Direct Response vs Branding: Like Oil & Water, or Can They Be Emulsified?
I come from a background with several years in direct response advertising and lead generation under my belt, and I noticed some key differences coming into a retail-centric environment at PMG. Retail companies can seem behind the curve compared to their pure-play peers. For example, many of them have a marketing mix focused on brand and offline channels. What I’m seeing is an opportunity for retailers to gain a big competitive advantage by leveraging the same technologies and strategies used in highly competitive lead gen or pure-play markets.
Imagine you take a retailer and give them that same focus toward their digital acquisition funnel. They start implementing things like lifetime value calculations at the ad group and placement levels, perpetual landing page testing and testing the conversion funnel. Imagine where that puts them. They now have an edge on their competition. So what approaches do we use to get there? Well, I know a few. Below are several ways I’ve found to bring direct-response strategy to retail PPC/PLA efforts:
Change your attribution model from last click. Last Click attribution undervalues the channels subjected to it in favor of direct type-in, because people are more likely to return to a website via the history suggestions that pop up as they type in their browser’s search bar. Under last click attribution, the acquisition channel loses that credit and it goes straight to direct type-in. If direct traffic were a channel we could invest more in, this might make sense — but it’s not. The most readily available solution for many is to look at the attribution modeling features available in Google Analytics which allow you to see the channel impact of various models. From there you can make a leap to the custom data-driven attribution model feature available to GA Premium users or take baby steps and switch from last click to last non-direct click.
After having your mind blown from seeing ROI jump in the model comparison tool, you can adjust your bidding easily with the right tools. Many tools have the ability to adjust attribution models, but DoubleClick for Search stands out for my example as they are rolling out the ability to bid based on Google Analytics attribution models. However, there are other integrations with some of the leading attribution platform vendors you can look into.
When your business model relies on the performance of 8 pages, like you see with direct response or lead gen offers, it’s easy to see the importance of landing page testing. For retailers who have literally thousands of pages and may not have the framework to scale tests across them it’s easy to see why landing page testing may take a back seat.
As a retailer, it is important to understand the impact of gaining one point on your conversion rate. Basically, every additional point to your conversion rate means one extra conversion per 100 clicks. To a conversion rate of 4%, one point can mean a 25% lift in revenue! Calculate the impact of an additional point to your existing traffic and you can see how beneficial it is to increase CVR.
You can get a lift like this from something as simple as changing the flow between the ad click and order confirmation page, or by uncovering a bug that caused the page to hang. It is just as important for an agency to test things that happened before the click as it is to test the pages that you’re sending them to.
No matter how good your products are, somebody out there with a similar product can get a leg up on you by providing a better flow. That flow translates to a better shopping experience, which turns into conversions. Turn the tables and beat them by making the shopping experience more enjoyable.
There are several solutions for landing page testing:
You can build an internal tool and framework based on redirects. This works if you have an abundance of dev and creative resources to crank out tests for you. It’s a pretty solid investment, fairly basic from a technology standpoint and gives you total control.
Purchase a testing platform. I don’t have experience on one of these yet, so no comment.
Once again I’m going to call out DoubleClick. Since DoubleClick uses redirects from the click through URL they are a natural fit for testing. They have a landing page testing tool available that is quite easy to use. The redirects mean that there is no interruption to the search URL and you can swap out pages in a jiffy!
Call up PMG and have them do it for you. This is my preferred choice.
Yes, this means build outs for days; but these may be different than you’re accustomed to. I’m not referring to haphazard buildouts that bog down your 3rd party bidding platform, but calculated precise buildouts with a solid structure. Investing time in tools and learning clever tricks will take you a long way, but learning scalable approaches is key for this. Everything should be broken out for profitability measurement, even if that means you have to look back a year or two to determine profitability for a particular keyword.
Of course we mustn’t forget cross negatives so that bid changes don’t divert traffic to other ad groups. This goes for ad group themes, match types, campaign themes, so on and so forth. We want to block all bad traffic, so if a search query drives garbage traffic, it isn’t just flushed downstream to the next available ad group to skew performance there too. Think of it as front loading effort to save time optimizing and maintaining down the road, with the added bonus of improved bidding efficiency.
For scalability and optimization purposes it is ideal to have labels or dimensions tagged for products, product categories and brands to build out analysis and reporting without headaches. Imagine being able to map PLA and search data together for cohesive optimizations, maintenance and reporting!
One huge point of separation between branding and direct response efforts is the popular opinion on affiliates. Yes, I’m referring to that dark underbelly of the online marketing world. Coming from experience in direct-response, not all affiliates are bad! Affiliates can be a dangerous tool in the online marketer’s toolbox, so we must watch them carefully, but not all are inherently bad.
Stay in touch
Subscribe to our newsletter
Affiliates drive huge traffic volume and are ready at a moment’s notice to pack up their caravans and move on to the next best offer (like carnies). I have yet to see affiliates worked into a brand-conscious marketing portfolio, but there are good ones out there that could be leveraged for a powerful and innovative emulsion of brand and direct response.
Posted by: Josh Parr
2 MINUTES READ | August 1, 2017