Experiments aren't just for scientists. In fact, ecommerce companies who conduct them can see a solid payoff, to the tune of 2% to 3% better advertising performance per experiment run, according to Facebook Marketing Science Research.

That's all the more reason for sellers to experiment with their advertising dollars—and with ad channels themselves. This experimentation provides you with data that helps you determine which ad strategies work for your business. Learn why this experimentation is worth it, and how to carry out these tests.

Why should sellers experiment with advertising channels?

We turned to Tim Mueller, head of business development at DealNews, for insight on how experimenting with ad dollars can benefit sellers' businesses. "By experimenting with new partners and channels, sellers can find new avenues to increase their sales and ROI (return on investment)," Mueller said. "For example, if you are only advertising on Amazon or on Google, you are most likely going to cap out at a certain return on your ad spend. You are rarely going to find a diamond-in-the-rough ad campaign that gives you an incredible return consistently in the long run."

The issue with sticking solely to Amazon, Google, Facebook, and the like comes down to competition. Due to the nature of these platforms, Mueller explained, competition will arrive and the return on ad spend will decline, as the cost will increase versus the sales the ads are able to generate. But by finding new ad partners, you’ll see the following benefits:

• Reducing how much you depend on a few sources of sales and traffic
• Discovering sources of a higher ROI for your ad spend

"In the long run, your business will be better off because you have developed a network of sources of traffic and sales," Mueller said. "You can use this network for different initiatives, such as product launches or closeout sales." Basically, the results of your ad-partner testing will make it easier to invest strategically, based on where each of these partners has excelled for your business.

How to experiment with ad channels

Now that you know why ad-channel experiments can benefit your business, the question becomes how to conduct them. First, you’ll want to decide which companies to run these ad experiments with. "Sellers should look for advertising with partners that have been in business for a while and have a background producing ROI for their clients," Mueller said. "The sales team at any new partner should be able to give you guidance on what works, what does not, and what to expect."

Compiling accurate data is key when conducting any experiment, and the way to get this data with an ad-channel test is via performance tracking. "The best experiments are those where you can test a percentage of your overall budget with a new traffic source and track performance," Mueller said. "Every new partnership that you source should be tracked either using a tracking link of some kind or a limited-time exclusive coupon code."

During your ad experimentation journey, remember that outcomes are never guaranteed. And an ad campaign that increases sales for one business may be a dud for another. But even if a campaign doesn't yield a positive ROI, the next one might. So keep experimenting with your ad dollars!

—Katie Roberts, Blog Editor, DealNews