Zapier is Love. Zapier is Hate.

I was asked recently via a LinkedIn poll “what’s your favorite middleware tool for Martech data?” and there was much love for Zapier in the votes and comments. Even though it has massive shortcomings.

I decided to lay out some of my my pains and experience below.

When you update a value or create a new object in a Zap, how do you communicate the change to the rest of the team? How do you know it doesn’t break another Zap? Or worse, an automation/workflow in your CRM that your teams depend on?

The answer is, you don’t.

529e72dc-3a02-4f7a-8097-6316c81ce85b.gif

This is why I hate Zapier.

As soon as you update more than one field on 2+ objects across many apps, you know you will run into issues down the road. It is near impossible to manage or maintain it for a longer period of time. Other tools like Workato or Tray.io don’t do much better either since all of these tools are blind to what happens downstream… But these tools still have some benefits, like easily triggering notifications in Slack. 

For now, we are all stuck with them so here are 2 recommendations I learned the hard way that makes this mess more manageable:

  • Create a Changelog. Create a Slack channel named “#Ops-Log”. Every time you change something in Zapier (or any other tools!) write a quick note in the channel. At least, when something breaks, you can go back to the channel and see recent changes. Bonus, this improves visibility and communication in the team!

  • Increase Error Visibility. In Zapier, create a Meta-Zap, which is a Zap that gets triggered when another zap fails. to do so, use Zapier’s "New Zap Error (Legacy) trigger <https://zapier.com/apps/zapier/integrations>". You then send an alert msg to an “#Ops-Alert” channel. Add alerts from other systems too (Salesforce, Zendesk, etc) to give you more visibility in your system.

Now stop reading this and go implement both of these tips now. It should take you less than an hour. What’s stopping you? Go! 

Next
Next

Analytics Engineers: The new Role that owns the Modern Data Stack