This is a fact:
No matter how good your website is — people are going to leave without converting.
Instead of getting frustrated — let’s squeeze every last bit of ROI out of that exiting traffic.
I mean… heck… they are already on their way out — might as well try something bold.
Here’s how it works…
We have two critical pieces of information about the people that visit our website:
- We know what they’re interested in – If they visit a page about peanut butter, we can safely assume they are interested in peanut butter. If they visit a page about jelly, we can assume they are interested in jelly. 🙂
- Where they came from – We know if they came from Facebook or Google search or a blog that linked to us.
And when we have this information … we can make them a relevant offer.
We call it “On Site” Retargeting…
Here’s the “On Site” Retargeting Process…
We’ve shared several of our ad retargeting tactics, and onsite retargeting works the exact same way…
Use the information you have about the visitor to present them an appealing offer.
You could have the best offer in the world, but if it doesn’t appeal to the right visitor at the right time, you’re sunk.
So, we set our website up to trigger relevant offers to exiting visitors based on…
- Keywords in the URL
- Visitor type (New visitor? Returning visitor?)
- Traffic type
When all was said and done, we were able to get 2,689 more leads from traffic that was intending to leave the site. Oh, and we didn’t just increase lead magnet consumption we also:
- Increased DM lab sales by 30.22%
- Decreased bounce rate by 8.52%
- Increased time on site by 35%
Let’s dig into exactly HOW we did this!
We used a third party tech called Optimonk for all of our campaigns.
There are a fair amount of great techs out there, and this one just happened to work wonders for us.
We created 5 total exit intent offer campaigns:
- Campaign 1 – Generic ‘Toolbox’ lead magnet for returning visitors that did not have any campaign keywords in the URL, e.g., facebook, blog,
- Campaign 2 – Generic ‘Toolbox’ offer for new visitors that did not have any campaign keywords in the URL, e.g., facebook, blog,
- Campaign 3 – Facebook Ad Template lead magnet triggered for URLs that included ‘Facebook’
- Campaign 4 – Social Swipe File lead magnet for URLs that included ‘Social’
- Campaign 5 – 212 Blog Post lead magnet for URLs that include the term ‘blog’
(For more information about lead magnets, read this article.)
For those of you scratching your head and saying ‘Wait, are exit offers cool with the Facebooks and the Googles?’
We’ve been assured they are in compliance because these types of exit offers are not an exit hurdle and provide value to the user (where a ‘WAIT DON’T GO’ JavaScript alert doesn’t provide inherent value and is a hurdle). Better news if you aren’t comfortable using exit offers on your PPC campaigns, just don’t show exit offers to your paid traffic sources (yep… you can do that) … and worry no more.
Let’s take a look at each campaign and their results…
Campaign 1: Generic Toolbox Offer Return Visitors
If you were a returning visitor on our site in the last 2 weeks, you probably saw this offer. Unlike some of our other campaigns, this was a delayed overlay not of an exit offer.
In an effort not to shoot ourselves in the foot, we made sure that our campaigns did not show up on our sales pages or dedicated landing pages.
We also didn’t want to mix up our messages from people who were visiting the site from our newsletters. Whenever we send an email, we have a particular goal in mind, and if they click through we don’t want to distract them.
Finally, we didn’t want to seem like that overzealous person you meet at a networking event that tells you WAY too much information WAY too fast, so we opted-to trigger this offer after the visitor had spent 15 seconds on the site.
Before I get to the results, let’s talk about the ‘New Visitor’ campaign and I’ll discuss this campaign’s global results.
Campaign 2: Generic Toolbox Offer New Visitors
This campaign had the exact same parameters, except for the visitor type. We targeted new visitors with the same offer but framed it slightly different.
Let’s look at some of the numbers, shall we?
There were 5,211 pageviews on the Toolbox’s landing page, and 2,485 out of it was delivered by OptiMonk.
The following screenshot contains the pageviews of the ‘Thank You’ page of the Toolbox. Visits to the “Thank You” page for this Lead Magnet indicate a conversion. So, for example — there were 1,802 visits to the Digital Marketer Toolbox “Thank You page” from the Optimonk campaign indicating 1,802 conversions.
