Reading Time: 7 minutes

The Conversion Rate Optimization Process

You’ve read plenty of LinkedIn posts about businesses prioritizing optimization and their successes. You’ve familiarized yourself with the barriers to optimization and have taken endless notes. You’ve even spoken to stakeholders about the benefits of experimentation and have achieved buy-in.

Where do you go from here?

To take advantage of the benefits of conversion rate optimization within your organization, you must implement an experimentation program. To do so, you’ll need to develop and follow an optimization process.

Optimization Must Occur Across All Business Departments

The nature of experimentation is cross-collaborative. There’s no room for siloes in a transformative optimization program. Although each department may speak about optimization and testing in different ways, the most successful programs are those in which all departments are on the same page.

“When we have a broad range of expertise, we can focus on the most impactful opportunities. If any departments within your organization aren’t involved, it begins to limit what you can do. Plus, being able to execute and coordinate across departments improves the efficiency of the optimization program.”
– Grant Tilus, Sr. Growth Product Manager, Cro Metrics

Marketing & Sales

Businesses in all industries dedicate significant financial investments to their marketing strategies. Marketing budgets equal 11.2% of company revenue on average. Why? Because business growth is critical.

Conversion rate optimization enables you to improve your marketing ROI. This goes far beyond a positive return. Optimization asks the question, “How do we lift outcomes to increase the performance of our marketing efforts?” For example, although you may be spending $10K in Google and receiving $20K in revenue, there’s plenty of room for improvement by improving quality scores and landing page conversion rates.

As for sales, you must optimize for a specific goal. What are you selling? Do you want to increase subscriptions? Let’s run a test to assess the impact of increasing awareness of the product supply length. Are you trying to move more product? Let’s look for ways to reduce user perceived and actual friction to purchasing. Overall, your goals will differ for each business and may even differ across channels.


Product optimization allows you to test and validate new features as they launch to maximize customer adoption and minimize churn. For example, if your goal is to get more customers to order from your app, there are elements of your UX/UI you should test to optimize the app’s performance or increase its utilization.

Customer Experience

What’s good for the customer is what’s good for business. It’s important to assess and tweak your UI/UX and serve offerings that best resonate with the user. This not only leads to a better experience for them but also a better outcome for your business. For example, is one of your brand/product value propositions more impactful than the other for driving purchases? Design an experiment and test it!


Your engineering department is responsible for execution. Yet, engineers are often the busiest folks within an organization. They’re always coding and building—optimization is often the last thing most in-house engineers have time for.

This is best illustrated when considering win rate. For example, if a win rate for an experimentation program is 30%, that means that seven out of 10 builds end up in the recycle bin. A stretched-thin team may not be able to prioritize seeing this work through in order to realize the remarkable benefits of those wins given the expected returns and mitigated risks.

Often, businesses have more success when they outsource their test development to a team trained for building experiments. After all, your internal team may need to push projects out by a quarter or further. An optimization program with external engineers can move much faster, allowing you to tap into the rapid response engineering you need for gathering insights in order to become more data driven.


The analytics team wraps everything together into data insights you can use to move your business further. Data informs everything from planning to audience behavior. Yet, it must be organized and optimized to be useful. The best optimization programs have tightly knit relationships with the analytics teams to use data to inform our strategies and understand our user’s behaviors.

The Overarching Optimization Process

Although the idea of business optimization through experimentation is relatively novel, the ideal optimization process isn’t. And when it comes to developing a program for your organization, there’s no reason to reinvent the wheel.  Optimization is simply control, variation, outcome and repeat.

At Cro Metrics, we’ve run 18,000+ tests for various partners to develop our highly efficient and impactful process. Beyond the steps involved, it’s important to remember that efficiency and impact are the keys to a successful optimization program. Yet, to help you become familiar with the optimization process, let’s break it down into five specific steps.

Conversion Rate Optimization Process in 5 Steps
Cro Metrics has refined this optimization process after running 20k+ experiments in the last decade.

1. Build a Foundation

The first step is to build a foundation for your optimization program. This means identifying your program KPIs that align to your business goals.

When selecting KPIs, it’s critical to map out your overarching business goals and then break those goals down into smaller KPIs to optimize for. For example, if your goal is to increase your revenue per website visitor, a KPI to impact via testing may include “add to carts” and cart checkout rate.

