Lifetime customer value - we consider LTV
“I saved my hottest hugs, kisses, smiles, respect and deepest admiration for marketers and analysts who calculate LTV,” exclaim web analytics guru Avinash Kosick. These are not simple emotions - this is the real state of things.
In our country, the first to consider the indicator LTV (Lifetime Value - customer lifelong value) were cellular operators. Their need was not accidental - against the background of a high level of penetration of cellular communication, the cost of attracting a single customer grew, it was time to get rid of unprofitable sales channels and change the distribution model.
Today, e-commerce is becoming more and more similar to the activity of a cellular operator: massive customer acquisition, serious outflow, and numerous lead generation and sales channels. This kind of e-commerce has generated a new paradigm for calculating work efficiency - from the standpoint of LTV, the total customer value. There are few followers of this paradigm, but in vain.
Most internet marketers and analysts use in their work a set of indicators that allow to evaluate the effectiveness of advertising campaigns: the level of failures, CTR, the number and share of conversions, churn (outflow), the cost of attracting a client. These indicators are able to give a general idea of the effect of marketing activities and the level of customer loyalty, but from a financial point of view, they carry almost no meaning if you don’t consider LTV together with them.
')
The formulas for calculating LifeTime Value are quite numerous and they depend on the purpose of using this indicator.
In order to simply determine the importance of a buyer for an online store, it is sufficient to use elementary summation based on
You can also use the classic formula for calculating:
LTV = (Monthly Revenue Per Customer) / Monthly ChurnRate
ChurnRate = Q / Nt , where
Q - the number of users left at the end of the period
Nt - the total number of remaining at the end of the period
To simplify this formula and some source comparison, you can use
GrossMargin per Customer = (TotalRevenue - Costs) / Nt
In general, you can find many formulas on the Internet and adapt them to a specific customer relationship management structure. Here, for example, is a rather universal formula that can be found:
LTV = AC × N × P × t ,
where AC is the average check, N is the average number of purchases per month, P is the profit share to the average check amount, t is the average lifetime of the user (how many time periods he is your customer - in months, days, years).
There are also formulas made up of customer churn. An interesting LTV calculation case can be found, for example, in a translated blog article.
The manager calculates the average cost of attraction, and then the average value of the client, the rest of the groups are distributed according to the “below average” and “above average” principle. This method is not the best solution, because does not give exact values and does not take into account additional factors associated with a specific promotion channel.
that provide advertising agencies. You can see, for example, a calculator from Netpeak , which counts LTV based on the data you entered. In principle, a good option, but it has general restrictions on the values, and also does not take into account industry specifics.
For example, in RealWeb we are calculating LTV for each of our clients and already proceeding from the combination of this and other indicators we are building a further advertising strategy.
As you have already seen, using simple web-analytics tools is not enough for solving the problem of estimating the lifetime value of a client. Unfortunately, Google Analytics in general does not know how to calculate the LTV indicator, but we’ll tell you about the private one a bit below. Accordingly, to calculate the indicator, you will need to carry out some training.
It is desirable to carry out value segmentation in order to understand which channels bring the most valuable customers from each segment. For example, take two buyers of a hardware store. Let 1,000 rubles be spent on attracting everyone. (say, AdWords). Client A came and bought a TV for 27,000 rubles. Six months later, he bought speakers for 3,000 rubles. Client B is the head of a software development company. Once a week he comes to the store and buys a USB flash drive, cable, network filter, or a cool mouse as a gift to a partner. On average, a week he spends about 1,200 rubles. It takes a year. Client A brought 30,000 rubles. He will not return, because he no longer needs new appliances. Client B brought 1,200 * 52 = 62,400 rubles. And will bring on. And who do you think, who received the loyalty card immediately, and who after the accumulation of a certain amount? Meanwhile, customer B could simply refuse from the store during this time, and in a particular case a loyalty card would serve as an additional incentive to maintain relationships.
It is necessary to analyze the channels and campaigns that brought medium and above-average customers in order to allocate funds for incentives.
It is necessary to determine the client's lifetime based on experience or available statistics, set the period measurement unit (for example, for a restaurant or grocery store it is a month and years, and for an online store - days and weeks, although it’s not all that simple) . In addition, it is necessary to measure the periods of repeated actions (purchases, payments) - so you can split the lifetime into intervals and predict profits or plan promotional activities.