See below…
On average the campaign had a 68,9% conversion rate (3,212/4,661). The OptiMonk group outperformed it slightly with a 80.7% conversion rate (1,802/2,231)
Campaign 3: Segmented Offer – Facebook Campaign
Similar to the generic offer campaigns, all of our segmented offers did not trigger on our sales pages or landing pages and we excluded all newsletter traffic.
If a person visited the site and the page they visited had ‘facebook’ in the URL, this offer would trigger instead of the generic offer.
In the image below, the visitor has entered a blog post about The 7 Step Facebook Advertising Game Plan.
This visitor will see the relevant Exit Offer called The Ultimate Facebook Ad Template Library.
It’s all about putting the right offer in front of the right people…
It’s also important to note, we set all of our settings to only show the offer once, we didn’t want to hit people over the head with an offer they’ve seen already.
And here are the results!
There were 8,617 unique page views on the Facebook template landing page (this is a heavily trafficked landing page from paid sources, which are omitted). Optimonk sent 471 unique views to this landing page.
Now let’s look at that conversion rate, here’s a screenshot of the ‘Thank You’ page for this campaign:
On average the Facebook campaign had a 58.4% conversion rate (5,038/8,617). The OptiMonk group greatly outperformed the standard campaign with a 90.6% conversion rate (427/471).
Campaign 4: Segmented Offer – Social Swipe File
Here’s the next segmented offer – this was triggered for any URL that contained ‘social’
And here’s what the offer pop looks like…
Here are the results!
There were 1,185 unique page views on the social swipe file landing page. Optimonk sent 226 people to this landing page.
Now let’s look at that conversion rate, here’s a screenshot of the ‘Thank You’ page for this campaign:
On average the social swipe file campaign had a 69.6% conversion rate (825/1,185). The OptiMonk group also beat out the standard campaign with a 84.5% conversion rate (191/226).
Campaign 5: Segmented Offer – 212 Blog Posts
This was the third and final segmented offer we created for this campaign. This offer was shown if the URL contained the keyword ‘blog’.
And here’s the offer…
Here are the results!
There were a total of 2,217 unique page views and OptiMonk sent 239 unique page views.
Now let’s look at that conversion rate, here’s a screenshot of the ‘Thank You’ page for this campaign:
The campaign as a whole had an incredible 93% conversion rate (2,064/2,217) Impressively the OptiMonk group saw a 91.2% (218/239) conversion rate, but did lose out when compared to the entire campaign. However, a 91% conversion rate from click-to-conversion is always okay by me.
The results for each campaign were fantastic, but we also wanted to optimize our offers further. We ran several split tests on all of our segmented offers, but that’s a story for another time!
Here are the results for the ENTIRE campaign…
Now let’s look at the entire onsite retargeting campaign as a whole (I’m only showing this to give you an idea of how the whole campaign functioned, when making decisions about where to add new offers, new triggering methods, etc… I always advise avoiding averages because they don’t tell the entire story).
You can also see, that those who have seen OptiMonk bounced less (53.39%). Those who have not met OptiMonk bounced with 72.65%[1], so OptiMonk decreased the overall bounce rate by 19.67%.
OptiMonk also increased other notable engagement metrics (seen in the chart above):
- Average time spent on site by 54% (breaking these down into seconds)
- Average pageviews per user by 26% (1.92 to 2.42)
In total, we generated 2,638 leads — and here’s the kicker — from traffic that was EXITING our website.
We were unable to track the last 51 leads in Google Analytics because they were acquired during the test period of the two-step form. So after we include that variation we’ve increased our total number of leads by 2,689 in only two weeks!
Final Thoughts
Onsite retargeting is a game changer, and my biggest prediction of 2015 was the growth of more refined personalization technologies …
Now I’m no Nostradamus (though I do have a beard) but if you implement this incredibly simple form of personalization on your own site, it will greatly help your business.
[1] 45,955*0.5339+97,004*X = 142,959*0.6646, where X = bounce rate of those, who have not seen OptiMonk
Have a question?
Ask the DM team and 9,036 other members in the DM Engage Facebook Group!
Not a DM Lab Member? Learn more here.
