Prototyping with Power Tools

Being a product manager in a company of talented designers and engineers is a special kind of frustration: you can see how every product could and should be better, but you can’t say you’d design or build it better yourself. Last month that frustration finally boiled over for me, and I rebuilt our sales CRM in a weekend. Well, kind of rebuilt.

I’m part of the sales team at Postlight, which means that twice a week we gather in the “pipeline meeting” to talk about all of our potential new clients. Until recently, we used a woefully overpowered CRM app (CRM stands for “Customer Relationship Management,” which is a fancy way to say “lead tracker”). All we needed it to do were three things:

1. A simple way to keep lists: of people; companies; and projects;
2. A way to link them together, and;
3. A way to view what we were working on at any given time.

Our CRM would do these things, sort of. It had tools for just about everything, and was materially designed to a fault. (We don’t want to name it because it’s no fun to beat up on SaaS products from smallish teams.) It could automatically import new leads by creepily reading your email, which made everyone uncomfortable. It could display and track individuals, companies, and projects with a never-ending supply of customizable fields, and create reports galore. We rarely used more than a third of the app — mostly we just looked at the kanban board view (if you’re not familiar with a kanban board, think of a Trello board).

Paul Ford is a co-founder at Postlight and he found the CRM so frustrating that one afternoon he wrote a small, angry treatise in Slack on the CRM app’s use of checkboxes/multiselects.

Every week we struggled with doing something that would, in any other tool, have been easy — horizontal scrolling in the kanban view was especially atrocious. Finally I decided that waiting for the company to fix the product wasn’t going to cut it — I had to fix it myself.

For other Postlight projects I often have a designer or engineer colleague I can rely on to help me build a great product. This time I had to fly solo. I decided to start simple and see what I could prototype myself, using my favorite tools. I exported our information from the CRM to get some workable data, fired up Google Sheets, and got cracking.

I started by building a simple view of all of the people and companies. Rows and columns galore, with some clever filling and filtering to make something presentable. I added in a space to track interactions, so we could keep track of if we had invited someone to one of our Postlight Sessions, or sent them the Practice book (they’re still available if you want one). I toyed around with Zapier and Mixpanel to automatically track every time we sent someone a Track Changes post. Things were flying, and it was still only Saturday afternoon.

Then I got to “projects,” and hit a wall. Google Sheets is great for managing data, but not for displaying it. Unless I could make a view as good or better than what our CRM had, we were never going to switch. Poking around, looking for open source self-hosted Trello clones I could hack to also house our data, I remembered Airtable.

If you haven’t used it, Airtable is amazing. The short version is that it’s a souped-up Google Sheets, combined with Google Forms and Trello. The long version is better explained by them:

In one quick copy/paste I was able to get all of our people and company data into an Airtable base, and it was a breeze to link up people, companies, interactions, projects, and internal team members (so we know who is responsible for what). For the projects view I used Airtable’s kanban view, customized what fields showed, and bam!

I wasn’t sure how well my pasted-together prototype would go over with the team. I didn’t want to rock the boat, and maybe everyone else secretly liked the tool we were already using and were just complaining. But after one of our meetings, with some trepidation, I demoed the Airtable alternative to our souped-up CRM.

I had copied all the data that we used in the meeting, so it looked very similar to before, but better. It had fewer features, but the ones we needed. It couldn’t run reports, but we had never run a report. It didn’t have the ability to assign tasks to team members, but commenting on a project and tagging a user still sent them an email, which is all we needed. (Everyone prefers to use their todo list app of choice anyway.) I got the best reaction to a demo I’ve ever gotten: Looking at the screen, the group reaction was “Well, yeah, duh. This is what we should use.”

There was no driving need for ditching the old CRM. It was bad, but it did its job. This was mostly an exercise in showing how much I, as a non-technical P.M. in a company of engineers and designers, could contribute with just the tools available to me. Still, the old CRM was expensive, so we did move, and we even saved some money in the process (though not quite as much as Adam did). And we still have all the features we need.

Our old CRM tried to please everyone with lots of features, and it ended up driving us away. It turns out that if you’re going to build a product, sometimes you don’t need anything more than a power user tool like Airtable.

—

Cody Cowan is a product manager at Postlight.