Most people working in marketing know what customer insight means: an unspoken truth—understanding customers and their behavior even when they do not express it.
But how, then, does customer insight differ from the research or customer data you already have? How can you generate insights about your customers and use them to improve business? What mistakes and misunderstandings commonly arise? Beyond insight and its application, what else can we do? This article will help answer those questions.
* In this article I use the English terms customer insight or insight for brevity instead of the Vietnamese phrase meaning “implicitly understood truth.”
What insight is and its characteristics
Customer insight involves interpreting behavior and trends of customers based on the data that we have about them, through which can take actions to improve product and service quality and increase sales revenue in order to both sides (brand and customers) benefit everyone.
Some essential characteristics of customer insight:
1. It Is Not an Obvious Truth If it were obvious, it would not be called an underlying insight. For example: From Google Analytics, you learn that 70% of visitors are aged 18–24, and infer that most website visitors are young. This is too obvious to qualify as an insight.
2. Do not rely on just one type of data. You need to combine multiple sources, indicators, data sets, and categories to generate accurate insights. For example: Judging a webpage's content as poor solely because it has a high bounce rate—the number of people who enter and leave without interacting—may be inaccurate. The page may provide such complete, useful content that visitors finish reading satisfied, with no need to search for or view anything else, and leave immediately, producing a high bounce rate. However, evaluating both bounce rate and time on page—the time visitors spend on the website—is more accurate. A high bounce rate combined with a high time on page means the content is good. A high bounce rate with a low time on page means the content genuinely has a problem. Combining multiple metrics therefore helps you uncover more accurate, deeper insights.
3. Practical actions can be developed from that insight. If it remains mere theory that cannot be applied or verified, it is not an insight either. For example: Your company has 2 business segments: B2B (business customers) and B2C (individual customers). B2B is doing very well, but revenue from B2C needs improvement. After researching the B2C group, you conclude that individual customers really like referring services to friends and receiving commissions. You therefore think a customer-referral system should be established to increase customer numbers and revenue. In reality, however, such a system cannot be implemented under the company's current conditions in terms of staffing (too many people needed to manage it), costs (investment in building the referral system), and time (6 months to complete and launch it) ==> If a conclusion cannot be translated into a specific, practical action—even before considering whether it is good or bad—it cannot be called an insight; it remains merely an observation or thought.
4. If implemented, the action mentioned above must be capable of persuading customers to change their behavior. For example: You discover that after buying a laptop, customers often look for a computer mouse. Based on this customer-segment insight, you can place the mouse you want to cross-sell right next to the laptop to increase the purchase rate -> if customers agree to buy both 2 products, you have changed their purchasing behavior: buying 2 products instead of 1.
5. Behavioral change must benefit both parties: the brand and the customer. For example: Buying a computer mouse together with a laptop provides greater value to the buyer and higher revenue to the seller, meaning both sides benefit.
Every customer has their own thoughts and behaviors, and we need to look beyond data or numbers. We also need to understand that customer insights change with time, trends, technology, timing, seasons, age, and many other factors. If you analyze and assess only past behavior without uncovering anything new about customer understanding, your customer insights will gradually become outdated and inaccurate.
How do you develop customer insights?
Once you understand what insight is and its characteristics, you can use those principles to develop customer insights and apply them to your business. This process has 3 steps:
- Collect data
- Interpret/analyze data to generate insights
- Take action based on insights
We will go through each step in detail below:
1. Collect Data
As mentioned above, insights come from data, and in digital marketing this data comes from:
- Website: sessions, time on site, bounce rate, etc.
- Mobile apps: screen views, time on screen, downloader information, etc.…
- Social media: followers, likes, shares, comments, etc.…
- Search / display advertising: impressions, clicks, conversions, CTR, CR, etc.
- Email: open rate, click rate, CTR, abuse/spam rate, list of unopened emails, etc…
- SMS: number of messages sent, open rate, list of phone numbers that could not be reached, etc.
- Online survey
These are just some common channels, not an exhaustive list. Insights can also come from other data sources, such as:
- Sales: information from CRM, order tracking files, contracts, etc.
- Customer service: information from call centers, telephone support lines, and web chat
- POS: information from systems at points of sale
- Customer reviews and feedback
- Market research
These are only some common channels, not all of them. Next, we will examine how to turn data into meaningful insights.
2. Interpret and analyze data to generate insights
Once you have data, you need to understand what it means and look for correlations between recurring patterns in certain metrics and your customers' goals (a better experience), as well as your own goals (making sales).
