Client Use Case: Measuring Marketing Channel Success

Pretty much any business owner will tell you they want to grow.  However, not all know the most efficient way to grow their business, particularly when it comes to spending money on marketing. 

That is the challenge one of our recent clients had.  As a newer chiropractor with an already successful business, they had minimal insight into which marketing channels worked best for their business.  They knew they needed to keep growing as a business, but also didn’t want to spend money on marketing that wasn’t working for them. 

That’s where Simplified Analytics was able to help.  We worked with them to learn their business, collect the necessary data, put it into a format that allowed us to analyze it, and identify the most effective marketing channels. They’re now able to make informed decisions about where to leverage marketing spend for the most return, and save money on those that weren’t performing well.

Messy Data

It can be intimidating to think about how you could end up getting value from information you collect as part of your day to day operations, especially if its in different places and in various systems. We worked with the Chiropractic practice to identify the systems and data necessary to analyze their marketing channels. To say their system wasn’t designed to easily extract data is quite an understatement. However, through a few rounds of trial and error, along with the company’s helpful customer service team, we were able to get all of the data we needed.
Once we had all the data we needed, the next step was to combine it . The bulk of the data came from two systems that don’t talk to each other, so we had to find a means to bring the data together. While usually not ideal, we were able to utilize the name of clients to bring the data together. With any dataset, oddities with the data will come up! From an analytics client perspective, this can be a challenging and potentially frustrating part of the process. However, this is a very normal process, regardless of how clean your data already is!

Analysis

With our goal of answering which marketing channel gave the best results, we now had the data we needed.  We opted for a common approach in marketing measurement called “attribution”.  This allows the revenue from a marketing channel to be attributed for a certain period of time.  For the Chiropractor’s practice, measuring in 30-day increments ended up making the most sense based on their business cycle. 

Measuring the success of marketing channels can be challenging, as there is a lot that needs to be considered and accounted for.  While retention of clients are important, how does retaining a client for 12 months that spends $50 a month compare to a client who stays for 6 months (half the time), but was spending $200 per month (spends 4 times as much in their time)?  In our analysis, we accounted for this by looking at the total revenue generated by each marketing channel over time.  The end result?  We found one of their marketing channels far exceeded the other marketing channels.

The below graph is a generalization of our client’s actual results to protect their identity and actual revenue numbers.

How the Analysis Can Be Used

For this Chiropractic practice, the first marketing channel far exceeding the other marketing channels allowed their team to not only know it was a successful channel (something they already suspected), but know how successful it is. When making decisions to spend on marketing, they’re now able to make a more educated decision when allocating marketing funds.
Beyond understanding their current state, they can also look for similar marketing channels to marketing channel #1 to see if the same success can be observed.

When in a Project Should You Think About Measurement?

I’ve seen the following scenario play out many times.

  1. Someone has a great idea
  2. The great idea gets approved
  3. People work on the great idea
  4. The great idea gets implemented
  5. The people in charge want to know if the great idea ended up actually being a great idea
  6. The data people get asked to help answer if this is indeed a great idea

Measuring if an initiative or project was successful can be hard enough, however it is made more challenging if the measurement conversation occurs after the fact, rather than before its implemented.  While starting a measurement conversation after implantation doesn’t mean measurement is impossible, it can limit the quality of the measurement.  This translates to being less confident about our answer to if the great idea was in fact a great idea.

In general, the best time to start talking about measurement is at the same time you’re discussing the other details of the project.  However, I know that isn’t always possible, so the best rule of thumb is the earlier the better.

Why is Earlier Better?

Having more options is (usually) better than having fewer options.  The earlier the process measurement conversations start, the more opportunity there will be to make a decision in one way or another that allows the measurement to be more accurate.  Here is an example to help illustrate this point.

Your company is launching a new loyalty program for its customers.  There are currently concerns about the benefit of the existing program, so the company’s leaders have given approval to launch a new program.  The team in charge does a ton of great research on best practices on programs, finding features that benefit both the customer and the company.  The new programs gets rolled out to all customers on the current program.  Unfortunately, the measurement conversation wasn’t a major emphasis, so the plan was to measure sales per customer to determine if this program was better than its predecessor.  Even more unfortunately, at the same time this new program was being rolled out, a global pandemic impacts your company’s sales.  How are you supposed to know if the new loyalty program is better than the last?

Since the measurement was a bit of an after thought in the above scenario, a change in sales made it difficult (or impossible) to accurately measure the new program compared to the old program.  Had measurement been a larger part of the conversation while the program was being developed, it could have been suggested to only roll out the new loyalty program to a random 50% of customers on the existing loyalty program.  Sparring you the details of exactly how this is better, this would have allowed to get an answer to if the new program is better than the old, even in the wake of a global pandemic!

The purpose of measuring things is to have better information to make decisions on, which is why this is important!

