Managing Customer Relationships Through Price and Service Quality

Managing Customer Relationships Through Price and Service Quality Agenda | Problem raised by the paper | Model description z z Individual customer behavior Aggregate customer behavior | System behavior in steady state | Profit optimization Agenda | Problem raised by the paper | Model description z z Individual customer behavior Aggregate customer behavior | System behavior in steady state | Profit optimization Introduction | Imagine that you are a computer hotline… z z | You want to increase your profit, which parameters can you adjust? z z | Price of the service Number of people who answer the questions (“service quality”) How do you play on these parameters? z | People call you when they have questions about their computers They pay for this service Decreasing price and increasing service quality can augment the number of customer and increase my profit But… z z More customers can lead to more delays and thus more customers who defect Lower price can lead to less profits General context | | | | Subscription-based, capacity-constrained service Customers choose the depth of their relationship with the company based on their level of satisfaction Pricing and service quality affect this level of satisfaction The intuitive reasoning: Service quality Pricing | But sometimes: Customer base and profit Service quality Customer base Price profit Complex relationship between pricing, service quality and profitability Agenda | Problem raised by the paper | Model description z z Individual customer behavior Aggregate customer behavior | System behavior in steady state | Profit optimization Service quality encountered and expected Customers interact with the company whenever they choose to do so w k evolves with: w k = αk wk + (1 − αk )w k −1 x is a random variable with distribution ik F(x| w ) = F k Interactions model ps subscription fee to have access to the service during a period [ pT; (p+1) T] pu usage fee. To be paid each time the service is used Rate of the interaction with the customer Customers use the service at the rate η : η (level of satisfaction, pu ) Utility of customer: (v(η ) − pu ) + c( w) Customers will renew their subscription when: ( ∞ ) ( i ) ⎤ − p > 0 with c( w i ) = c( x)dF x | w i T ⎡η (υ (η ) − pu ) − η c( w s ∫ ⎣ ⎦ 0 We can then define b, the expected net utility per unit of time: ( ) ( ) i (⋅) = T ⎡η (υ (η ) − p ) − η c( w i) ⎤ − p b η ; ps , pu , F u s ⎣ ⎦ The customer will choose η * = arg max(b) i (⋅) = b* < b And he will defect whenever: b η * ; ps , pu , F min ( ) ) Results M onotonocity of custom er's usage rate: η * 2 w ith p u / η * 2 w ith c / η * 2 w ith w / M onotonocity of the custom er's ut ility: b * 2 w ith p u and p s / b * 2 w ith c / b * 2 w ith w / T he com pany can use w as a concrete and m anageable m easure of custom er utility b * Agenda | Problem raised by the paper | Model description z z Individual customer behavior Aggregate customer behavior | System behavior in steady state | Profit optimization Aggregate customer behavior Customers stay with the company if b* ≥ bmin That corresponds to w ≤ w max with b* (w max ) = bmin We can construct a Markov process where each state is described by the number of previous interactions and the customer's current level of satisfaction represented by w . To discretize the set of level of satisfaction we use: I= [ 0;w max ] = [ 0; u1 ] ∪ [l1; u2 ] ∪ ...[ls ; us ] The states of theMarkov process are (i,k) where after the k th interaction, w ∈[li ;ui ] State transition probability | Recall that: w k = αk wk + (1 − αk )w k −1 | In term of probability: i k +1 ≤ x | w i k = y ) = Pr ⎛ w ≤ x − (1 − α k ) y ⎞ = F ⎛ x − (1 − α k ) y ⎞ Pr( w ⎜ k ⎟ ⎜ ⎟ α α k k ⎝ ⎠ ⎝ ⎠ Where F is the true distribution of the company service quality |  is uniformly distributed in [li; ui] then: If we assume that w i k +1 ≤ u | w ik ∈ I ) = p = Pr(l j ≤ w j i k ij 1 ∆i ui ⎡ ui ⎛ u j − (1 − α k ) y ⎞ ⎛ l j − (1 − α k ) y ⎞ ⎤ ⎢∫ F ⎜ ⎟ dy − ∫ F ⎜ ⎟ dy ⎥ αk αk ⎢⎣ li ⎝ ⎠ ⎝ ⎠ ⎥⎦ li System behavior in steady state | Now that we know how the customers react to the decisions of the company, we can calculate the number of customers who pay the subscription fee and the service demand rate from customers to the company λ jk = ∑ i: I i ∈I { } i ∈ I ⎤ → ⎡ b* ( w i) ≥ b ⎤ pijk −1 λik −1 with I = I i ∈ I : ⎡⎣ w i⎦ min ⎦ ⎣ Customer response to a change in the service quality Let λ the total service demand rate: λ = ∑ λik i ,k λ 2 with W / BUT... There is no such rule for the total number of customers N λ jk ( ie: N calculated with N= ∑ ) j,k η j Intuitive explanation | | As the service quality decreases, the probability that the customer leaves the company increases. At the same time his number of interactions decreases Let’s study the effects of this on the length of stay: E (length of stay)=E (total number of interaction) x (Average interval between interaction) Decrease with a lower service quality | Increase with a lower service quality The length of stay can thus increase with the decrease of service quality and thus make the number of customer increase Customer response to a change in the company pricing policy pu and ps affect directly wmax which in turn affects λ . For a higher price, customers expect a better service Consequently: λ 2 with pu and ps / BUT... There is no such rule for the total number of customers N Intuitive explanation | | | | Based on the monotonicity of η*, the increase of pu will produce an increase in the time between customer interactions But at the same time, the increase of pu will make the criteria for staying in the company more strict and thus the number of interactions will decrease There is therefore no direct relation between the length of stay and pu, and consequently, no direct relation between N and pu On the other hand : N 2 with ps / Agenda | Problem raised by the paper | Model description z z Individual customer behavior Aggregate customer behavior | System behavior in steady state | Profit optimization Problem description | Revenue: R = λ pu + Nps | Cost: C (λ , W ) | Thus we have the following nonlinear program to solve: max Π = λ pu + Nps − C (λ , W ) Such that: λ = g ( pu , ps ,W ) N = h( pu , ps , W ) W ≥0 Solving and interpretation | Using the Lagrange multipliers we obtain the equations: | Those equations equate marginal revenue with marginal costs Each of them can be decomposed into several intuitive parts | Conclusion | | | This paper gives the analytical tools to understand the complex relation between pricing policy, service quality and customer base and behavior The main originality of this paper is the introduction of this factor of relation depth: η Principal criticism: the paper does not take into account the fact that there is a difference between subscribing to the whole period and subscribing for one interaction. At the end of each period, even if the customer had been satisfied enough to make another interaction, he could decide that overall the service is not good enough to pay for another whole period