Part 1: Deep customer empathy
The state of understanding of the developer partners' needs outside of the API was sparse and undocumented. We couldn't charge forward because we didn't know who we were serving on a qualitative, empathetic level.
Taking a step back and finding the "Jobs to be Done"
We already had quantitative data on our partners, but little qualitative. The method I chose was a "Jobs to be Done" style interview with a variety of partner customers, believe it would give me deep, high-resolution insights through the human interaction and conversational exploration. A "thick data" (deep with few) approach.

The interview format was adapted from the JTBD "switch" interview. The questions gave room for the participants to go deep into their motivations, emphasizing exploration and introspection. Some questions from the interview:
When do you first remember considering becoming an Intuit Developer Platform partner? What prompted this consideration?
What problem did you see being solved by this partnership? And, has the partnership fulfilled that promise?
Was there a “last straw” moment you remember that gave you the final nudge to engage with Intuit?
Before you experienced the relationship Intuit, what do you remember what you hoped you would end up with? And having experienced it, how does the reality compare to the hope?
The interviews were meaningful conversations that allowed for the deeper motivations behind wanting to use our platform and API to be revealed. After finding the overlap between interviews, three key findings rose to the surface:
- Partners place a heavy emphasis on the relationship between them and us, believing an active relationship to be critical to their success.
- They believe they require qualitative guidance and quantitative feedback for their integrations to excel in the platform ecosystem.
- The widespread brand-equity of Intuit is one of the biggest attractors.
Armed with this clear, but broad, direction on the deeper needs of our customers, it was time to narrow in on specifics. For that, I turned to a co-creative activity in Mural where the customers would help design the solution themselves.
Part 2: Go broad then go narrow
With the assistance of our Partner Management and Developer Platform stakeholders, we collected a broad list of benefits Intuit could offer to our partner customers. The list was of things we already do and things we could potentially do but aren't.
Finding resonance through co-creation
I had the idea of using Mural as a co-creative space for the customer to design the tiered benefit program alongside us. An activity was created where 9 general categories of benefit would be stack-ranked by importance to the participant, and then the individual benefits would be sorted into tiers based on the participant's input.
The activity started with this Mural of the 9 categories and 80~ benefits, with the invitation to add ad-hoc as needed:

We wondered if sorting nearly 100 items over the course of an hour could be done. Our participants proved that not only could it be done, but that by having them drive the activity they were able to reach a confident end-state in an average of 40 minutes!
The process looked like this (this is a 36 minute time-warp):
The results were surprising and pleasing
We conducted 20 of the activities, allowing a wide variety of partners types (company size, maturity, industry) to each contribute their own vision of what a meaningful tiered partner program looked like to them.

There was a concern that the output would be biased toward customers wanting everything available in the entry tier, but participants were all cognizant of the downsides of asking for something unreasonable vs. something realistic, even if they moved benefits they wanted into tiers they currently did not qualify for based on the model.

Pivot point: Interviewees adding new groupings
After a few sessions, we found that interview participants kept adding their own ad-hoc items that all had to do with performance analytics of their intergrations. We added this new category and options for the remainder of the sessions and it was one of the most used. A blind spot was identified and remedied.
Adding a layer of quantitative data via large-scale survey
To help bolster the quantitative value of the customer input, we also sent a version of the activity to several thousand partners via a survey. The aggregated survey response further validated our hypothesis about partner benefit and gave us a solid foundation of data to move forward.