Simplified Analytics Podcast Coming 3/9

While starting our company, we wanted to identify parallel ventures that could server as both a marketing and educational outlet.  Our blog posts are one of these avenues, while a podcast is our newest.

Both Kelsey and myself listen to podcasts from a variety of genres, so we wanted to take what we liked from our favorite podcasts and use that when creating our own.  The key difference between our podcast and others though is our focus.  When looking at the various podcasts available today, there aren’t any that focus on small businesses using data.

Our focus will be to interview small business owners and decisions makers, parallel entities to small businesses (Chamber of Commerce and government programs), and anyone else who has a tie to small businesses.  The interviews are conducted to assume minimal experience with data and analytics concepts.  When there are more technical explanations, you’ll see we stop the conversation to define and clarify the word or phrase. We want this to be an educational outlet for people who are interested in the small business space.

The target audience for this podcast is anyone in or aligned to supporting small businesses.  Using data is common practice at most large companies today and many of the most successful small businesses are finding ways to integrate data into their businesses.  Our goal with this podcast is to provide relevant resources and examples on the topic.

We’ll be dropping 4 initial episodes to gauge interest in the podcast concept.  Each episode will be coming from quite a different perspective on how small businesses can leverage data and analytics.

The first episode is with Mary Hodson, the Chamber of Commerce President in Hutchinson Minnesota.  We focus on using some examples from the businesses in her community and how they use data.

The second episode is with Cory Schmid, Employment & Training Specialist at the Minnesota Department of Employment and Economic Development (DEED).  We talk about how DEED uses data to support the initiatives they run, helping both small businesses and unserved communities in Minnesota.

The third episode is with Cameron Wonchoba, a Data Scientist at Verdigris and HoopSpots.  Our conversation focuses on the application HoopSpots.  This is an app that allows users to find pick up basketball games in their areas.

The fourth episode is with Sean Dotson, General Manager at Dancing Goat Distillery.  We discuss how a multiple family generational company like Dancing Goat Distillery has been able to evolve to start leveraging data to inform their decision making.

We enjoyed having these conversation and we hope you enjoy listening to them as well!

A brief trailer podcast is available now! You can either follow this link, or search “Simplifying Analytics for Small Businesses” on Spotify, Apple Podcasts, or Stitcher.

How to Know What Data to Capture?

In order to use data and analytics in your business, you need to have the right data.  But how do you identify what is the “right” data?  The one sentence answer is “start with the important questions or challenges your business is facing”.  In this post we’ll provide a general framework for thinking through what data to collect within your business.

1) Start w/ Business Questions

Whenever thinking about using data, you want to start with the business questions and/or challenges.  Data and analytics is most valuable when its helping address the most valuable questions and challenges in a business.  Even if you already are collecting a ton of data, trying to simply mine the data for insights is extremely time consuming and rarely yields valuable returns.

All businesses will have questions, concerns, and pain points.  Brainstorming a list of these items, without thinking through the context of data and analytics is best.  If you start by thinking about the data, you’ll be limiting the list to what can be done right away vs. what is most valuable to address!  Putting together this list and reviewing with others in the company will provide a variety of perspectives to ensure you’re not over focusing on your perspective.

2) Assess Current Data

Once you have your list of items you’d like to address, now you can start identifying if the data you’re capturing today can be used for your list of items.  A common mistake is to focus on perfection at this stage.  If you have data that answer half of the question, or data that will generally address an item on your list, make sure to note that!  You can now prioritize the list of items to determine which you’ll address with the data you already have.  These are the quick wins!

After this exercise, you’ll start seeing where some of your data gaps are.  Document these data gaps.

3) Address Data Gaps

Knowing your data gaps is a significant part of the data battle!  With the list of data gaps, start prioritizing these gaps.  Two important items to consider are the value closing the data gap would bring, as well as the effort required to capture the information.  Even if the data would be highly valuable to capture, if it is extremely expensive and difficult to capture, it may not be something worth addressing. 

Once you have prioritize the data gaps, you can begin updating process and implementing the necessary changes to collect this information.

In Conclusion

Each of these general steps will vary greatly between businesses.  Some businesses may already be capturing most of the data they need, however most likely won’t be.  Some businesses will have a lot of easy changes in processes to capture additional data, while others will have a difficult time and need to only prioritize a few updates.  In general, the goal for all businesses should be to use data to answer their most important business questions. 

Breaking up the process to start addressing what you can answer today is important.  It can take some time to update processes and collect enough data to do something with it.  So starting with what you can means you will start getting value out of your data sooner!

You May Already Have The Data You Need to Start Getting Value from Data and Analytics

Are you thinking about using data and analytics in your business, but not sure you’re capturing enough data?  You may be surprised how little data is required to get value out of your data!  A common objective we heard when starting Simplified Analytics is most businesses aren’t capturing the necessary data they need to get value out of investing in analytics.  While this is sometimes true, it isn’t always true.  There are a number of systems capturing data that you may not even be thinking about.  We’ll briefly outline a few of them below.

