Questions tagged [customer-lifetime-value]

Customer Lifetime Value (LTV) is the projected, discounted revenue that a customer will generate during their lifetime. LTV is a key business metric for many types of businesses.

Customer Lifetime Value (LTV) is the projected, discounted revenue that a customer will generate during their lifetime. LTV is a key business metric for many types of businesses, historically the domain of marketing functions. The prediction model can have varying levels of sophistication and accuracy, ranging from a crude heuristic to the use of complex predictive analytics techniques.

Customer Lifetime Value (CLTV) takes the following general form:

$$ {\text{CLV}}={\text{GC}}\cdot \sum _{{i=1}}^{n}{\frac {r^{i}}{(1+d)^{i}}} $$

where ${\displaystyle {\text{GC}}}$ is yearly gross contribution per customer, ${\displaystyle n}$ is the horizon (in years), ${\displaystyle r}$ is the yearly retention rate, ${\displaystyle d}$ is the yearly discount rate. Extensions of this include marketing costs.

${\displaystyle {\text{GC}}}$ and ${\displaystyle r}$ are often the quantities of interest. One can use simple heuristics based on historical data, or complex predictive model to estimate them.

17 questions
14
votes
3 answers

RFM & customer lifetime value modeling in R

Can anybody tell me how to do recency, frequency & monetary value (RFM) modeling & customer value modeling in R? Also, can somebody refer me some literature on it?
Beta
  • 5,784
  • 9
  • 33
  • 44
4
votes
0 answers

Estimating variance of MLE estimate of Beta-Geometric/NBD without MCMC

I'm using Fader's BG/NBD model for customer LTV calculations. The log likelihood is the following: $$ \sum_{i=1}^{N} \ln L(r, \alpha, a, b|X_i=x_i, t_{x_i}, T_i) = \frac{B(a, b + x)}{B(a, b)}\frac{\Gamma(r + x)\alpha^r}{\Gamma(r)(\alpha + T)^{r+x}}…
ilanman
  • 4,503
  • 1
  • 22
  • 46
4
votes
1 answer

How to fit newer cohorts using Pareto/NBD or Beta/Geo for CLTV

I am trying to fit the popular Pareto/NBD or Beta/Geometric models for non-contractual, continuous customer data. On top of that I then fit the Gamma/Gamma model for monetary value (using the very helpful lifetimes package). There are a few…
3
votes
0 answers

Calculate CLV / TLV for contractual Business

I am desperately trying to apply a CLV/TLV (Customer Lifetime Value) algorithm to my dataset in R. Unfortunately, the more I read about it, the less confident I get if all makes sense. =) Do you have an advice on how I can calculate the CLV? I have…
2
votes
0 answers

Brainstorming ways to calculate customer lifetime value for a long term subscription model

Business Model: Customers sign up for 12 month contracts with the option to incorporate add on services (these are also contractual). If they cancel, they still pay for 12 months and their account will be terminated once their contract is…
madsthaks
  • 277
  • 3
  • 11
1
vote
1 answer

Which regression metrics fit best cases with very large values and exactly 0 values?

I'm working on a CLTV problem, where the objective is to predict the future spending of the customers, given their past behaviour. According to arXiv:1912.07753, paragraph 4 EVALUATION METRICS, I'm measuring calibration (difference between actual…
1
vote
0 answers

How can I predict customer lifetime value in 12 months?

I was trying to predict 'customer lifetime value' but only for the first 12 months of him/her in the platform I work for. However, most models, such as Pareto/NBD, are good in order to assess the real lifetime value, not only for this short…
dsbr__0
  • 707
  • 2
  • 7
1
vote
2 answers

What is a good modelling approach for Customer Lifetime Value when consumption intensity matters?

I would like to produce a customer livetime value (CLV) analysis where the CLV is tied to the intensity of the use of the product, i.e. it is not a subscription business (where the relevant event would be the cancellation of the subscription), but…
clog14
  • 181
  • 10
1
vote
2 answers

Why do Pareto/NBD models require custom likelihood functions in PyMC3 and Stan?

I'm interested in Bayesian modeling of customer lifetime value (CLV), preferably via PyMC3. I've found that research in this area started mid-to-late 1900's and has remained active since. It would seem that some combination of Exponential, poisson,…
jbuddy_13
  • 1,578
  • 3
  • 22
1
vote
0 answers

Chaining/combining logistic and linear models

I have a analysis that is looking to predict the total customer value based on a customer's first purchase amount. I am noticing that a set of features predict whether the customer will purchase ever again after their first purchase (first purchase…
1
vote
1 answer

Best way to forecast retention over a certain timespan

Let's say I have data, import io data_string = """ retention 0 1.000000 1 0.915227 2 0.905325 3 0.895019 4 0.882389 5 0.872285 6 0.861069 7 0.853087 8 0.848237 9 0.843387 10 0.838537 11 0.835708 12 …
Pwon
  • 25
  • 4
1
vote
1 answer

Pareto/NBD and New/Existing Customers

I am looking at implementing a Pareto/NBD model to forecast customer lifetime value in a non-contractual business setting. One thing I haven't got my head around yet is whether such a model is equally applicable to both new and existing…
0
votes
1 answer

What do I need to read/study to understand Peter Fader's papers that deal with CLV, Pareto/NBD models etc.?

I would appreciate if someone could tell me which subjects these papers listed below "belong to", and how do I get started in understanding them - what do I need to read/study and in what sequence. In general, I would like to be able to understand…
RahPah
  • 1
0
votes
0 answers

life time value prediction with logistic regression in r?

I have a mobile game dataset as below and I m expected to make analysis from users first 3 Days activity and predict/classify users’ Lifetime Values (LTV) from the user-level dataset with implementing logistic regression and randm forest models in…
0
votes
1 answer

Fader/Hardie BG/NBD model. Interpretation of $a$ and $b$ of the beta distribution drop out process

On R's CLVTools package documentation, there's a sentence referring to the pareto/NBD model I'm working with the BG/NBD model not the pareto/NBD model. I'd like to understand if I can interpret the parameters of this model in a similar way to the…
Doug Fir
  • 1,218
  • 1
  • 13
  • 27
1
2