For example: You see that customers accessing your website on mobile phones currently have a lower purchase-conversion rate than desktop users. You infer that your website's mobile version may not yet provide a very good user experience, and that improving it could help increase revenue.
We can see that customers' satisfaction with their experience of buying a product or using a service brings revenue to a business, directly or indirectly. Therefore, not every insight necessarily needs to generate immediate revenue. Sometimes, simply focusing on improving the user experience is enough, and customers will then return more often of their own accord or refer new customers.
3. Develop actions based on insights
Once insights have been generated by analyzing the available data, you can start using them to take actions that bring you closer to your business goals. At this point, the insights interpreted and analyzed from the data must be checked against the characteristics described in part 1 to ensure they are accurate and suitable for application.
Actions derived from insights will differ depending on your goals, industry characteristics, company situation, market conditions and trends. There is therefore no universal standard or template for this.
Some common applications of insight
As mentioned above, although there is no universal standard, here are some ways to use customer insights generated from data, with illustrative examples:
1. Assess the Degree of Impact: helps businesses understand how their actions have affected customer behavior while allowing them to predict customer responses to proposed changes.
For example: Before implementing a new pricing policy, you can use insights gathered from various sources to assess how customers may react to the price, positively or negatively. Based on that insight-driven assessment, you can adjust the price more appropriately before bringing it to market.
2. Increase lifetime value: assesses customers’ lifetime value and allows businesses to measure factors such as customer acquisition costs and churn rates.
For example: Insights gathered from customers show that, on average, customers aged 15–22 replace their phones 1 time every 1 year and prefer expensive new models, while those aged 23–30 replace theirs 1 time every 2 years and do not need the newest model. Based on this insight, the manufacturer decides to release a new product about once every 1 year for customers aged 15–22, while lowering the price of the previous model so that customers aged 23–30 feel motivated to replace their old phones sooner rather than wait until the following year. This increases the second group's lifetime value and secures revenue from the first group.
3. Trend analysis: helps predict customers' future behavior based on past actions and helps businesses understand the likelihood that customers will behave in a certain way.
For example: Based on historical data, a real estate company knows that during Vietnam's 7th lunar month, the ghost month, customers limit buying and selling because of superstition, reducing demand. The company decides to cut advertising costs during that period to avoid waste and offer more attractive promotions to stimulate transactions.
4. Cross-sell/up-sell analysis: identify relationships between products and services to understand how best to combine products. These combinations can then be used for cross-selling or up-selling. Cross-sell: selling additional related products to customers who have bought a product. Up-sell: selling a product of the same type but more premium than the one the customer currently uses.
For example: An education company specializing in English instruction offers products such as a flexible adult English course with 15 levels, where 1 is the lowest and 15 the highest; an online business course from a prestigious foreign university; and IELTS, TOEFL, and SAT preparation courses. By analyzing data, the company discovers that students at higher levels (10-15) of the flexible adult English course often want to enroll in the foreign university's online business course, while intermediate students (6–9) often enroll in IELTS preparation. The company launches a promotion offering discounts when advanced students also purchase the online business course and intermediate students also purchase the IELTS course. This program increases cross-sold course purchases by nearly 200% during the promotional period.
These are only some common business applications. Different needs call for different applications and approaches.
Going beyond customer insight
However, application is not the end. You also need to understand the effects of those actions and what data can be drawn from them. There are 2 more things you SHOULD DO:
Reviews
- Effective or ineffective? How will you measure it? Using which tools? Which indicators? This is what you need to define clearly before applying and implementing activities based on insights.
- If it is effective, what percentage does it achieve, and could anything be done better?
- If it is ineffective, where did you go wrong? Was the action you developed from the insight inappropriate? Was the insight itself wrong? Or was unreliable data the source of the mistake from the beginning? Work backward through each stage to find the problem.
Collected data
- What data did the recent application activities generate?
- What additional insights can we derive from this new data?
- How can new data be combined with existing data to generate further insights?
- An insight can lead to not just one but several practical actions. What else could you do?
Summary
To sum up, a ‘good’ insight must be non-obvious, based on multiple data sources, actionable, and capable of changing customer behavior while benefiting both sellers and buyers. To create insights, begin by collecting data from different sources, then interpret it to form observations about customers. These observations qualify as insights only if they meet all the characteristics above, and only then should you apply them to improve business performance. After applying them, evaluate their effectiveness and collect data to continue deriving newer, clearer insights.
This article hopes to give readers a relatively comprehensive and clearer understanding of insight and help you improve what you are doing.
* Source: Conversion.vn
>> Great Articles on Entrepreneurship