Point of Sale (POS)

If you are in retail, the point of sale (POS) system is likely where most things within your business run through.  Regardless of the specific system, your POS system is capturing and storing information that goes through it.  However, are you getting value out of this data?

Most POS systems will have some reporting capabilities, however these reporting capabilities often lack insights.  From the software developer’s perspective, they’re trying to create features that everyone purchasing their product can use.  The issue with this for you becomes the lack of insights specific to your business.  This means there is an opportunity to expand beyond the software’s reporting capabilities to get more value out of your data.

Whether you hire an employee, learn yourself, or hire an outside company, the opportunity to take the data from your POS system and turn it into actionable insights is immense.  Most POS systems will allow you to extract the majority of your data from it.  This can be an overwhelming amount of data.  However, you don’t necessarily need all of this data.  It is best to start with some of the challenges or opportunities within your business then determine what data from your POS system can help answer this question.

POS systems can have a ton of features, so you may not currently be utilizing all of its capabilities.  As you dig further into your data and are able to start answering questions, other questions may pop up that you want additional data for.  While you may not have the data you need to answer all your questions today, you can then update your internal processes to start collecting the additional information, allowing you to answer the questions in the future.

Customer Relationship Management (CRM)

If your business is heavy on the sales and account management, its likely your business has some type of customer relationship management (CRM) system.  While you may not think this is an opportunity to use data and analytics, it actually has a number of opportunities to! 

If you’re tracking sales success within the CRM, this is likely an area the software has built in metrics for (this is also probably more apparent where data can be used).  However, there is also another category of analytics that CRM data can be used get value from: Relational/Network Analytics.  There are techniques that can identify communication and network patterns, allowing you to understand the patterns in a more digestible way.  This can then be used to identify more successful habits and communication patterns to train your sales force on.

Same as with the POS systems, as you start diving into the data and answering questions you will likely find additional information you’d like to capture.

Website/Google Analytics

A popular analytics with small businesses right now is Google Analytics, which is used to track activity on your website.  There are a number of reporting options available within Google Analytics.  However as with any data, leveraging the data is an important step to getting value out of the data.

Bringing It Together (Literally)

There are 3 common systems used within businesses that capture data.  Once you’ve made the effort to understand the data, an extremely valuable and powerful next step is to find ways to connect the various data sources within your business.  When you’re able to do this, its likely you’ll begin to find patterns you had no clue existed.

Why Should I Use Analytics if My Business is Already Successful?

Most small businesses end up failing, but yours hasn’t.  So why would you consider changing how you’ve done things and introduce something new like analytics into your already successful business?  Fair question! 

In the graph below you can see that even between 10 and 15 years of being open, 10% of business end up failing.  So even more mature new businesses are still at risk of failing.

BLS Data from Investopedia Article

As you’re already well aware, the key to continuing a thriving business is evolving and changing.  Social media is a great example.  20 years ago, social media didn’t exist.  But today its now used by the vast majority of businesses.  Analytics is a tool to help your business in 3 major ways.

  1. Increase speed of decision making
  2. Increase quality of decision making
  3. Scale/streamline processes

Increase Speed of Decision Making

Things in business move at 100 mph.  If you’re the owner or a major decision maker in a business, you’re often asked to make decisions as quickly as possible to capitalize on an opportunity.  What if there was a tool to help you make those decisions even faster?  There is and it’s called analytics.

A skeptic may say something along the lines of “won’t adding something to the decision-making process slow it down?”.  Fair point, however when correctly implemented, analytics will speed up your decision making.

The main way analytics allows for increased speed of decision making is by taking a lot of information and digesting it.  What if to make a decision, you need to understand current sales, future forecasted sales, previous ordering history, and anticipated shifts in purchasing behavior?  By capturing the data for this and applying analytics to it, you’ll end up being able to make a quicker decision than you otherwise would have.  The speed of decision making comes with a major caveat though.

The speed of decision making is most relevant with common and repeated decision making.  If you’re having to make the same type of decision on a recurring basis (sales expectations, inventory purchasing, marketing spend, etc), then you’re able to set up an analytics process to “automate” this decision making process. 

I use the term “automate” loosely here, as its usually still recommended for you to put eyes on a decision vs. just letting the analytics be the sole decision maker.  By leveraging a hybrid approach, you’re able to reduce the time required to think about highly complex situations, while also keeping your insights and knowledge of a situation included in the decision-making process.

Increase Quality of Decision Making

This is the most universally beneficial quality of analytics, regardless of your type of business.  Analytics adds another perspective that is often times more objective and able to handle more complex situations better than humans can.  To be clear, the goal isn’t to replace you or others in making decisions, its to combine them to generate a higher quality decision.  We have an another post on this topic if you’re interested in going really deep into the topic: 1+1 = 3, The Value of Combining Business Knowledge with Analytics in Decision Making.

