• Home
  • Using Data to Improve Your Sales & Marketing Process

Using Data to Improve Your Sales & Marketing Process

A successful sales & marketing process is a science and an art.  This article helps uncover some of the ways to use data and testing to help improve your sales process.

Develop a List of Questions & Hypothesis

The purpose of data is to answer a question (Ex: Who is my best customer?).  To specify what you are really trying to solve, it’s best not only to develop a question but also develop your hypothesis for thee question (Ex: Our best customers are 35 year old females in healthcare).  By developing a hypothesis, you know exactly what information to collect and there is less room for ambiguity.

Invest in Systems to Collect Data

Look for systems that can easily help you collect, share, and analyze your data with minimal effort.  Top of the line CRMs have contact form features to help you collect data.  Cheaper solutions include Google Docs/Forms, Survey Monkey, and Excel that work fine with a little more human effort.

Record Your Results in The Proper Format

When your done collecting data, you will need to analyze it so you can use it to answer your hypothesis and learn more about your company.  When developing surveys or forms to collect data, try to ask questions that can be answered with yes/no, numbers, percentages, rankings, drop downs, options, etc.  If all your questions allow you to fill in unique responses, it’s going to take a long time to identify trends.

When You Aren’t Quite Sure, Use Qualitative Testing

Qualitative testing is best when you have multiple variables that are unknown.  For example, if you are trying to validate an idea or test the user experience of your product, qualitative testing is the best option.

You usually don’t need to test with more than 3-5 subjects to start to identify relevant patterns.  Results do not need to be statistically significant and the process should be iterative.  Qualitative tests include:

Validation Interviews

To validate a concept, the best method is sitting down with prospective customers or talking to them on the phone.  Forewarning, it is very tough to get honest answers from customers and prevent them from being biased.  Tricks to prevent bias include asking the same question in multiple ways and throwing in random questions that disguise your objectives so your audience won’t cater their responses to what they think you want to hear.

Product Usability Testing

Observe customers using your product and identify their pains and problems.  Great for helping improve your product and the user experience. Watch Steve Krug Usability Testing on YouTube.

Competitor Usability Testing

Observe customers using your competitor’s products and identify their pains and problems.  You can do this before you even have a product.  Great for helping to validate an early concept and discover what workarounds people have created to avoid their problems.  Run Google, Facebook, or Craigslist ads to find five people using your competitor’s products.  Watch and ask questions.

Marketing Usability Testing

Observe customers using your marketing website and identify their pains and problems. Helps test your marketing message and how strangers will perceive your product.

Five-Second Messaging Tests

A smaller form of marketing usability testing.  Ambush strangers.  Go to your local coffee shop, ask if someone will give you 5-10 minutes and you’ll buy their beverage.  Explain why it helps you.  Tell them you are not selling.  Show your landing page and ask the following questions.  What does the user think this product does?  Who does the user think the product is for?  Can the user figure out how to get the product?

Wizard of Oz Testing

If you really want to get a good idea whether a concept will work, build a front-end design but no backend functionality.  It requires a human doing all the work on the backend to service the client.  Zappos tested out their product buy taking pictures of shoes at local shoe stores and selling them online.  When someone purchased they would have to drive to the shoe store, buy, and ship the order.  They lost money but validated a billion-dollar concept.

Prototype Testing

Show them something that looks like your product and ask for their reactions.  Never ask them if they think it is a good idea or would buy.  A lot of times they either don’t know if they would buy or they are going to say yes just to be nice.  The closer you can make it look to the real product the more accurately you can predict whether people will use that product.

When Your Trying to Prove Something, Use Quantitative Testing

Quantitative testing is best when you have only one variable that you want to test.  Goal is to develop results that are statistically significant by collecting a larger sample of data.  Examples of quantitative testing:

Landing Page Testing

Setting up multiple landing pages to test out how many people are interested in a product, your messaging, etc.  Use Google analytics to track which pages have the greatest success.

A/B Testing

Good for getting statistically significant data that you can hang your hat on, proving hypothesis, understand what customers are doing and making decisions on what to cut and what to improve.  Not good if you are constantly making many changes on your product, message, etc.  In other words, not great for early testing unless you are isolating your changes and can test them individually.

Feature Stub (Fake Door)

Add a button that says upgrade or buy to test whether users will pay additional for a certain feature or version of your software.  When they click on the button, take them to a page that says: “we are currently working on this feature, if you’d like to be updated when it is complete, please send us your email.”


Surveys are not good at getting in touch with your users because you will either have to ask closed ended questions that will bias the results or open-ended questions that won’t get you to what you need to know.  Surveys are okay later in the game when you know your market and are asking them to prove or disprove certain hypothesis.

Other Methods of Collecting Data

Testing Sites

If you have the budget, there are sites that helps you do testing like,,,,,  They are not cheap.

Google Adwords

A great way to test who would be interested in buying your product or service is to setup multiple Google Adwords campaigns that target different markets or include different messaging on your product or service.  Over time you will be able to see who’s looking for you and what they are looking for.

Google Keyword Planning

Similar to above.  Instead of setting up Google Adword campaigns, go to the Google Keyword Planner and type in words that describe your product or service.  You will be able to see which words and messaging are more popular than others.

Google Website Analytics

If you place a Google tracking code on your website you can see which pages are performing best.  This tool is extremely helpful if you segment your site into landing pages and content targeted for different audiences.

Old Fashioned Research

Sometimes you don’t have to re-invent the wheel and can find answers to what you are looking for by doing some quick online research.


One of our favorite lines in our office is “better wrong than vague.”  It means that it is best to pick a path, test that path, and then discover if you are right or wrong.  Companies that are vague in what they do and who they serve never amount to much because they cannot allocate all their resources to their best opportunities.


Leave a comment

Your email address will not be published.