Customer Journey Maps Must Come Before Transactional Customer Feedback

cjm-partial

Transactional customer feedback is a very effective way of improving business performance. With it you can diagnose problems and processes that are driving customers away and reinforce the drivers of customer loyalty in the business.

But how do you decide which transactions should be included? If you start the design of your customer feedback process top down with a Customer Journey Map, you will make better design decisions.

What is a Customer Journey Map?

Customer Journey Maps document the interactions that customers have with your business before, during and after their relationship with you.

With so many practitioners using them, there are many different forms of customer journey maps; none of them right or wrong, just used in different ways.

Some are circular, some include the Awareness phase of the marketing cycle, and others include estimates of the customer’s emotional state at each point in the process. All however, attempt to understand and document the customer’s experience with the organisation.

My preferred approach is to map out the trackable elements of the customer experience and include some internal processes. This allows clients to easily apply the customer journey map information to the design of the transactional customer feedback data collection process.

cjm-partial

As you can see above, we split the map into two sections: Customer initiated and Company initiated. We also include some internal processes that the client does not see so that they can be picked up in the next stage.

Why Use Customer Journey in Customer Feedback?

One of the key benefits of transactional customer feedback is gathering information around each of the key touchpoints in the business. If done in a consistent manner you can then apply Pareto analysis to the feedback on each touchpoint to determine where to invest your time for maximum benefit.

Merging Journey Maps and Customer Feedback

In complex businesses you should not attempt to translate the entire touchpoint map into active transactional customer feedback data collection on day one. This is simply because some data will be harder to access than others. You are better off prioritising your data collection to focus on the more important, easy to collect data first and then build out to the complete map over time.

If you try to start with 100% coverage of your customer journey map you may never get the process up and running. At the very least you will delay the launch beyond when you could be acting to improve the customer experience.

When you have constructed your journey map you should analyse it to prioritise touchpoint data collection using the following four factors:

1. What Data Do We Have Access to Today?

First determine what data is actually being collected in a usable manner. For each touchpoint identify which system (manual or automated) records the customer’s transition across that Touchpoint. Then determine if you can extract that data in a useful way.

Examples of a transition across a touch point include:

  1. Account manager switches the prospect status to “sold” or “lost” in the CRM system. This is relatively easy to extract.
  2. Shipping system issues a consignment note. This would also be a very easy transaction to capture.
  3. Contact centre representative leaves a text note in the customer record about a call from a customer. This would be very difficult to extract and use because it is a free format text.

You can see that even though the organisation might record the transition across the touchpoint the data might not be easy to use. Focus initially on data that is easy to extract and use.

2. Which are the High Value Interactions or Moments of Truth

In any customer journey there will be more important touchpoints (e.g. customer places order) and less important touchpoints (e.g. customer downloads user manual).

Try to focus initially on the higher value touchpoints for your business. This may be difficult to determine empirically because you will have no initial data but use the knowledge and skills in your organisation to make an educated guess.

3. Identify the High Volume Transactions

Initially you should also look for high volume transactions. In a transactional customer feedback environment this will provide more reliable data more quickly and allow you to act sooner.

Also, improvements in high volume transactions will often have a larger combined impact on the overall business than changes in low volume transactions.

4. Capture High Value Customer Segments

Focusing on important or large customer segments is like focusing on the high volume transactions. It will naturally focus you on the important parts of the business.

Update the Customer Journey Map with Targeted Touchpoints

At the end of this process you should have a customer journey map that shows the prioritised data feeds that you want to drive into your customer feedback process.

Now the task is relatively straight forward. Simply work with your data analytics or IT team to methodically get each data feed up and piped into your transactional customer feedback system.

They don’t all have to be completed on day one. Just lay out a plan to bring them all into line at some point.

Using the customer journey map to plan out your transactional customer feedback execution ensures that you start collecting the right data about the right touch points in the right order. That will help you to drive the maximum impact from the continuous improvement process.

It’s Finally Here: Customer Feedback for Small Business

RunOurSurvey - Small Business Customer Feedback

RunOurSurvey - Small Business Customer FeedbackWe have been helping our clients to collect and use customer feedback for more than 12 years. We have worked hard to help organisations to better understand and act on feedback from their customers. And, in that time, we’ve seen some really rewarding successes.

But for us there was one big problem: what we did was never really accessible to small businesses.

As a boutique consultancy ourselves this really bothered us. Through a combination of cost, time needed and the process; the work we have done just has not been available to small businesses.

So we decided to change that.

Today we are launching RunOurSurvey – A site and support system dedicated to helping small business to collect and use customer feedback.

The site’s mantra is simple: Customer Feedback:

  • At a small business price;
  • In a way that small businesses can use effectively without becoming customer feedback or statistics experts

Please head over and have a look as we have some great content and tools ready to go:

We are really excited to be part of the solution not the problem here.

