A frequent dilemma I hear education portals have is not knowing what the best traffic channels are to generate quality leads. They either don’t want to upset their lead buyers with lead fluctuations or they don’t have a team in place that can test new channels well.
Most of the top portals have the budget and lead buyers in place to extensively test each traffic channel but it takes time, the right analytics team and a watchful eye to get it right. After studying the demographic and info preferences of education leads, my team has spent the last several months testing many different traffic channels.
We had meetings with all our big lead buyers prior to testing and basically told then exactly what we would be doing. We told them we’d need them to be patient through the process as some sources may drop while others increase in quality – but we wanted to stay in constant contact with them and get as much analytics back in real time as possible. We told them the exact metrics we would need. They had the ability to return below-benchmark leads right away, and after all was said and done, they would most likely come away with higher enrolling leads.
We wanted this to be a meaningful test knowing that once we were done optimizing, split testing and stress testing, the remaining traffic channels and sources would be of the highest quality.
Surprisingly, every one of our buyers was excited for our test and was willing to be patient with us through the process.
Here’s a highlight summary:
Email Marketing
We had a small amount of email marketing experience but nothing nearly as extensive as we did over the last few months. We went to some of the biggest list managers with great reputations and worked with them in building creatives to run on a CPL basis. We assumed this was going to be the highest quality traffic but it wasn’t.
Using the same creatives across different lists can dramatically change quality. And just because we mailed through some of the most reputable list managers, didn’t seem to matter. At certain times, we were promised lists that “were great quality” but turned out not to be. And lists they were not sure about ended up being great finds.
Once we weeded out the poor lists and their data sources, email marketing proved to be high quality (but not the HIGHEST quality).
The age demographics of our email leads tended to be older (40 years old+). Lists that had origins from grant sites did extremely poor. Geo-distribution was skewed toward the east coast. Enrollment rates were 3.8%.
Social Media Marketing
The best part about social media marketing; mostly Facebook these days, is that you can target your audience via their exact demographic. Any age or interest you want to target, you have it at your fingertips.
Does that mean this type of marketing is effective?
Yes, it can be. But you have to have the budget to do it. Facebook is very expensive to advertise education these days (well over $1 per click) unless you are creating ads like this targeting 16 – 18 year olds (sarcastic)-
Trust me, you don’t want to do this. These leads don’t back out.
You have to target older ages and that is where clicks get very expensive. However portals with well established lead buyers can afford this, and I recommend testing social media marketing with older age groups.
Display Marketing
The king of education display marketing based on visibility is ClassesUSA, mostly due to their long standing Microsoft deal.
For us, display marketing had the biggest amount of controllable variability.
Quality is purely based on the sites you target. For example Yahoo Education is a gold mine (tip), while sites like AddictingGames.com is a waste of time. Most people aren’t going to buy media directly from sites so you have to make sure you go with a reputable display network and negotiate your terms carefully.
This is where having a skilled internal media buyer comes in handy. There are many tricks of the trade that come with media buying that you only learn through experience. My experience as an affiliate marketer years ago taught me the ropes, at a price tag exceeding 100k.
Display marketing surprisingly turned out being our second best traffic source at an enrollment rate of 4.1%.
Search Marketing
Hands down the best converting traffic source during this test. By converting I mean enrollments (and I realize affiliates don’t care about enrollments). But enrollments are most important to colleges, AORs and well established portals.
Sure bids are $8+ in some cases, but if you have a solid lead gen path in place, don’t be scared to try it. Search marketing is the only source that nails “user intent” on the head. Who better to sign up for your college than someone searching “business degree at online colleges”?
Enrollment rates were 7.4% on our search campaigns.
Affiliate Traffic
Yes, we tested affiliate traffic. Remember when I said display channels have the highest amount of controllable variability? Well with affiliate traffic you have the highest amount of NON-controllable variability.
We tested five of the best reputed affiliate networks and results were very mixed.
Patterns were the most difficult to spot and I felt too many layers removed from having control of the campaign. The best advice I can give if you are going to test affiliate traffic is to start by allowing email, search and display publishers, and limit each new publisher to a 50 lead initial cap. Yes the networks will give you pushback asking for more, but just tell them you will open the cap as soon as you see the first 50 leads come in per publisher. When that happens, be sure to keep your end of the bargain and give immediate feedback to the network and publisher.
Another tip is to take your planned CPA and build in an assumed XX% bad-data scrub. Then lower your CPA by that amount and give it to the affiliate network at the lower rate. If the networks’ percentage of bad leads exceeds your built-in assumed scrub rate, cut that publisher or network. If you find a publisher that is well within your allowable scrub, think about giving them a custom payout. It is hard to find good affiliates, so when you do you’ll want to keep them promoting.
Conclusion
That’s about it for our test. Although it took several months, it was well worth the effort, time and analysis. We now have a four fold increase in overall traffic and leads at a much higher quality level than before.
I’d like to get my team doing another test in about six months. I’m a big believer in constant testing and scaling.
Are there any traffic sources you have experimented with that I didn’t mention? Leave any comments you have below.
{ 2 comments… read them below or add one }
Great insight into the edu lead gen mentality. So which of these channels are you going to stick with and which would you recommend most?
I personally think that affiliate traffic is the best among these various channels of traffic. If you simply go by numbers, you might initially get mixed results but over the long haul… affiliate traffic medium is a lot more consistent in terms of quality leads.. Again, just my two cents!