2. Validate With Data

With KPIs selected, it’s critical to identify opportunities for experimentation by looking at your data to better understand the customer journey and overall performance. This data will point you towards opportunities to focus on versus simply basing your next move on someone’s opinion. This is what it means to be data-driven.

There are many ways to gather user data. For example, you can use Google Analytics to find holes in your website’s performance. Other techniques include heatmaps, customer surveys and usability tests. In the future, you’ll be able to use previous experiment results to inform this step in the process as you iterate.

3. Ideate & Prioritize

Time, money and other resources are always limited. So, you must stack, rank and prioritize. During this step, you must develop hypotheses for each test concept using the data you’ve gathered and previous knowledge or gained insights. For our clients, we use our client dashboard to easily track each hypothesis.

As a result of prioritization, you’ll develop a testing roadmap to help guide your experimentation efforts. Now is also the time to outline the prioritization process you’ll use now and for future experiments. At Cro Metrics, we prioritize our in-house tests based on several factors, including:

  • Business objectives
  • Potential lift
  • Previous test outcomes
  • Technical complexity

A custom prioritization framework allows you to place the tests that are more likely to move the needle first in your queue. With each test you complete, you’ll then use those results to further prioritize the tests inside your roadmap.

4. Design, Build & Launch

With a clear roadmap, you can begin to execute by designing, building and launching your tests into the wild. To do this, you’ll need an in-depth A/B testing tool. At Cro Metrics, we use any and every testing tool, depending on project needs. For example, some of the tools we use frequently include AB Tasty, Optimizely and VWO. We also use personalization to build and launch experiments on behalf of our customers.

To build an experiment, you must create your variation, select your audiences, add in your metrics and allocate your traffic sources. We use our proprietary Optimization Local Interface (OLI) tool to streamline test builds.

Quality assurance or QA is also a critical component of this step. Your experiment must look and work the way it should before you show it to users. Otherwise, the test is null and void. Your QA process should verify that your experiment functions properly, that the changes don’t negatively impact your website or app’s functionality and that your production environment matches your staging environment.

5. Analyze, Codify & Iterate

Finally, you’ll need to analyze the results from your experiment to quantify project impact. For winning variations, you can work to implement the changes ASAP if the change will support your goals. 

For losing results, don’t discount them too quickly. Win or lose, a well designed experiment will always deliver an insight based on your hypothesis. For example, is there another hypothesis to make or something you can try to optimize based on the result? This is what experimentation is all about—continuously iterating to improve outcomes.

Each result or any insights gathered should be documented and shared with your entire team. We also recommend reviewing the performance of your optimization program quarterly.

Staying Organized & Keeping Operations Moving Smoothly

An optimization program requires collaboration for success. To stay organized, you need a solid dashboard tool to help streamline testing ideation and reporting across your organization. At Cro Metrics, we use our proprietary client dashboard to review program performance and to see which tests are currently live and upcoming.

GrowthMap also enables our customers to approve tests and see results. A dashboard tool greatly reduces your internal overhead as it streamlines communication into a single source.

Possible Results From a Solid Optimization Process

Chances are, you’ve heard of The Sierra Club, the longest-standing and influential grassroots environmental organization in the US. Fighting for our right to live in a healthy world, donation volume is critical for the health of their organization. When a website redesign threatened that volume, they turned to experimentation.

Fast forward to the end of their program, the Sierra Club received a 22-40% increase in monthly donations and 14% in monthly gifts as a result. Experimentation also prevented a major loss by visitors who would have otherwise missed specific website cues. In addition, personalizing the donation experience drove a 30% lift in monthly donations.

This is only one example of the incredible business-changing results that come from a solid optimization process. Strategic optimization positively impacts your finances, operations and brand.

Big results are only possible by building a culture of experimentation within your organization. Our friend Stefan Thomke says it best:

“When management aims for big results, they cannot rely on lucky guesses, experience, or intuition alone…Running experiments should be as normal as running the numbers.” – Stefan Thomke, Harvard Business School Professor and Author

Start Scaling Your Optimization Program Today With Cro Metrics

These insights are enough to get you started, but the future of your program depends on your ability to continuously improve your strategy and scale. Allow Cro Metrics to use our proven optimization strategies to help you grow your program across your organization. Speak with an expert today.