As I say please go ahead and have a look. Then let me know what you think – we are keen to help build a useful and accessible support system for small businesses.

Practical Statistics: How to Test if Your Customer Feedback Score Really Changed [Excel]

t-test-2

DoctorIn customer feedback we often run the same survey to different sets of respondents and we are very interested in identifying whether the responses are different between different groups. Typically those groups are either different sub-segments (male vs female customers) or different time periods (last quarter vs this quarter).

However, too often the different/not-different determination is done overly simplistically using a simple average of the feedback scores. If the sample average is different then we assume the group average is different. If it’s the same we assume the group average is the same.

While simple is often good, in this case the risk of making a false positive or false negative error can be high. The cost of that error can be substantial: changes that are actually working can be discontinued (false negative) or initiatives that are not working continued (false positive).

So while it is tempting to just look at whether the average score has changed we need to do a little more work to validate that decision.

Enter Student’s t-Test (Independent)

Nothing to do with education, Student’s t-test was named for its inventor’s pen name and is a relatively straightforward test you can perform on data to determine if there is a difference between two sets of customer feedback responses.

Assumptions

As you might expect there are quite a few assumptions when using this test but they are assumptions that are usually valid when applied to customer feedback so you should be fine:

  1. Normal Distribution: The data needs to be normally distributed. This is usually the case in customer feedback data.
  2. Homogeneity of variance: The variance between the two samples needs to be the same. Again this is commonly the case for feedback data in the situations we will look at below.
  3. Data is independent: For our purposes we will be using an independent t-test and that means that the samples should be independent of each other. For instance, you should not survey the same people in each sample. Again we are normally fine on this assumption for sample based time period surveys as the chance of exactly the same people responding to each survey is low.
  4. Random samples: The process of taking the sample should be exactly random, i.e.the chance of receiving a response from every person should be exactly the same. Now we know in practice that there is some skewing of the response curve because very happy and very unhappy clients are more inclined to respond than those in the centre. For this assumption we probably just have to accept that the sample is not truly random.
  5. Data must be continuous: In general the data we deal with in customer feedback is Ordinal data:

“How responsive is Company X in closing the loop after problem resolution? Where 1 is very unresponsive and 5 is very responsive”

The response is one of 5 numbers. Now technically you should not use the t-test for this type of data but there is quite a lot of statistical discussion that indicates it’s perfectly fine to do so.

What Can You Test?

The t-test test compares the two samples and provides the probability that they have the same mean.

Commonly you would test for changes in the overall score (“would recommend” question, customer satisfaction, customer effort score, etc.) and the scores for the different attributes in your customer feedback questions: responsiveness, accuracy, professionalism, etc. The questions are typically on a scale with a number as a response.

Note that this test does not work for Net Promoter Score: see this article for more information on how to test for changes in NPS.

As noted before, the test can be used for changes over time (monthly numbers) or between groups (contact centre team A and contact centre team B).

Using Excel

For the detailed maths you can check out Wikipedia but here I am going to simply show you how to do the calculation in Excel as this is what most people will use.

Step 1: Get Your Data and Run the Test

Below we have some sample data that shows the response scores for 3 months along with their average.

Looking at the simple average some people would be inclined to say that the score rose each month 3.4 -> 3.6 -> 4.7. Maybe, maybe not. Before we make any big decisions based on that information let’s test it and find out.

t-test-2

First we’ll compare May and June to see if there was change in the score.

From the Data menu in Excel, select Data analysis and you will be presented with the selection box. Scroll down to “t-Test: Two Sample Assuming Equal Variances”.

t-test-1

t-test-4

Enter the data for the first and second range of responses. These are your two datasets and in our case are May and June but for you could be Team A and Team B, etc.

Set the hypothesised mean difference to 0 (zero) (we are testing for no difference) and put the output range to where ever you would like. Hit Okay and Excel does the work.

t-test-3

Step 2: Interpret the Result

This is the result of the May to June test. As you can see there are lot of data provided in the results. The trick is to find the information that you really care about and that’s right at the bottom: “P(T<=t) two-tail”

t-test-5

This value is the probability that the two populations have the same mean and that the difference in the sample means (3.4 to 3.6) is simply statistical noise. This probability is 18.8% for our example.

In statistics we generally use 5% probability as the decision point (don’t ask why). So based on this test, because the probability that they are the same is 18.8% (i.e. greater than 5%) we would accept that they are the same.

Aside: This is termed accepting the null hypothesis.

Now let’s look at the June and July data.

t-test-6

Here we can see that the probability that the two populations are the same is 0.00000000135%. As that is less than 5% (by a long way) we will decide that there really has been a change in the average for the underlying population.

