New Product Update! Deliverr has expanded product dimensional requirements Learn More

Blog > Conversions > eCommerce seller strategies: When to use data vs. experiments

eCommerce seller strategies: When to use data vs. experiments

Your eCommerce seller strategy is the key to your success, but how much of that strategy should be data-driven, and how much should be experimental?

With plenty of research on product optimization, pricing, advertising, and shipping available, adopting a data-driven approach is easy, but not always the best for your particular business. You could get stuck using a strategy that works for another seller, but is ineffective for you.

On the other hand, experimenting on everything instead of learning from others’ mistakes means you’ll be starting at ground 0, instead of getting an advantage (even as a new seller). Taking an experimental approach to everything can be costly and unnecessary in many cases.

In this article we explore a few different ways to develop your seller strategies by balancing data and experiments, and where they make the most sense.

The different types of selling strategies

Lets start by breaking down the two seller strategies we’ll dive into; data-driven and experimental.

1. Data-driven selling strategies

Data-driven selling strategies use existing data to fuel direction.

The benefits of adopting a data-driven approach include:

  • Easy availability: Thanks to digitalization, there’s a vast amount of data available both internally and externally
  • Proven results: By following steps that have proven success in the past increases your chances of replicating that success
  • Objectivity: Numbers don’t lie. Data overcomes any assumptions that may be curbing your success
  • Technology: Advances in machine learning mean that you can leverage data to predict patterns, anticipate needs, and personalize your messaging

The disadvantages of adopting a data-driven approach can include:

  • Cost: You may need to purchase data or the tools to capture advanced customer data, which can be expensive
  • Determining the wrong metrics: It can be challenging to identify and track the right metrics for your business and goals, and if you follow the wrong data it won’t help and could even hurt your business
  • Backward-facing: Data is a snapshot of what’s been successful in the past, overlooking the fact that customers and trends change
  • Context: Data doesn’t always provide context, which can lead to incorrect conclusions and assumptions

2. Experimental selling strategies

Experimental selling strategies, on the other hand, experiment with different approaches to learn about customers and find the strategy that works. In the words of Mark Zuckerberg: “Move fast and break things.”

The benefits of adopting an experimental approach include:

  • Responsiveness: You can quickly realign your sales strategy to new trends and patterns, without waiting for corroborating data.
  • Competitiveness: Adopting a different approach to other online sellers can lead to new insights that put you ahead of your competitors
  • Tailored: Experimental strategies embrace individuality and avoid the one-size-fits-all nature of data
  • Discoveries: With the freedom to experiment, you can discover new approaches that may work even better than data-backed methods

The disadvantages of adopting an experimental approach can include:

  • Cost: As with purchasing raw data and data analysis software, it can also be costly to experiment with different strategies, especially when they don’t work
  • Time: Significant time can be wasted before finding the right direction for your strategy, giving your competitors a head start
  • Bad habits: Experimenting without guidance or the ability to shift your thinking can lead you to make the same types of mistakes repeatedly, getting you nowhere in a hurry

Data-driven vs. experimental selling strategies

As you can see, both data-driven and experimental selling strategies have their perks and considerations. We recommend combining a mix of the two to keep your business optimized, yet innovative.

Your own best implementation depends on what you are trying to achieve, the amount of data available and the time you have for experiments.

Let’s take a look with some of the most common sales strategies for eCommerce.

Optimization

Listing optimization should be a key focus of any online business – getting your products seen and clicked on. The technical nature of optimization, coupled with the amount of data out there, makes a data-driven strategy crucial.

Without using the data available via keyword tools, identifying the profitability, competition, and search volume of search terms would be near-on impossible, and certainly time and budget consuming.

Once you are successfully targeting short-tail keywords for your listings using data-backed insights, then you can allocate some of your time and budget to experimenting with long-tail keywords to increase your competitiveness and long-term success.

Pricing and promotions

Your pricing and promotions strategy is the main determiner of how profitable your eCommerce business is. Price too low, and your limited profits will prevent business growth; price too high, and your customers will look for cheaper alternatives.

Ideally, you want to use existing data to understand the impact of price and value (i.e. free shipping) on demand, and the impact of promotions on conversions per sales channel and per product. Gathering this data, however, requires experiments.

Therefore, new sellers will need to adopt an experimental approach to gather the insights necessary for a data-driven approach in the future.

Advertising

Like optimization, the success of marketplace advertising is greatly improved with a data-backed strategy, working with information such as audience segmentation, keyword competition, and the best timing.

However, that data isn’t always available. Amazon Ads has swarms of available data that can help you to implement a successful Amazon advertising campaign. Walmart, on the other hand, doesn’t have as much data available – giving you the opportunity to play around and discover your own advertising analytics and trends.

Experimental advertising strategies are also great for other forms of advertising, such as content marketing and social media posting – allowing you to work with your audience to discover what they like and don’t like.

Shipping

Your shipping strategy is an interesting one, working best as a combination data-backed and experimental approach. Data tells us that fast and free shipping is a strong purchase driver – making fast shipping programs such as Amazon Prime and Walmart Free 2-Day Shipping  highly profitable.

How you go about achieving fast shipping requires a more experimental approach – discovering the quickest, cheapest, and most reliable service for your business. This might be in-house fulfillment, marketplace fulfillment (i.e. Fulfillment by Amazon), or outsourced fulfillment. We could call this approach “data-driven” – using data to direct you, with experiments to determine the best method.

Closing message

How should sellers approach their business: armed with data or ready to experiment? It all depends on how much data is out there and how much time and space you have for conducting your own experiments.

We enjoy adopting a combination approach that uses data to guide but allows the freedom and fun to make discoveries of our own.

Subscribe to stay updated with new articles!
Share
Tweet
Share
Pin