Data challenges

Challenges

Network effect

Problem description:

On the one hand, we want our data to be reusable between different retailers. This is why allowing them to create their own questions might be a bad idea, since in this case the data will be messy. Moreover, retailers may create bad quizzes which spoil customer experience, and this might negatively impact our brand.

On the other hand, prohibiting retailers to set their own questions might have a negative impact on sales, because retailers likely want to have a freedom in their actions.

Question:

How can we solve this dilemma? Ease of sales + the reusing of data between different retailers + great customer experience?

Thoughts:

TBD

Product tagging

Problem description:

We need to ease customer onboarding as much as possible. Product annotation might be the most time-consuming part of it.

Question:

It would be great if we could come up with a product tagging solution based on ML

Thoughts:

TBD

Not go into this space rn.

0-1000 skus =manual

algos for 1k+ SKUs

Product tagging part of item: e.g. charge per sku 20c per sku

An additional revenue stream -

Need-based attributes

Problem description:

Some filters are need-based

Question:

How can we understand the relevance to a need-based filter based on product features?

Thoughts:

TBD

Saagar TBC

Data collection

Problem description:

We’ll need to create a user-centric recommendation system. However, it might be hard because of:

1) 3rd-party cookie-death - we might not have a good way to identify users between different sites

2) not enough relevant data/messy data

Question:

1) How could we tackle a 3rd-party-cokkie death problem?

2)What data do we need to use in our recommender system?

Thoughts:

TBD

Want to print your doc?
This is not the way.

Try clicking the ⋯ next to your doc name or using a keyboard shortcut (

CtrlP

) instead.