Day 550: browserflow + zapier

product

the questions i’ve been struggling with over the past few months (i mean, basically ever since i started thinking about how to make a business out of browserflow) have been the existential ones:

  • who is my target customer?
  • what problem am i solving?

i initially planned it to be non-technical people but learned that being able to build flows often requires being able to think like a programmer (knowing to put a conditional on a Click command to handle the case where the element being clicked might not exist, etc.) and being familiar with certain concepts (variables, CSS selectors, etc.). in other words, the learning curve is too steep for most casual users to build their own flows, so i’d need to get technical users to build flows that non-technical users can benefit from.

so the plan thus far has been to target developers. browserflow makes it easy to specify what elements you’re interested in without having to figure out the selectors yourself in many cases, which is quite convenient. my current plan was to show off one use case (web scraping), market it a whole bunch, and see what other use cases people are interested in.

but there’s another approach that i’ve been getting more excited by lately: browserflow could be the solution for anything zapier doesn’t do.

zapier is the de-facto solution for automating business workflows, but it’s limited by the APIs exposed by their partner companies. you know what can be used to automate things that aren’t available via API? browser automation.

take, for example, facebook. many zapier users are interested in triggering a zap when a new member is added to a facebook group, but that’s currently not supported by zapier. what if i built a flow that synced new members to a spreadsheet? then zapier can be used to further process those users.

this makes the target customer super clear: zapier users who want an integration that doesn’t exist.

ways to find what integrations people want:

  • look through the feature requests in the zapier community
  • ask what people want on zapier facebook groups / subreddit
  • look at the list of top requested apps that aren’t integrated with zapier

what i’d need to build:

  • browserflow api so you can run a flow from a zap
  • webhook or command to make http request to run a zap from a flow

i really like this approach because:

  • these are users who are willing to pay for automation solutions (since they’re zapier customers)
  • i know where to find them (fb groups, subreddit, zapier forums)

the only “downside” is that since these will be nontechnical users, i’d have to be responsible for building and maintaining these flows, but if this approach works, i can just hire people to make sure popular flows continue to work

plan:

  • build the flow that syncs new facebook group members to a spreadsheet
  • try to find people willing to pay to use it
  • build out the features needed for full integration with zapier (api, webhooks/http command)