Ever wonder what the secrets to good ROAS (return on ad spend) really are? Well how about starting from your website and finding ways to improve conversion rates.
Better conversion rates mean more people complete your sales/lead generation funnel – without having to increase traffic or spend any more on ads!
Here’s one cool way so many people miss.
Have you been tracking the conversion rate from “outbound clicks” to “landing page views”?
We just nearly doubled ROAS by getting the ratio up from 55% to 65% (65% is still too low aim for 80%+)
Niche: Ecommerce; collectables/memorabilia
Problem: after a promising first few months, performance dropped.
Solution: below.
When it comes to knowing Facebook and being able to manipulate it to get results, me and my team are pretty hot, perhaps even amongst the best in the world.
BUT working as an agency we often find it’s the things out of our control that dampen performance. If you run ads for clients you’ll know what I mean.
And if you’re working with an agency, you’ll see what I mean below.
=========
If you notice below the good ROAS days (ROAS column, green = good) are pivoted around good site conversion rates (Sales CR% column) more than CPC (Cost per click).
(tip: if you want to create reports like this for your Facebook data go check out my YouTube explainer video)
However as we’ve been doing a lot of testing on Facebook the CPC had started fluctuating.
C.R.O. – Conversion Rate Optimisation.
We analysed Hotjar (we use it for Polls, Heatmaps and recording user journeys) and Google analytics to pinpoint the exact issue and one of the key metrics was conversion from product page to cart (VC > ATC in the data.. View Content > Add to Cart).
What were the changes?
- Site speed improved by removing Shopify plugins delaying the pixel load. The site owner liked to install new Shopify apps he’d see in Facebook groups, completely unaware that it often negatively impacted the load times on his website
- Product page improvements with visual changes to make it easier for prospects to find what they need such as making the returns/delivery information clearer and making the option to select variants easier to see.
After analyzing our ads performance trying to figure it out any guesses what had happened?
The client installed a plugin which the app developer gave them a guarantee won’t negatively impact performance. 🤕
Guess what it did.
The worst thing about this is when Facebook sees traffic being sent but not converting, it tries to switch traffic/targeting to find ‘better’ audiences.
The problem is if the traffic/audience is NOT the problem, and your site is, getting stability on Facebook ads performance is a nightmare, hence ensuring your website speed and conversion rate are stable/strong is SO important.
Facebook’s algorithm is almost too powerful in that it is too sensitive to your website performance that a good campaign can go bad because of website issues.
It’s like your website is sending happy or sad signals back to Facebook based on whether conversion is high or low – and if Facebook sees too many sad signals, the algorithm often thinks its the traffic that’s the issue so it tries to find better users… when often in reality Facebook’s working fine and it’s your website with the issues.
So whether you’re running your own site campaigns or for others:
– ALWAYS ensure your pixel is loading fast. I use tools.pingdom.com and test from multiple locations or GT Metrix.
– ENSURE your landing page experience is slick. I use GA and Hotjar mainly to go deep in analysing and understanding painpoints to improve landing page and funnel conversion then look to run split tests to improve conversion rates.
– MONITOR the conversion rate between outbound click and LPV (landing page view) to ensure it remains fairly high (80%+). A low number is usually because of either spammy traffic (ie fake profile clicks that don’t want for the site to load) or high bounce rate because your site (and therefore pixel) takes too long to load
Let me know if this helps you too!