When using analytics to improve decision making, I like to view it as an extra employee that you pay little money for, never needs a break, and continues to learn new information at an incredibly fast rate.  The cost and benefit trade-off for the impact of analytics is often easy, particularly when dealing with decisions close to the financials of a business.  Say you’re working on analytics specific to your marketing spend.  Your annual budget is $5000 and you don’t want to increase the budget, but want to get more sales out of your marketing efforts.  You can evaluate your previous marketing efforts to understand where you’ve had the most success, allowing you to put an emphasis on those successful marketing channels over others that have been less successful.  When making these changes you’ll often have an idea of what impact these changes will have before even making them.  If you haven’t addressed this type of optimizing in the past, you may be surprised with the extra value you can get without changing your budget.

Scale/Streamline Processes

Many analytics tools and techniques have a secondary benefit that can end up being highly valuable for some businesses.  Leveraging code can be a very powerful tool to remove manual and repetitive tasks, allowing your employees to spend time doing other tasks within the business.  Copying information from one format to another is a good example of how code can be leverage.  If an employee is taking information from a PDF and putting it into a spreadsheet, this type of task can be streamlined to be done much quicker, reducing the amount of time an employee needs to spend on it.

Final Thoughts

Even if you’ve be running a successful business, leveraging analytics can still be a powerful tool to build upon and continue your success.  The value a business can get from analytics will vary from company to company.  Regardless of the industry or type of business you run, decision are made, so increasing the speed and quality of decision making is valuable.  Additionally, most businesses have processes that could be streamlined, allowing employees to focus on other tasks within your business.

Interested in talking more about how analytics could help your business?  Reach out to me at jason@simplifiedanalyticsconsulting.com and we can have a conversation about where analytics may be able to add value to your business!

Should I Use Analytics? The Costs and Benefits of Analytics

If you’re a business owner and haven’t yet started using analytics, a natural question will be: How do I know if this is something I should pursue?  Here I’ll provide you some thoughts on how to evaluate the costs and benefits of investing in analytics.  While the specific monetary amounts will vary, I think you’ll find this general enough to give any business a starting point for weighing the costs and benefits.

Costs

The cost of analytics can vary significantly, however there are a few different places these costs can come from when you’re getting started with analytics.

1: Free (But Not Really Free)

The first option is to upskill yourself or one of your existing employees.  While this can be viewed as a “free” option, there are costs associated with this.  This cost mainly comes from the time to do this.  If you or an existing employee already has some background in data and analytics, the cost of this may be low.  However, if you’re starting from scratch, the cost to upskill on analytics can be incredibly time intensive.

If you’re the owner of the business and already have an analytics background, its unlikely you haven’t already taken steps to integrate analytics into your company.  Its more likely that this situation comes up when you have an existing employee who also has an interest/background in data and analytics.  When you have this type of employee, it can be a cost effective and also timely solution.  The majority of leveraging analytics is understanding the business questions and challenges, so an existing employee with an analytics skillset will likely be able to make quick progress on where analytics can be leveraged first.

However, if you don’t have any existing employees who are into data and analytics, the cost of upskilling them will take a long time.  This will also take them away from their “day job”, which may have impacts to the business.

2: External Help

Hiring an employee to do analytics can be a significant cost commitment.  It can also be a risky one when you’re unsure how/if you’re going to see value from analytics.  The next cost option is hiring external help from a consulting company (like Simplified Analytics😊).  Using an external company provides you immediate access to analytics expertise with a fixed and short-term cost commitment.  Especially in the early stages of proving value with analytics, this approach reduces risk by limiting your costs to a specific engagement.  If you begin seeing the value of analytics and continue investing, you will likely see this benefit until your analytics work reaches a point where it becomes a full-time job that could be hired into your company.  Hiring an external company is likely cheaper until this threshold is reached because you’re not either paying an employee a full salary when there isn’t a full analytics workload or you’re not trying to hire for someone who knows both analytics AND how to do something else within your business.

The major negative to this approach is hiring an external company means there will be some learning curve for the consultants to understand the specifics of your business.  This will take some time, but much less than upskilling an existing employee with no analytics experience.  As mentioned above, another negative can be the long-term costs of using an external company.  If your analytics work reaches a full-time job workload, then its likely more cost effective to hire a full-time employee with a data and analytics skillset.

This is the most flexible option from a cost side and reduces long term commitments should analytics not prove valuable to your business.

3: Full-Time Employee

This is mentioned in the “External Help” section above, but will touch on this in a bit more detail.  As mentioned above, it makes the most sense to hire a data and analytics person once you have enough analytics work for a full-time job. 

My rationale for waiting to hire a full-time analytics person vs. hire someone who does analytics and something else in the business is related to the work and the typical analytics person.  Analytics work requires a disproportionate amount of heads down time compared to many other functions in a company.  Additionally, the typically analytics person isn’t going to be interested in working in other parts of the business (like customer service).  While it is possible to find a situation within your business that you could make work, it will make hiring for the role much more challenging.