As you already understood, the data for calculating LTV are taken from the outside, so you will need to obtain data from CRM / ERP / 1C or from the financial (economic) department for further calculation. Here is a sample list of data you may need:
The higher the client’s value obtained during the period of its life, the more space you have for further actions: developing loyalty programs, opening new channels of attraction, marketing activities and research activities. That is, in fact, you can more freely spend money on such customers. If LTV shows a downward trend - this is a dangerous sign, you need to take action and study the accumulated customer base and target the activity on pre-sales and complementary sales. It's simple: to keep a client cheaper than to attract a new one.
Indeed, the platform provides for an LTV section (in English and Russian). The report, which is available only for application views, calculates the value of LTV in the context of attraction channels, based on the life cycle and revenue volume. LTV comparison of different user groups is also available. The maximum evaluation period is 90 days. This is quite a small interval, however, it is quite suitable for mobile applications and their dynamics. This is not to say that this report is ideal, but it gives some idea for further analysis.
In principle, according to his logic, he is close to the cohorts that we have already considered in our blog . It remains to hope that the tool will develop and web analysts will receive at their disposal a report for calculating the LTV customers who come from various online channels. In conjunction with attribution models, this will give a strong impetus to the development of e-commerce analytics.
The calculation of LTV cannot be replaced by a pool of other indicators - this is a valuable value, which has an economic and marketing sense. Of course, it requires data collection for substitution into a fairly simple formula, but the effort is worth it - no wonder some companies enthusiastically tell that they saved tens of thousands of dollars after introducing LTV into the system of monitored business intelligence indicators. Awareness of the value of customers allows us to competently and reasonably develop loyalty programs, to allocate truly “right” customers. And, as you remember, only the right bees carry the right honey.
In our country, the first to consider the indicator LTV (Lifetime Value - customer lifelong value) were cellular operators. Their need was not accidental - against the background of a high level of penetration of cellular communication, the cost of attracting a single customer grew, it was time to get rid of unprofitable sales channels and change the distribution model.
Today, e-commerce is becoming more and more similar to the activity of a cellular operator: massive customer acquisition, serious outflow, and numerous lead generation and sales channels. This kind of e-commerce has generated a new paradigm for calculating work efficiency - from the standpoint of LTV, the total customer value. There are few followers of this paradigm, but in vain.
Most internet marketers and analysts use in their work a set of indicators that allow to evaluate the effectiveness of advertising campaigns: the level of failures, CTR, the number and share of conversions, churn (outflow), the cost of attracting a client. These indicators are able to give a general idea of the effect of marketing activities and the level of customer loyalty, but from a financial point of view, they carry almost no meaning if you don’t consider LTV together with them.
')
Why and how to calculate LTV?
- His calculation focuses on success - you will know exactly which channels bring you the best customers.
- You will know exactly the effectiveness of each channel to attract customers and be able to reallocate costs based on the needs of your business.
- You will see the value of each customer group in the long term.
- You will be able to estimate savings points, as well as understand how much additional funds can be spent on attracting and retaining a client (for example, through remarketing, mailings or advertising campaigns in social networks).
The formulas for calculating LifeTime Value are quite numerous and they depend on the purpose of using this indicator.
Ready-made formulas
In order to simply determine the importance of a buyer for an online store, it is sufficient to use elementary summation based on
- the volume of repeat orders, which was in the period under review compared to the previous one:
- we receive data on orders for the reporting period
- get data about orders for a longer period (for example, six months)
- aggregate user data with the inclusion of all records from (1) and only matching from (2)
- summarize the order data for each user and find the average value.
You can also use the classic formula for calculating:
LTV = (Monthly Revenue Per Customer) / Monthly ChurnRate
ChurnRate = Q / Nt , where
Q - the number of users left at the end of the period
Nt - the total number of remaining at the end of the period
To simplify this formula and some source comparison, you can use
GrossMargin per Customer = (TotalRevenue - Costs) / Nt
In general, you can find many formulas on the Internet and adapt them to a specific customer relationship management structure. Here, for example, is a rather universal formula that can be found:
LTV = AC × N × P × t ,
where AC is the average check, N is the average number of purchases per month, P is the profit share to the average check amount, t is the average lifetime of the user (how many time periods he is your customer - in months, days, years).
There are also formulas made up of customer churn. An interesting LTV calculation case can be found, for example, in a translated blog article.