Aside: This is termed rejecting the null hypothesis.

Handy Dandy Interpretation Summary for t-Test in Excel

The handy dandy summary card for this test:

  • If “P(T<=t) two-tail” < 5% then a change has probably occurred in the overall population
  • “If P(T<=t) two-tail” > 5% then a change has probably NOT occurred in the overall population

Conclusion

Too often when reviewing customer feedback data and looking at charts people see changes in the sample average score and then interpret that the overall response of all customers has changed. This causes them to make poor decisions and spend hours trying to find the “reason” that the change occurred.

Don’t make that mistake.

A 30 second t-Test can validate if change has probably occurred and allow you to confidently state: “yes it has increased”, or, “I know it looks like it’s increased up but it’s not a significant change and we should ignore it.”

By Adam Ramshaw BusinessLeadersGuideCTA

The Only Statistical Analyses You Need to Use On Customer Feedback Data

bigstock-Single-red-apple-floating-abov-38968036

bigstock-Single-red-apple-floating-abov-38968036Statistical analysis is a big, complex and fascinating area of study. Okay, okay maybe it’s not fascinating for everyone and I can already hear a few yawns at the back of the room. But the good news is that if you are analyzing customer survey feedback there are just a few key tools that you need to master.

Get those sorted and you’ll be way ahead of your peers and really understand what your feedback is telling you.

So let’s step through some of the big statistical topics needed in customer feedback analysis.

[Read more...]

Writing the Perfect Customer Feedback Survey Invitation

bigstock-Vintage-Typewriter-41977552

bigstock-Vintage-Typewriter-41977552You already have a great survey invite subject line and now you need to follow that up with an email body that drives people to start the survey.

The invite doesn’t have to be long and complex, in fact it should be short and to the point but to be most effective it must include some key information.

At all times remember the goal of the invite: to persuade the respondent to provide their feedback on your organisation. That’s it. Nothing more. Don’t add words into the invite that do not directly help you achieve that goal. [Read more...]

Here are the Most Effective Customer Feedback Survey Incentives

close up of man hands holding gift box

close up of man hands holding gift boxSitting with management at the final review for your customer survey questionnaire someone is going to ask about the incentive for survey completions. It’s a good question but what is the right answer?

Here are the most effective customer feedback survey incentives and when you should use them for maximum ROI.

[Read more...]

Email Subject Lines that Drive Customer Feedback Survey Responses

bigstock-delete-button-18380024

bigstock-delete-button-18380024Before a customer can complete your meticulously developed customer feedback survey, they need to open the email invitation. You have precious few seconds to prevent their finger jabbing at the delete key and your subject line is your first defence. So make it a good one.

There is lots research into what drives higher general email open rates and we can leverage this to make sure that we can maximize the open rates of our client surveys. So let’s take a look.

[Read more...]

The Best Practice Net Promoter Roll-Out Process

best-practice-nps-rollout

Okay so you know what you need to implement but how do you get there? When I talk to most new clients they want to start with Listen. Let’s send some surveys they say.

However, best practice is to hold off on that and start by deciding how you’re going to use the information.

Here are five steps to the roll-out process:

best-practice-nps-rollout
[Read more...]

[Webinar] Using 5 Whys Root Cause Analysis on Customer Feedback

5-whys-webinar-thumb

5 Whys is one of the most commonly used quality system tools. It is a simple and methodical way to identify the root cause of an issue.

When applied to Customer Feedback you can convert “interesting feedback” in to root causes and actions plans to drive improvement in your customer experience.

In this 30 minute webinar, we’ll teach you how to use this high value Customer Feedback tool:

  • Defining The 5 Whys Approach – What exactly is The 5 Whys Approach?
  • Customer Feedback Application – How to apply The 5 Whys Approach for Customer Feedback with practical examples.
  • Is it really for me? – When should you use The 5 Whys Approach?

[Read more...]

The 6 Critical Success Factors to Customer Feedback Success

do-net-promoters-tell-us-anythingThere are six key requirements for successfully implementing Customer Feedback or Net Promoter that you need to keep in mind as you launch the process.

Senior Management Buy-In

Firstly you must have buy-in from senior management. Getting a Net Promoter program up and running takes cultural change across the organisation. It also takes internal (staff) resources, budget and the commitment to follow through on delivery.

If you do not have support from senior management (up to and including the business manager or CEO) then you will struggle to succeed beyond adding a question or two to your customer survey.

If they are not already on-board you will probably need some business value based case studies to show what is possible. Lots of links between NPS and business value have been made but the case studies can be difficult to find.

However, we have collected a range of real world statistics from different industries for just this purpose. So if you need ammunition check out this post of Net Promoter Case Studies and Success Stories.

[Read more...]