This is the least flexible option from a cost perspective, as you’re committing to a full-time employee.  However, once you have enough analytics work, this ends up being a more cost-effective option than relying exclusively on an external company.

4: Blended

Once your business has enough analytics work for a full-time employee, does that mean you don’t need an external company to assist?  It really depends on your goals and budget for analytics within your business.

The value of having both a full-time employee an external support resides within specialization.  There are many different specializations within the analytics space.  The full-time employee you hired for analytics may have a specific specialization in which they’re an expert at (and you hired them because that’s the specialization you need most).  However, if another project comes about that requires a different specialization, there can be value in leaning on an external analytics company for that specialization.

Once you have enough analytics work for a full-time employee, this ends up being a flexible and cost-effective option.  You’ve hired an analytics person full-time to do what you know is most common.  You’re then able to leverage the flexibility of an external company for additional expertise when needed.

Cost Summary

In summary, if you have an existing employee who is interested and knowledgeable with data and analytics, start with them.  That’ll likely be a cost effective and timely solution.  If you don’t have that knowledge with an existing employee, hire an external company to start testing out analytics.  Once you have proven out the value of analytics and have enough for a full-time job, hire a full-time employee to run your analytics.  If your analytics investment continues to grow, leverage a blend of your full-time employee for your “common” analytics needs and an external company for new or less common analytics work outside your employee’s expertise.

Benefits

Analytics ends up being a nice service to sell because its benefits can be both immediate and long-term.  The benefits for each company will vary drastically, so you’ll find this section quite generalized in an effort to be relevant across businesses.

1: Short-Term

The consultative approach we use at Simplified Analytics highlights the short term benefits you could expect from leveraging analytics.  This focuses on the specific situation for the analytics work.  If the goal is to optimize the marketing budget of your business, then we’ll aim to measure the increase in sales.  If the goal is to reduce costs within the business, then we’ll measure the decrease in costs.  These immediate benefits are measurable as long as there is forethought and intention around what to measure and how to measure it. 

When analytics is measured in this way, its easy to weigh the benefits and costs.  If the cost of hiring an external analytics company is $1000, but the expected increase in sales is $2000, that’s a 100% return on investment (and hopefully an easy decision)!

2: Long-Term

The benefits of analytics on a business in the long term and much more abstract.  In general, a culture that values analytics makes more informed and better decisions on a daily basis.  However, how do you measure the value of all these decisions?  The time and effort of doing so would be a significant expense by itself!

Using conservative estimates in these situations are often the best way of quantifying the potential value in the long run.  If your business does $300,000 is sales each year, can you expect a more data driven culture to increase sales by 1%?  If that is the case, then that’s only $3000 of increased sales each year.  However, what if you can expect data driven decision to increase sales by 10%?  The expected return of $30,000 seems a bit more attractive.

Using the long-term benefits to justify an analytics investment is much more challenging, however should be part of the consideration.

Final Thoughts

If you just focus on the cost of analytics, it can look to be an unnecessary expense.  However, if you bring together the cost and benefits of analytics, you can start seeing where sizable expected returns can come from integrating analytics into your business. 

The largest and most successful businesses in the world place a high value on testing and learning, so I hope the thoughts on how to minimize your costs based where you’re at in your analytics journey allows you to feel more comfortable with experimenting with analytics to get to the short and long term benefits of analytics within your business.

1 + 1 = 3, The Value of Combining Business Knowledge with Analytics in Decision Making

There are certain combinations in the business world that are greater than the sum of their parts.  Those familiar with the history of Apple, know without both Steve Jobs and Steve Wozniak the company wouldn’t have been successful.  Steve Jobs lacked a deep technical skillset to build software or hardware, but was an incredible marketer and salesperson.  Steve Wozniak had an incredible technical skillset, however fit into the standard stereotype for computer programmers, being introverted and unable to speak to a “normal” person about their work.  Without each other, its unlikely Apple would have become what it did today.  Their skillsets complemented each other, creating a combination that was greater than the sum of its parts. 

Business knowledge and analytics share a similar relationship.  By themselves, each have strengths and weaknesses.  When brought together, they’re complementary, allowing for better results than each by themselves.

Pros and Cons of Business Knowledge and Analytics In Decision Making

Let’s discuss the pros and cons of both business knowledge and analytics in the context of decision making within a business.  This will provide a foundation to gain insights into how the combination of the two creates a sum greater than its parts.  For the sake of wordiness, I’ll introduce the term BKA, short for “business knowledge and analytics”.

Three dimensions will be used for each evaluating BKA individually.

Objectiveness: How susceptible to bias is the approach?  If two individuals independently are given a question, how likely is this approach to result in both people giving a similar answer?

Speed: How quickly can the decision be made with this approach?

Scalability: As there are more an more decision makers in a company, how well can this approach be implemented across the company?

