Sitting on an OMMA panel, Daniel Redman was joined by approximately 300 peers all discussing scaling ideas and the many successes of Behavioral Retargeting. The topic of the conversation was ‘Retargeting to Scale’, exploring both the profit and pitfalls of Retargeting. As this marketing strategy begins to make a prominent presence in Internet Marketing, there are many things that could be taken from the panel.
Among Daniel, others at the discussion contributed their own inputs on the many advantages and qualities that retargeting has played in the online space. While the technology remains relatively consistent, many have found the benefits of retargeting to be extremely successful by employing a host of strategies and product couplings. As questions at the panel were being answered, those present and listening online were able to take away more than just responses from the panelists.
As Daniel stated in a response, “It truly is a double-edged sword, as you certainly want to isolate and segment your site traffic, however you don’t want to make your prospect pool too thin.”
The panel was moderated by Alan Chapell, Chapell & Associates. Below are Redman’s formulated responses/notes to the questions posed by the moderator and some additional ones that time could not permit.
As I think back to my days at DCLK with the Boomerang product, retargeting always sounded good in theory, but as a practical matter, it was so hard to find that shopping cart abandoner downstream. What are some unique and innovative ways that you’ve addressed the scale issue?
With the ability to segment your creative by site action, you can easily isolate behaviors such as abandonment. You can literally distinguish between any number of activities that are site usability deficiencies. Though, it truly is a double-edged sword, as you certainly want to isolate and segment your site traffic, however you don’t want to make your prospect pool too thin. The important thing is to continue adding segments rather than making them a singular campaign strategy. For example, rather than excluding visitors, who DO transact, develop an offer or cross-sell program for that group. This isn’t allowing for scale up, per say, but it’s at least keeping you from trimming to deeply. A method that eVisibility will be employing in 2009 to replenish Retargeting visitors with quality traffic is an integration with Social Media channels stimulated by organic traffic. In testing, this has rendered exciting results.
Provide an example of a how retargeting helped achieve one or more of your client’s goals. (e.g., driving sales, shortening the sales cycle) What is the metric for success?
An ecommerce client selling big ticket furniture had a sales cycle that turned 11% of all of his transactions after 9 days from first visit. As it’s a vertical with quite a few competitors, I expected that I would improve conversion rate after 9 days and see an increase in the long cycle transactions at the same time. That wasn’t the case; total sales cycle decreased, volume increased, and now even though we’re winning $1,600 average transaction values, we’re only picking up the post 9 day sales at a 2% clip.
As we are our own best client, eVisibility is achieving a 50% decrease on lead cost versus paid search. Coupled with what we call the ‘They’re huge’ effect, retargeting has been a stronghold in our approach. We’re earning more call-in leads and simple reconnections from lost prospects that attribute in part to retargeting.
The pattern here is that Retargeting seems to perform most startlingly well in a competitive space.
When thinking about conversions, what % of conversions are click thru vs View Thru conversions? And what are some of the challenges you see when addressing attribution – particularly with view-thrus?
Of course with view-thru, clients are concerned that they are paying multiple times for the same sale. There is some validity in this of course, depending upon what other marketing is concurrently taking place.
The message is simple on CPA campaigns; take into account click costs in determining a whole acquisition cost. It’s very nominal.
What impact are retargeting campaigns having in terms of increasing Search queries?
Ive certainly seen and felt an impact on branded queries, though it’s not a 100% infallible data comparison, because we have very little insight to the origination of a query string in most cases. It’s a near improbability to truly track the influence of marketing actions on a searcher. WE can gather a lifetime story and gauge tendencies with current technology, but until our brains have ‘Intel Inside,’ pinpointing actual influence is a pipe-dream.
How many segments (hp, search, product pages, and conversions) on average are you observing for any given brand and how does the message change for each user on these segments?
Retargeting is highly reliant on the quality and quantity of traffic and which goal or goals are established from the onset. Campaigns have been a success driving merely a few conversions a week. Others successful campaigns are driving over a hundred per day. 9/10 times clients aren’t seasoned enough to be able to take the next step with retargeting and allow for deep segmentation. Most are simply getting their feet wet with a basic retargeting model. Typically, clients run retargeting, experience great success, and aren’t interested in investing to get more from what is ‘already outperforming’.
What are some mistakes someone can make when running a re-targeting campaign?
I know we are having a conversation about scaling retargeting, but simply speaking, retargeting can fail when it is not coupled with a significant traffic source and all consistency/relevancy factors are ignored. For all intensive purposes, this is not a stand-alone marketing initiative. Retargeting provides a wealth of ways to customize in order to gain accellerated success. Landing pages, messaging consistency, frequency modification, etc. I imagine by dumb luck, sometimes you can have a well performing campaign by dropping a return visitor on the homepage with the same banners you were running in 1995, but it’s not likely.
During this down economy are pricing models changing? What % are CPM and what are acquisition/rev share?
As an agency, we are exploring more CPA driven deals. The more savvy prospect/client understands the mathematics of a CPA and will directly request CPM, but the majority of business owners these days really want to feel like they are getting what they pay for and CPA makes a ton of sense for them. If a client can sleep at night knowing that they are only paying for conversions, than more power to them and I will support that strategy 100%. CPA is really another way to attempt to play your cards against the house (a retargeting vendor). You’re challenging the technology vendor to ‘go ahead and beat me’, knowing that in your pocket is your site conversion rate (whether good or bad).
How do you see retargeting evolving over the next few years?
In the immediate future, I’d like to see greater integration across channels. The ability to control messaging in Paid search seems like the logical fit for me. More dynamic insertion capability, especially in use with sequential serving you could virtually create a new world for the end-user. Tag-teaming with the long but dead co-registration could be another avenue. Cookie lifespan is going to continue to be a hotly debated topic. How we, as marketers, can extend the life of a prospect and limit attrition will be critical as prospects become savvier. Overall, in the short term, retargeting shouldn’t try to second guess itself. It works; it’s great, let’s simply make a more efficient wheel and the keep the fed off of our backs.
What role (if any) does data play a role in the targeting process? And here I’m talking about data IN ADDITION TO the info typically used for retargeting (e.g., abandoned the shopping cart.) Is there value in pulling other data points, or does that become more trouble than it is worth? Why?
I recently delivered a webinar on data analysis. The summation of that was that data will set you free. This gets back to the point of isolation and segmentation. To give a specific application; a large volume publisher has an interest in retargeting but not an interest in footing a sizeable bill. By data-mining, it’s possible to identify the most engaged traffic by usability cues, isolate from outside navigation as much as possible, and retarget only the group that you can really knock out of the park.
Similarly, what role does creative play in retargeting? How much testing (e.g., A/B, sequencing, multi-variet, Etc.) is really going on at this point? Why / Why not?
In my experience, messaging and destination are the two critical elements in a retargeting campaign. Aesthetics finish 3rd by a great distance. Relevancy should be your grabber, not the imagery. This has been the problem with Myspace advertising in the past. You can serve a ton of impressions and if you use wild rich-media with in-banner vids and games, you’ll grab clicks; but they are empty clicks. A/B and multivariate are both valuable and should be taken advantage of. But, I can’t say that I am a poster child for doing so. It’s difficult to squeeze banner costs out of a client when they are already succeeding with their current set and their prospect pool is recycling frequent enough to not saturate.