Evaluation method
The manager calculates the average cost of attraction, and then the average value of the client, the rest of the groups are distributed according to the “below average” and “above average” principle. This method is not the best solution, because does not give exact values and does not take into account additional factors associated with a specific promotion channel.
Finished calculators,
that provide advertising agencies. You can see, for example, a calculator from Netpeak , which counts LTV based on the data you entered. In principle, a good option, but it has general restrictions on the values, and also does not take into account industry specifics.
Specialist calculators
For example, in RealWeb we are calculating LTV for each of our clients and already proceeding from the combination of this and other indicators we are building a further advertising strategy.
Terms of calculating indicator LTV
As you have already seen, using simple web-analytics tools is not enough for solving the problem of estimating the lifetime value of a client. Unfortunately, Google Analytics in general does not know how to calculate the LTV indicator, but we’ll tell you about the private one a bit below. Accordingly, to calculate the indicator, you will need to carry out some training.
It is desirable to carry out value segmentation in order to understand which channels bring the most valuable customers from each segment. For example, take two buyers of a hardware store. Let 1,000 rubles be spent on attracting everyone. (say, AdWords). Client A came and bought a TV for 27,000 rubles. Six months later, he bought speakers for 3,000 rubles. Client B is the head of a software development company. Once a week he comes to the store and buys a USB flash drive, cable, network filter, or a cool mouse as a gift to a partner. On average, a week he spends about 1,200 rubles. It takes a year. Client A brought 30,000 rubles. He will not return, because he no longer needs new appliances. Client B brought 1,200 * 52 = 62,400 rubles. And will bring on. And who do you think, who received the loyalty card immediately, and who after the accumulation of a certain amount? Meanwhile, customer B could simply refuse from the store during this time, and in a particular case a loyalty card would serve as an additional incentive to maintain relationships.
It is necessary to analyze the channels and campaigns that brought medium and above-average customers in order to allocate funds for incentives.
It is necessary to determine the client's lifetime based on experience or available statistics, set the period measurement unit (for example, for a restaurant or grocery store it is a month and years, and for an online store - days and weeks, although it’s not all that simple) . In addition, it is necessary to measure the periods of repeated actions (purchases, payments) - so you can split the lifetime into intervals and predict profits or plan promotional activities.
As you already understood, the data for calculating LTV are taken from the outside, so you will need to obtain data from CRM / ERP / 1C or from the financial (economic) department for further calculation. Here is a sample list of data you may need:
- average check per customer
- average number of purchases per customer during the reporting period
- advertising costs per month and for the period
- the number of new customers per month and for the period
- monthly outflow of customers and period
- return percentage
- percent of discounts and margins on client groups.
The higher the client’s value obtained during the period of its life, the more space you have for further actions: developing loyalty programs, opening new channels of attraction, marketing activities and research activities. That is, in fact, you can more freely spend money on such customers. If LTV shows a downward trend - this is a dangerous sign, you need to take action and study the accumulated customer base and target the activity on pre-sales and complementary sales. It's simple: to keep a client cheaper than to attract a new one.
Google Analytics has LTV for mobile apps.
Indeed, the platform provides for an LTV section (in English and Russian). The report, which is available only for application views, calculates the value of LTV in the context of attraction channels, based on the life cycle and revenue volume. LTV comparison of different user groups is also available. The maximum evaluation period is 90 days. This is quite a small interval, however, it is quite suitable for mobile applications and their dynamics. This is not to say that this report is ideal, but it gives some idea for further analysis.
In principle, according to his logic, he is close to the cohorts that we have already considered in our blog . It remains to hope that the tool will develop and web analysts will receive at their disposal a report for calculating the LTV customers who come from various online channels. In conjunction with attribution models, this will give a strong impetus to the development of e-commerce analytics.
The calculation of LTV cannot be replaced by a pool of other indicators - this is a valuable value, which has an economic and marketing sense. Of course, it requires data collection for substitution into a fairly simple formula, but the effort is worth it - no wonder some companies enthusiastically tell that they saved tens of thousands of dollars after introducing LTV into the system of monitored business intelligence indicators. Awareness of the value of customers allows us to competently and reasonably develop loyalty programs, to allocate truly “right” customers. And, as you remember, only the right bees carry the right honey.
Source: https://habr.com/ru/post/293904/
All Articles