A landing page testing program has to be established before becoming an ongoing or periodic activity in the online marketing department. The success of the first test is critical for establishing the long-term momentum for subsequent tests. Let's look at some strategies you could pursue as you get started.
Start with a small test. Once you've demonstrated your ability to pull off an end-to-end landing page experiment, you should have enough support to continue. The key to this approach: appear nonthreatening and don't ask for a lot of help.
Components of a small test may include:
- Unimportant landing page: Don't try to fix your home page or highest-traffic landing pages first. Such efforts may be actively resisted before you've proven yourself. Pick a secondary landing page that has a reasonable amount of traffic (perhaps from a single online marketing campaign).
- Small diversion of traffic: If you only have one landing page, you can still conduct your test by diverting a small percentage of the traffic to testing alternatives. This will guarantee that, even if your tested alternatives underperform the current baseline, the overall drop in conversion rate will be small.
- Simple page changes: If you make simple changes (e.g., headlines, sales copy, call-to-action buttons), you won't need a lot of outside support to create your alternative test elements. You can also make such changes in a short amount of time.
- Basic test structure: Use A/B split testing (either on a granular or coarse level) for the test. The data analysis is simple -- there are no complex variable interactions or complex design matrices to worry about.
- Low-cost testing platform: Several low-cost tools are available for running basic A/B split and multivariate tests. Many of them are hosted on the Internet and don't even require installation. An excellent choice for this purpose is the free Google Website Optimizer tool.
But remember, your test must still produce meaningful results. Your data rate must be high enough to complete the test in a reasonable period of time, and the financial impact (or at least the percentage of improvement in the conversion rate) must be significant. Don't run your test if the data rate will be too low, or if the proposed test elements are unlikely to produce conversion improvements on the landing page.
Stay Below the Radar
Another option for running your first test is to do it out of public view. Sneak it into your normal activities.
To do this, divert a portion of your existing budget to testing activities. You and your testing team can also run tests "off the clock" in addition to your normal activities.
Many online marketing programs run on a CPA basis. If you bring in a reasonable volume of leads or sales at the target CPA, then people won't care exactly how you do it.
In this environment you can spend some of your media dollars on testing and still hit your volume numbers by squeezing more conversions out of the lower ad spend. Once you have a case study pointing to a real success, you can go into the open and get additional budget specifically for testing.
Sneak It Through Your Affiliate Program
You can also consider testing as a part of your affiliate program. Because improved landing pages increase lead volume without changing the affiliate's payout structure, they have a "force multiplier" effect on all of your current and future affiliates. You can legitimately funnel off some of your affiliate program management budget to the landing page optimization effort.
One way to bypass the need for budget approval is to roll your testing into your performance-based affiliate payouts. Try to find a testing company that will run landing page optimization projects on a performance basis (most offer guaranteed-performance fixed-price engagements). Many companies prefer this approach, because there is no financial risk to them.
The decision to handle landing page optimization in-house or to outsource can be complicated. But if you're considering outsourcing strictly for the initial proof-of-concept test, it has several clear advantages:
- Guarantee that resources and skill sets are available. Any reputable testing company that you hire will have all of the necessary skill sets in-house. They will often have access to specialists that you can't afford to hire and don't need on a full-time basis.
- No need to ride the learning curve. The testing company will also have extensive experience in writing test plans, running tests, and analyzing the results. You don't have to learn based on your own mistakes and are much more likely to have a positive outcome for your first test.
- Allows you to shift blame or take credit. If the test doesn't go smoothly, it can be blamed on problems with the testing company. If it goes well, you can take full credit for having had the foresight to hire them.
Make the Financial Case
Most top managers are receptive to financial arguments. They understand that there are two basic ways to improve profits: increase revenues and lower costs. They like and appreciate subordinates who also speak the language of money and understand their contribution to the company's financial success.
Focus on the mission-critical parts of your Web site and understand the value of the intended conversion actions. Take the time to dig up financial information that allows you to estimate the lifetime value of the conversion action.
Build your own financial model once you understand the revenue and variable cost percentage for your landing page. Remember to use conservative assumptions and estimates to make your business case stronger and more defensible.
Landing page optimization can often mean large potential profit increases. Your project may be the most financially promising marketing initiative available to the company. Once you get high-level executive buy-in based on the financial case for the project, you should have the political support to move it along and get the active cooperation of other team members.
Meet Tim Ash at SES Chicago on December 7-11. Now in its 11th year, the only major Search Marketing Conference and Expo in the Midwest will be packed with 70+ sessions covering PPC management, keyword research, Search Engine Optimization (SEO), social media, local, mobile, link building, duplicate content, video optimization and usability, while offering high-level strategy, keynotes, an exhibit floor, networking events and more.