Link the attendees and the decision-makers @ PyCon TW (part I)
This is the second year since I joined the preparation team of PyCon Taiwan, an energetic and vigorous community composed of Python enthusiasts. As a member of the newly born data-driven strategy team, the following is my tasks and duties:
- Build key metric from brand value propositions
- Design the questionnaire and plan the release channel
I spend most of the effort to figure out what do we really care about, and how could we got the answer by a short interaction with our attendees. In this article, I will share how I think and execute the tasks above.
Backtrace Thinking
I am a big fan of First Principle Thinking, which indicates that we need to focus on the essence of the target. It can help me get rid of the framework of constant comparison, place myself on the origin of the coordinates, and face the requirements directly.
A questionnaire is an efficient way to make our team obtain the feedback from our attendees, but we do not have enough time to talk with each of them, so we sort out all questions into a list and send it to them, then hope they are willing to finish it.
But what is the essence of the questionnaire?
The essence of the questionnaire is the question inside the mind of its designer, they wanna get the answer from the respondents. In my case, it’s the decision-makers of PyCon Taiwan.
What is the essence of the questions in the mind of decision-makers?
Why does a decision-maker will pop up question marks in their mind? Because they need to make decisions and inspect results depends on their cognition, whenever they reach something unclear, they will need to do some hypothesis and look for the answer from the TAs.
What is the essence of the decisions?
The community is composed of people, each member has it’s own brand value proposition, it implies that what they wanna bring to the attendee during the conference. We achieve that via make decisions on all the parts and details, just like a physical projection of abstract proposition in the real world.
That’s it! It’s pure enough for me to think out of the framework, and I think I and not able to dig deeper.
The essence of the questionnaire is to check how our brand value proposition achieves during the conference.
The Derivative Process
- I want to bring the attendee some value, so I join the preparation team, make the decision for the conference, and take responsibility.
- I encounter something uncertain while making decisions, so I do some hypotheses and hope it could be verified in the future.
- I need the feedback from the attendees to answer my questions during making the decisions.
- I do not have enough time to chat with them one by one, so I sort out a questionnaire and release it.
Some Specific Example
I hope PyCon Taiwan can help attendees enjoy the characteristics of each city, so I decide to host the conference in Tainan, instead of Taipei in the past few years. I not sure whether this can make our attendees enjoy the characteristics of Tainan, so I ask them to evaluate the foods in Tainan within the questionnaire.
I hope PyCon Taiwan can bring attendees various knowledge via a highly diverse agenda, so I give positive feedback to proposals of soft power. I am not sure whether this can make the attendees feel the increase in diversity, so I ask them to score for it within the questionnaire
Metrics
The value proposition is an extremely abstract concept, it’s too hard to say whether we achieved that successfully. On the surface, everyone has gained something from the event and excited to share what he/she has seen.
We can not assert “The attendees enjoy the food in Tainan” objectively just because we host the conference in Tainan, which is too subjective and the answers possibly be different between each member. We need to collect the evidence from the attendees and prove that “They really enjoy the food in Tainan”.
It sounds reasonable, but what evidence should we focus on? I tend to think towards the essence…
What is the difference between an attendee who enjoys food in Tainan and others?
If we can know how many attendees show the behavior of enjoying food, it is the evidence for “The attendees enjoy the food in Tainan”.
Interpretation
However, problems and indicators cannot be simply mapped in many cases, it is easy for people to think in different directions if your question is not specific enough. If we ask “Are you enjoy the food in Tainan?”, what would the attendees think?
- McDonald’s food is delicious, so my answer is YES.
- The beef soup of the vendor near the PyCon Taiwan Venue is a little too sweet, so my answer is NO.
In the first case, actually they can find McDonald's in almost every city, not only in Tainan, and it’s also not the dishes we expect our attendees to enjoy. In the second case, the attendee really tries the famous vendor, this is actually what we like to see, although he gives negative feedback because of the unpleasant experience.
Concrete
To avoid those unexpected biases, we need to make the question more concrete for the attendees.
What local delicates did you eat during the conference?
A. beef soup
B. fried noodles with eel
C. steamed egg rice
D. other
As far as an attendee who has tried the delicacies in Tainan is concerned, he will respond to the topic, and we can take it as positive results. It’s not the point whether they satisfy or not, everyone has different preferences, but we care about if they are willing to eat.
An abstract metric may cause the corresponding question too hard to answer, and cause problems with different interpretations, this could be improved by making it more specific. Reliable, objective and high-quality data are more critical than an answer to the value proposition directly while building metrics.
To Be Continued
The story has been a bit long so far, and the rest is left to the second article. Because some plans may be adjusted until the conference, I can also write more leisurely after the event.
