Release of 'LTV Analysis Diagnostic' to Identify Causes of Inability to Calculate LTV by Initiative and Determine Current Status
INVOX Inc. has released the 'LTV Analysis Diagnostic,' a tool to determine the extent to which your marketing LTV analysis is currently being implemented.
To what extent are you utilizing LTV (Lifetime Value) in your business?
By answering 10 questions in about 2 minutes, you can determine which of the four stages your company's analysis environment is in and receive suggestions on what to work on next.
This is intended for those who make decisions using data, such as managers in marketing, CRM, e-commerce, growth, and business operations.
LTV Analysis Diagnostic (Free, approx. 2 minutes)
Background
We are receiving an increasing number of inquiries about wanting to analyze internal company data using Claude or ChatGPT.
On the other hand, when we actually speak with them, we find that in the majority of cases, they are stuck before they even reach the AI analysis stage.
What is holding them back is not how to use AI.
It is that basic analysis, including LTV, is not in a state where it can be used for decision-making.
And in many cases, they are not even aware of it.
The reason is that the managers' perception is not that 'LTV cannot be seen,' but rather that it is in a state where 'it can be seen if I want to see it.'
LTV analysis stops at 'it can be seen if I want to see it'
Many companies list LTV as a key KPI.
However, when asked specifically, it almost always falls into one of the following categories.
Overall LTV can be calculated, but LTV by initiative or media cannot be calculated
It can be calculated, but it requires exporting to Excel and aggregating every time
Only the analyst can produce the numbers, and the manager themselves cannot
Sales and marketing see different numbers when looking at the same metric
None of these four are called 'cannot be seen.'
They appear if you request them, and they can be aggregated if you spend the time.
However, from the perspective of decision-making, the evaluation changes.
If the numbers for deciding next week's ad allocation take three days to arrive, they are not useful for decision-making.
If you have to postpone a request to 'see these numbers by media' during a meeting, that meeting will not reach a conclusion.
There is a greater distance than you might expect between 'being able to see' data and 'being able to use it for decision-making'.
The LTV formula and the reasons why it cannot be calculated by strategy
The LTV formula itself is not complex.
For subscriptions and recurring purchases, it is calculated using the following formula.
平均継続月数 ＝ 100 ÷ 月次解約率（%）
LTV ＝ 平均顧客単価 × 平均継続月数For one-time purchase e-commerce, it takes the following form.
LTV ＝ 平均購買単価 × 購買頻度 × 継続期間All you need are three things: unit price, frequency, and duration (or churn rate).
Most companies are already able to calculate company-wide average LTV.
The problem arises when you break this down by strategy.
For example, to calculate the 'LTV of customers acquired from Google Ads,' the following three must be connected by the same customer ID.
Which customers came from Google Ads (linking acquisition strategy to customers)
How much those customers purchased after acquisition (linking purchase history to customers)
When those customers left or canceled (history of retention status)
Group customers by the strategy used to acquire them and the time of acquisition, then track how much that group purchases later.
As an analysis method, this is called cohort analysis.
The formula does not change. What changes is how the necessary data is connected.
The reason LTV by strategy cannot be calculated lies not in the calculation method, but in how the data is stored.
Four reasons why LTV by strategy cannot be calculated
The points where analysis stalls are generally predictable.
1. Costs for each media are separated in their respective management consoles
Ad spend is in each media's management console, sales are in the e-commerce cart or CRM, and email or LINE are in separate tools. To line them up, you need to copy and aggregate them every time.
Because there is a limit to the data retention period in management screens, you may find that data is no longer available when you try to look back at the performance of advertisements placed last year.
2. The same customer is being counted as a different customer
People who came from advertisements, people who purchased via LINE, and people who came through referrals.
If customer IDs are not integrated, these three are counted as three separate people.
In this state, you cannot identify which strategy acquired which customer on an individual (or company) basis.
You can only compare the number of acquisitions and the acquisition cost.
3. Purchase and cancellation history is not linked to acquisition strategies
This is a case where even if you can see acquisition by strategy, the number of purchases and the duration of retention after acquisition are not connected to the strategy.
In this case, you cannot distinguish between strategies with low acquisition costs and strategies that generate profit.
This is where the LTV metric is most effective.
4. Even if you can generate the data, the person in charge cannot access it themselves
This is a state where a dashboard exists but conditions cannot be changed, or where a request to an analyst is required every time you want to check it.
Even though the numbers exist, you cannot see them when you want to.
If checking the data itself incurs a cost, the frequency of checking will decrease.
There is an order to these four points.
You cannot fix point 4 while point 1 is still missing.
Conversely, companies that have reached point 4 have necessarily passed through points 1 to 3.
Therefore, 'whether LTV analysis is being done' should not be viewed as a binary choice of yes or no, but rather as the stage of how much has been built up to better reflect reality.
Contents of the LTV Analysis Diagnostic
There are 10 questions, and it takes about 2 minutes to complete
Questions are divided into toC (EC, D2C, subscription) and toB (corporate, SaaS)
The choices are three options: 'Done (O) / Partially Done (△) / Not Done (X)'
There is no cost
We are not asking for tool names or budgets, but only things you can answer based on your daily experience.
For example, we ask questions like this
Q. Can you check the acquisition cost for each advertising medium at any time?
Q. When someone asks during a meeting, 'Can we see these numbers under different conditions?', can you check it on the spot?(toC)
Q. Do the numbers match when Sales and Marketing look at the same metrics?(toB)
What you will learn from the diagnostic results
After answering, the following content will be displayed immediately.
Your current stage (LEVEL 1-4) and an explanation of that status
What you are already doing well
Where you are currently stuck
Where to start (up to 3 steps)
Changes you can expect after setting up the environment (time to calculate LTV, effort required for checking, frequency of checks, etc.)
Under 'Where to start', specific action items such as 'Advertising data integration,' 'Customer ID integration,' 'Purchase data connection,' and 'Self-service analysis environment' are listed in order, starting from the foundation.
Companies for whom this LTV analysis diagnostic is suitable
This diagnostic is suitable for companies like the following:
Companies that have set LTV as a key KPI but cannot calculate it by initiative or medium
Companies that have to aggregate data in Excel every time to calculate LTV
Companies where you cannot check the numbers without requesting them from an analyst or the IT department
Companies where Sales and Marketing have conflicting numbers, and discussions always start with reconciling the data
Companies considering data analysis with Claude or ChatGPT but are unsure where to begin
Related case studies
Case studies where LTV analysis and advertising data integration were actually implemented are introduced below.
Other case studies are compiled in our Case Study Magazine.
Summary
Having 'LTV as a key KPI' and 'making decisions based on LTV' are two different things.
The former can be written in a presentation, but the latter is determined by whether weekly budget allocations actually change.
Even if the LTV formula is simple, breaking it down by initiative requires that acquisition tactics, purchase history, and retention status are all linked to the same customer ID.
The reason why LTV by initiative cannot be calculated lies not in the calculation method, but in how the data is stored.
First, please try checking your current status.
Take the LTV Analysis Diagnostic
LTV Analysis Diagnostic (Free, approx. 2 minutes)
Consult based on diagnostic results and challenges
If you would like to specifically consider 'where your company should start' after the diagnostic or because your challenges are already clear, please use our 30-minute free consultation.
We will ask about the tools you are currently using, the themes you want to analyze, and the areas where you are stuck, and then propose an order of operations.
Book a 30-minute free consultation
At INVOX, we provide support ranging from the design and construction of data infrastructure to the building of AI analysis environments using Claude or ChatGPT.
For those who want to build an environment capable of AI analysis beyond just LTV analysis, we also offer the following free support.