Business Knowledge

ProsCons
Objectiveness-Small/simple decisions
-Where data isn’t collected
-Varies significantly between decision makers
-Past experiences can influence a decision
-Can vary from day to day for the same person
Speed-Doesn’t require data collection or time to run an analysis
Scalability-Small/simple decisions can be easily implemented via company norms/culture-Large/complex decisions don’t scale
-Requires preparation to transfer knowledge

Analytics

ProsCons
Objectiveness-Highly consistent from day to day
-Highly consistent from person to person
-Some degree of subjectivity with decisions on approach
Speed-Repeatable decisions where data is collected and process already created-New decisions where data hasn’t been collected or a process already created
Scalability-Code based approaches are easily repeatable and scaled-If version control not properly managed, can created inconsistencies

Combined

The above gives you an idea of what the pros and cons for BKA when used independently.  In this more detailed format it may be more challenging to see, but the pros and cons of both approaches are compliments.  To help make this easy to see, the below graph provides a summarized perspective of the above pro and con tables.

The combination of BKA provides you the best of both worlds!  If a decision requires speed, go with a heavier approach of leaning on your business knowledge.  However, if a decision is highly complex, go with a heavier analytics approach, as this will provide a more objective viewpoint that can be scaled across your company (if need be).

Decision Making Framework

Popularized by Jeff Bezos at Amazon, I’ve found this approach to decision making the most useful.  When a decision is reversible, make it fast.  This implies only using easy and quick to gather data vs. conducting a large and complex analysis to collect more information.  Since the decision is reversible, the value of speed often outweighs the risk of being “wrong”.  Thinking about this in the context of the pros and cons above, this means using your knowledge of the business is likely the primary source of information for this decision.  This can often be the majority of decisions made at a business each day.  Not all decisions should require data and analytics, as this would slow decision making considerably. 

When a decision is not reversible (or extremely expensive to do so), there should be a thoughtful effort on which information is necessary to make the decision.  If you’re a restaurant looking for a building, this should be a thoughtful process.  Identifying the relevant decision-making factors (foot traffic, space, layout, etc) and then weighing the various options will result in a much better decision than picking the first building you see!

What to Expect When Combining Business Knowledge and Analytics

The above sections make the points about BKA being compliments to each other in the decision-making process.  However, what are some of the outcomes you’d expect to see if today you’re primarily using business knowledge, but made the decision to start using analytics?

1: Confirm Your Existing Perspective

Sometimes your existing knowledge of a situation and analytics end up giving the same answer.  This doesn’t mean implementing analytics was a waste of time or money!  You now have two different approaches that led you to the same answer, meaning your confidence in that answer can be greater than each independently.  Everyone is susceptible to biases, so the value of “double checking” important decisions can be significant.

2: Finding New Insights

Using analytics can help you find new insights and knowledge you may not have been aware of before.  This is most likely in new areas you haven’t explored in the past, but can also occur in spaces you’re already familiar with. 

In new areas, finding new insights with analytics should be relatively straightforward.  If you have no base knowledge of an area, using data to better understand it is likely to generate novel findings.  However, when exploring areas you’re already familiar with, it can lead to uncomfortable questions: Why wasn’t I already aware of this? How could we have missed this?  This is part of the power of analytics (and value in combining with your business knowledge), as it provides a different approach and perspective that may find something new with fresh approach.

3: Finding Insights Counter to Your Perspective

While rare, sometimes analytics can help find a situation in which your existing perspective is simply incorrect.  These can be the most challenging situations for a business owner or decision maker, especially for a business new to analytics.  The natural first question is to challenge the analytics, which is a reasonable first step.  If you and your business have a long held belief and a new analytics approach comes in and tells you something opposite to that belief, we should make sure the new analytics approach is being done correctly.  However, if the analytics approach is done correctly, the next steps can test the culture of a business in a way it may not have been tested in the past.

4: Consistency Across Decisions

As you further integrate analytics into your business, you should expect to see a higher degree of consistency across decisions and decision makers.  While there will always be room for interpretation and disagreement, analytics often shrinks that window.  You should find more productive disagreements since the scope of potential disagreement is much smaller when a business adopts the findings from analytics as the “truth”.

Final Thoughts

If you’re a business owner and less familiar with analytics, I hope these thoughts provide you some things to think about within your business’s decision-making processes.  The objectives of using analytics aren’t to replace your knowledge of your business, they’re to compliment and enhance your knowledge of your business.  Knowledge is power, and analytics aims to increase your knowledge.

If you’re interested in how analytics can help your business, reach out to Simplified Analytics for a free consultation about how analytics may be able to help your business.

Analytics Doesn’t Need to Be Complicated

I’ve observed data and analytics often times get a reputation for being extremely complicated.  Individuals who have made careers in the data and analytics space are often referred to as some name associated with magic: wizards, magicians, or miracle workers.  While data and analytics can be VERY complicated (Tesla’s software for autonomous driving being an example), there are also many ways to leverage data and analytics in your business that are incredibly simple. 

