Method · Offer design
Menu-based conjoint: which package do your customers put together themselves?
In menu-based conjoint, participants build their own offer in every task: a base plan – or none – plus any add-ons, each with a price. Prices change from task to task. 121surveys shows which base plan wins, which add-ons are taken at which price and where revenue per person is highest.
Building and testing is free. To publish, you choose a paid plan.
Menu-based conjoint explained
What it's for
For offers customers assemble from building blocks – software with add-on modules, plans with options, menus. You see which base package wins, which add-ons respondents take at which price and where revenue per person is highest.
What respondents see
In each task respondents see all base packages and add-ons with prices and put together their package – or choose none. Prices change from task to task.
How many responses
You see the observed choice rates right away. For the simulator you need at least 30 answered tasks; in practice it becomes reliable from about 200 people, so that the survey has offered each add-on often enough at each price.
Reading the results
The tables show, per item and price, how often the survey offered it and how often respondents chose it – if the rate falls as the price rises, demand reacts to it. For add-ons only tasks with a base package count. In the simulator you set prices and see the shares of the base packages, the take rates of the add-ons and revenue per person.
Limitations
The model looks at each add-on on its own: whether two add-ons replace each other or are only bought together is not captured. Prices outside the tested range are an extrapolation, and the shares are choice shares within the survey, not a market forecast.
What 121surveys does for menu-based conjoint
Base plans and add-ons
Create base plans and add-ons with a description and a list of possible prices. If you like, participants can also choose “none” – just as in a real purchase.
Observed choice rates
Right after the first answers you see, per item and price, how often the survey offered it and how often it was chosen.
A model per plan and add-on
From 30 answered tasks, 121surveys estimates the choice between base plans (multinomial logit) and, for each add-on, the probability of purchase depending on price.
Simulator with revenue
Set prices and instantly see the shares of the base plans, the purchase rates of the add-ons and the expected revenue per person.
Frequently asked questions
When is menu-based conjoint a better fit than classic conjoint?
When customers combine things themselves in real life: software with add-on modules, plans with options, menus. Classic conjoint has people choose between finished offers.
How many base plans and add-ons make sense?
As many as you would show in your real offer – usually two to four base plans and a handful of add-ons. Every additional price level needs more answers so that each combination is offered often enough.
What does the simulator show?
For the prices you set: the expected shares per base plan and for “none”, the purchase rates of the add-ons and the revenue per person. That lets you compare pricing models before you launch them.
Does the model detect whether add-ons replace each other?
No, it looks at every add-on on its own. Whether two add-ons replace each other or are only bought together shows up in the observed choices, not in the model.
What does a menu-based conjoint study with 121surveys cost?
Setting up the design and running through it with the test link is free. To publish it you need a paid plan; menu-based conjoint including the simulator is part of every one. What responses cost is on the pricing page.
Test your modular offer
Create a menu-based conjoint and run through it for free with the test link.
