Building a Custom Framework

A framework is the rubric Backdrop scores your calls against. It is a list of elements — Metrics, Economic Buyer, and so on — and, for each one, the evidence a call has to contain before that element counts as covered. What the elements grade is up to you: MEDDPICC grades deal health, and a framework you write can grade anything a call should have proved.

MEDDPICC ships built in and is what you get until you change it. If your calls turn on something MEDDPICC never asks about — a security review, a procurement gate, a champion who has never bought software before — you can build your own framework and have Backdrop score your calls against that instead.

Every Backdrop user can build one and choose which framework scores their own calls. You write it by answering a few questions about how your deals go wrong, and Backdrop writes the elements from your answers.

Your frameworks

Select Frameworks in the sidebar. The list is the built-in frameworks plus every framework you have written yourself — frameworks your teammates wrote stay on their own lists.

Select any row to walk its rubric one element at a time. Built-in frameworks are read-only — the rubric reads the same, but nothing in it can be edited. Duplicate one from the ⋯ menu and the copy is yours to change, or build a new framework from it, which is what the rest of this page covers. The ⋯ menu also carries Rename and Delete framework for the frameworks you wrote.

Choose what scores your calls

The star beside each name is the one that matters: it marks the framework your calls are scored against, and exactly one framework can hold it.

Select the star on any row to move it there, and your next call is scored against that rubric. A framework needs at least one element before it can take the star.

Talk: pick what it's built on

Select New framework and the builder opens on its first question: what the framework should be based on.

Each tile shows the framework's element badges and how many elements it has. Pick a built-in framework, one of your own (marked Yours), or Nothing — an empty framework built only out of what you say.

The builder runs in four phases, tracked across the top of the window: Talk, Building, Review, and Done.

Talk: answer the questions

Backdrop then asks about your deals — where they stall, what a good call actually proves, which deal shapes you'd rather walk away from. Answer in your own words, in plain prose. Specifics are what make the difference: the questions are there to find the failure your current rubric doesn't grade.

Each answer gets a collapsed Thinking line above the reply. Expand it to see what Backdrop took from your answer and which element it decided that belongs to — worth a look if a later element is not what you expected.

Building: it writes the rubric

When the interview is done, Backdrop plans the changes and then writes the elements one at a time, with a counter of how many are finished.

Elements it decided not to touch are copied in as they were. Elements it invented from your answers — here a Security review readiness element that MEDDPICC has no equivalent for — appear alongside them. Next turns on once the rubric is whole.

Review: read it and fix it

Review is the rubric as it will score your calls. The rail on the left is every element; select one to read it.

Each element is made of the same parts:

• Assessment — what Backdrop is asked to judge on this element.
• Definition — what the element means, and what it explicitly is not.
• Graded attributes — the table that does the work. Every attribute is graded across the same three bands: Fully, Partially, and Fails, each spelled out as the evidence that band requires.
• Scoring bands — what a score out of 10 means for this element.
• Anti-patterns — the mistakes that should not earn credit, such as counting a seller-supplied number as buyer evidence.
• Post-call question — the question this element answers on the post-call scorecard.

Some elements also carry optional sections: a litmus constraint and its tests, role disambiguation, a framework caveat, or extra constraints.

Everything here is editable in place: select any field to rewrite it, drag elements in the rail to change their order, and use + at the bottom of the rail to write an element of your own. Two ways to change it:

• Edit the text yourself, field by field.
• Select Refine with AI and say what is wrong with it — an element reps skip, a score that turned out to mean nothing, a deal it let through. Backdrop asks from there and rewrites or adds the elements you agree on.

Done: put it to work

Select Save when the rubric reads right. Your edits and the element order are written, and the framework joins your Frameworks list.

Select its star there and your calls are scored against your elements from then on: the same post-call scorecard, reading your rubric instead of MEDDPICC.

---

Next step

See Post-call briefing and MEDDPICC scoring for how a framework's elements are scored and read after a call.