If you’re a business owner who hasn’t started leveraging analytics and are a little intimidated by the complexity of analytics, my hope is to help you feel more comfortable with the idea of starting simple.  There is still value to be gotten out of analytics with these simple approaches.  Once you’ve gotten the simple approach down, then you can build upon that to gain more and more value from analytics.

Analytics Maturity Model

A commonly used framework to explain analytics maturity is the analytics maturity model.  There are a number of variations of the below graph, but I prefer this one from Gartner as its simple and includes the questions trying to be answered at each maturity stage.

Gartner (2012)

To illustrate what this maturity model looks like in the real world, I’ll use this from the perspective of a local restaurant.

Stage 1: Descriptive Analytics

Stages 2, 3, and 4 are all based on how well you can do stage 1.  You need to be able to know what has happened in the past to understand why it happened, and then predict what will happen.  This makes descriptive analytics very important, as it sets the foundation for what we usually view as the more complicated (and valuable) analytics.

Descriptive analytics is often referred to as reporting.  It tells you what your sales and costs were over the last 12 months, how many customers you had, and what your website traffic was.  Understanding the past puts you in a better position to advance down the analytics maturity path, both from a knowledge and understanding perspective, but also from a data perspective.  The same data you use for descriptive analytics is used for the more advanced analytics stages.

For our restaurant example, this includes labor costs, inventory, rent, food cost and sales, etc.  This data is then put together to show the restaurants profits.  By putting it into the below graph, we can see most months they’re making a profit. 

This is great information, but what can you do with it?  By simply showing this type of data without any context provides essentially no value outside of:

  • In March the restaurant lost money
  • In July and September the restaurant had a minimal profit
  • In January, May, October, November, and December the restaurant made a large profit

If you stop with descriptive analytics, the next questions are likely to be “why should I care?” or “so what?”.

Stage 2: Diagnostic Analytics

Some business owners may view analytics as a replacement for the intimate knowledge they have of the inner workings of their business.  This couldn’t be further from the truth. The diagnostic analytics phase is a great of example of how powerful bringing analytics and contextual knowledge together is.

The diagnostic analytics phase is answering the question “why did this happen?”, which is often the step in the analytics journey you start seeing real value and return on your investment.  If you’re collecting a lot of data in your business, you can look further into the data to start trying to find the why.  However, this type of data collection isn’t always done, so applying your contextual knowledge can be a great (and sometimes better) substitute.

The owner of the restaurant suspects the presence of college students in town may be a significant reason for some of the changes in profit.  The owner gets some estimates from others in the community for how many college students are in town for that month.  When you put that profit and % of college students in town together on the same graph, you can see the profits and college students being in town generally follow each other from month to month.  Its safe to conclude the changes in college students coming and going throughout the year plays a significant role in the profits for the restaurant.

With this insight on the relationship between profit and college students being in town, now we have something that is actionable and can be used to improve profits!  Can we change our advertising focus when college students leave?  Should we reduce our costs during these slower months?  Should we run specials targeted at the non-student population?  We’re able to get to this insight without any complicated statistics, models, or autonomous driving AI.

Stages 3 and 4: Predictive and Prescriptive Analytics

Since the goal of this post is to focus on how analytics can be simple, I’ll be brief about predictive and prescriptive analytics.  These by nature are more complex, however if you’ve set the appropriate foundation with descriptive and diagnostic analytics, you’ll find these to be less complicated to understand.

The goal of predictive analytics is straightforward, being able to predict the future with some degree of certainty.  Prescriptive analytics however can be a bit harder to understand.  Think of prescriptive analytics as what diagnostic analytics is to descriptive analytics: moving from “cool information, but what do I do” to “here is why and what we can do to change it”.  Predictive analytics will tell you what to expect in the future, while prescriptive analytics will tell you what is driving that prediction and then how to influence it.

Using the restaurant example, you can predict sales for the next year and get a number.  What happens if you don’t like that number and want to know how to increase sales?  Prescriptive analytics uses the predictive model and tells you what is driving that the predictions.  This should sound similar to what diagnostic analytics is (because it is).  The key difference is diagnostic analytics is a lot simpler and faster to get started than prescriptive analytics.

Final Words

In short, analytics can be relatively easy and fast to start getting value out of it.  There is a general model you can reference for analytics maturity to understand the difference levels of analytics.  While the model isn’t perfect, as each business varies from the next, it provides a  general guideline for analytics.  If you’re pursuing the analytics journey, expect the initial stages to feel less and less complicated as you continue to use them.  This then opens the door for more advanced analytics (that hopefully don’t feel as advanced as they do today)!

Interested in pursuing analytics in your business?  Fill out our contact form and we’ll set up some time to discuss how Simplified Analytics may be able to help!

Customer Segmentation and Marketing Campaign Analysis

We’re firm believers in using examples to help explain new topics to people. As we continue to build our customer base, we will continue to put out examples like this to provide potential customers examples of how analytics can help increase sales and decrease costs by enabling better decision making. This examples is using real data from a grocery store in Brazil that was made available via Kaggle.

