Bing - Optimizing and Scaling an Old Campaign (3)
Hi guys,
I have an old campaign in Bing that's been running for years in one form or another. We own the funnel and it was profitable last time I looked. The campaign spent $650 in October (that's without me having logged in for months) and cost us $1.94 per opt in - which takes us a couple weeks to break even on, then start churning a profit.
Goals:
- Reduce the cost per opt in;
- Increase volume.
Initial Analysis:
Read through redrummr's awesome thread and realized that the data we have is not worth much because the moron we paid to optimize it a while back make the following mistakes:
1. For some campaigns he forgot to limit to only the USA. From reading on here, I believe it's best to split into east and west coast anyway to facilitate day parting.
2. Left mobile and tablet traffic on. We're an info product company, people ain't buying our shit on mobile yet.
3. Seems like there are two different conversion pixels setup and firing...and I believe there is no way to tell the difference between them from just the normal campaign window.
What the data is useful for is identifying negative keywords, so I'll be using that.
Also picked up Mehdi's suggestion that if QS drops place keywords in the header, title, h1, footer etc. That's going to help because I recall us having killer QS's and now they're all 5 or 6. Plus I've seen it mentioned that volume is highly dependent on QS.
Plan moving forward:
1. Create new campaigns based on the highest volume keywords and adgroups
2. Desktop Only. Split by east and west coast - USA only. Separate campaigns for Search and Display (could use an opinion on this if anyone has one?)
3. Only one tracking pixel - we'll use sale of initial product. Lag can be between 0-14 days after opt in, but I figure in the long run it'll give better data and we'll be optimizing for QS and CTR in the beginning anyway.
3. Look through "Search Term" Dimensions and pick out words that have nothing to do with our offer and add those as negative keywords in the new campaign. Going to just use common sense for matching type - if anyone has specialized advice here I'd love to hear it.
Screenshot of campaign that was running, last few months worth of data:
As a prelim test, I created 2 new campaigns each with the highest volume/best converting keyword. No negative keywords for now because I used the phrase match instead of broad match - don't see how any combination of those 3 words other than the exact phrase could have anything to do with our niche, very open to feedback. Created quick keyword "stuffed" squeeze to see if QS goes up. Also using this opportunity to split test ads.
Questions:
1. Any good reason to use broad over phrase for a 3 word keyword that needs to be in that particular order?
2. Any reason not to separate Search and Display campaigns?
3. Leave Search partners on or turn off?
4. Anyone prefer tracking the opt in instead of initial sale?
Cool Follow-Along! I don't have Bing-specific advice, I'm afraid, but I'll be watching this one with a lot of interest!
Keyword stuffed hard squeeze type landing page came out with low quality scores. Main reason is "landing page relevance" which is weird, I would expect it to be "landing page experience."
So going to try a landing page on our hub site which is well established among search engines. If quality score goes up I may need to look at using a cloaker to show this page but redirect to our hard squeeze because the cost per optin is about 1/4 (plus I've been looking for a reason to play with one for a while.)
Quick note: If the QS on the new LP is low then:
1. need to test with opt in thank you page on same domain (currently redirects to our direct response funnel type domain.)
2. remove lightbox from LP
Home > Search / PPC (Adwords, Bing, etc.) >