easy' route of uploading content of n-quality in an it all looks similar bog awful stock CMS and cutting and pasting ad/af network code means that one is competing with 99% of the same stuff against 99% of competitors. It is, quite frankly, a race to the bottom jerked up by an occasional Google algo 'quality' change. Yes, it can still work; although mostly it doesn't. Or so say the numbers. And forum angst.

To succeed - and I don't just mean with Google - one needs to offer significantly better value than your next few competitors combined, to have both the quality and the ability to become a niche destination and not merely a query result.

You may not be 'killing it' with Google but someone in your niche is.
You may not be 'killing it' with mobile but someone in your niche is.
You may not be getting 5-10-20-40% return traffic but someone in your niche is.
You may not be achieving 5-10-20-40% conversion rates but someone in your niche is.
And that means that you have a lot of work to do.
Me too.