The Business Questions

The grocery store has two questions they’re looking to have answered.

  1. How have our previous marketing campaigns performed?
  2. How can we simplify our marketing efforts to be more targeted by providing relevant campaigns and deals our customers will be more likely to act upon?

Their objective in answering these questions is to increase the impact of their marketing budget without needing to increase the budget allocated to marketing.

The Data

This grocery store collects a lot of data via their customer loyalty program. While more data is usually better, by no means do you need to be collecting all of this information to get value from this type of analysis. This means they have some basic demographic information about their customers.

  • Age
  • Marital Status
  • Education
  • Number of kids or teens

The loyalty program also enables the grocery store to track purchasing activities across their website, catalog, and in store. The information they have tracked includes:

  • Purchase amounts for various categories (Wine, Fruit, Fish, Meat, and Jewelry)
  • Where they’re purchasing the products (Website, In-Store, or Catalog)
  • Website visits

Lastly, they’ve run their marketing campaigns through their customer loyalty program, meaning they’re able to identify which customers use which coupons from the campaign.

The Analysis

To answer both of these questions, a customer segmentation analysis will be done to provide a view point from the perspective of customer profiles.

Customer Segmentation

Executing an analysis to segment your customers into specific groups is actually quite simple. However, there are a number of easy to forget/miss steps that can give you an inaccurate result. Customer segmentation is usually done with a “clustering” algorithm. There are a number of specific clustering algorithms, however they generally work by measuring the distance between data points. This means the scale of your data needs to be consistent between data elements. For example, if you’re comparing the annual income of a customer to the amount of fruit they purchase each year, their income will be much higher than the amount of fruit they purchase. Without adjusting our data to ensure having a high income and purchasing a lot of fruit, our customer segment results will not be accurate.

Where the major work comes in with customers segments is digesting what each of the segments mean. It isn’t helpful to call a customer segment “Customer Segment 1”, its common practice to evaluate the group and create a “profile” for what the typical person in that profile is like.

Customer Segmentation Results

The output from the algorithm identified 4 customer segment groups. The existing data is then split into the 4 segments and analyzed for trends or unique traits about those customers.

By putting profiles and stories to each of these groups, the grocery store can then be more targeted in their marketing efforts to each of the segments. This part of the process can be time consuming, but also fun to flexible some creative muscles as well!

Group 1: Channel Agnostic Frequent Shoppers

This group spends the most amount of money on average across all the product categories compared to the other segments. They’ll shop in store, online, or via the catalog. They enjoy options for purchasing from the grocery store. This makes them a highly desirable customer from a profit standpoint. They’re also the group that capitalizes on promotional sales the most. They’re overall a highly educated group.

Group 2: Large Family

This group is primarily distinguished by the size of their family. They’re the most likely to have kids and teenagers. Given the size of their family, they generally don’t purchase much from this grocery store, suggesting they’re looking around at potentially other discount grocery stores. We conclude this by their frequent visits to the stores website, but lack of spending money.

Group 3: Highly Educated Couples

This group is the most likely to have an advanced degree (Master’s or PhD) and most likely to be married. They’re the least likely to have teenagers, suggesting if they have children, they’re younger in age. They frequently browse the company website, but don’t spend a lot of money with the grocery store.

Group 4: In Store Frequent Shoppers

Similar to first group, these customers spend a lot of money with the grocery store and also leverages promotional sales as well. However, this group specifically shops in the store vs. online or via the catalog. While profitable, this indicates digital or catalog advertisements won’t make an impact on this group.

Previous Marketing Campaign Results

The other question from the grocery store is how their previous marketing campaigns have performed. We can use the newly created customer segments to understand the performance of the marketing campaigns.

The grocery store has run 6 marketing campaigns in the last 2 years. Overall, they see about a wide range of customer engagement with the campaigns.

Campaigns 1, 3, 4, and 5 were generally successful with customers in all segments engaging with relative consistency between 5-7%. The campaign that performed the worst was 2. This saw engagement rates of around 1%. The most successful campaign was the 6, which saw engagement rates between 7-24%. Within campaign 6, the Highly Educated Couples segment was most likely to engage with the campaign.

The grocery store can learn from these results by not spending time and money on campaigns like 2 in the future. They can also learn that campaign 6 was by far the most successful, particularly with the Highly Educated Couple segment.

The Conclusion

By doing an analysis like this, the grocery store can have better insights into their customers and which marketing messages will be more effective with certain groups rather than a “mass marketing” approach by giving all customers the same marketing touchpoints. While some of the insights may not be novel, some are likely to be new. At the very least, businesses can confirm their suspicions and thoughts, increasing their confidence when making decisions.

Is this type of analysis something you think would be interesting for your business? If it is, reach out to us to get a free consultation scheduled!