Would you use your own products and services? A critical lesson from Slack’s Stewart Butterfield

Reading Time: 3 minutes


Tanya Newhouse, CEO at ClevertarWritten by Tanya Newhouse
CEO at Clevertar

One of my favourite podcasts is an interview with Slack founder Stewart Butterfield, in a series called ‘How I Built This with Guy Raz’. Within a captivating hour-long conversation, Butterfield describes how he created both Flickr and Slack from ideas that dramatically deviated from the business’s original intention, but which created huge success by following customers’ interests and requests, even if these weren’t planned for.

Of course, this is the aspiration of any founder, but the interview is great for many reasons. Butterfield shares his doubts and vulnerabilities, and how he made decisions to pivot with his back to the wall. He talks about going home and crying while lying in the foetal position, as his world shifts beneath him, and I think we can all relate to that.

Our team are Slack users too, and so I was fascinated to learn that the messaging app was originally created as an internal tool for Tiny Speck, a game developer and Slack’s precursor. The tool was so useful for internal communication that the team decided to commercialise this software rather than the online game they had begun building, when it became clear that it was heading for failure. Fortunately, the move to Slack works, and Butterfield’s achieves success. His authenticity, tenacity and courage are an inspiration.

Now, perhaps the Slack origin story has become rather glossy as history is reworked for mass consumption (as many founding stories are) but I do enjoy the idea of using one’s own product before releasing it to the market, and learning from the experience.

As you may have read in my last post, Clevertar recently employed new staff and we’re delighted to be expanding our team to support new virtual agent installations in a range of use cases.

What you may not know, however, is how we use our own Clevertar virtual agents as part of the recruitment process itself. At the suggestion of our partner Harrison McMillan, our team created a virtual interviewer called ‘Sophia’ to ask candidates questions to help qualify, vet and recommend them for interview.

Applicants were sent a link to Sophia who asked questions about their background, interests and skills, all within a short, standardised interview that could be easily scored and results objectively compared. The process was straightforward, and it worked.

But like most successes, we didn’t do this on our own. We were fortunate to have subject-matter experts to help craft the interview questions, a partnership model we use when developing Virtual Agents with our customers too.

For example, our team is small and agile, which suits some people and not others, so Harrison McMillan suggested the following question to help us identify those with the best fit, along with a rating:

Question: What type of work environment do you enjoy working in?

Question Rating
Fast pace, dynamic environment with high energy, ever changing 1
I like to work in a quite environment, following a plan or process in a logical manner 0
I enjoy working with people, I like to have a plan but if urgent priorities come in, I am happy to change my focus 2
I enjoy working for a large corporate business with lots of people 0

There are no right or wrong answers – just what works best for us at Clevertar.

Sophia also asked applicants:

  • ‘Why do you want to work with us?’ and
  • ‘What is your experience with virtual agents like me?’

This allowed us to hear the person’s ideas and approach too. It was a valuable, and efficient way of qualifying candidates and I look forward to next time.

So what can we learn from this and Slack’s success story?

  • They built a tool for themselves. As technologists, we should be building technology that we actually need as well. And then once it’s proven, you can realise it’s commercial potential elsewhere. We built Sophia because needed her to help scale our team and automate the recruitment process. If we need her, perhaps others do too!
  • Everything we do should be customer-orientated. Slack’s rapid growth is testament, at least in part, to the company’s dedication to listening to its customers and building a solution that fixes problems. What do clients need help with right now and how can your technology or business fill the gap?
  • You’ve got to know when to pivot – are you married to unpopular features and offerings that aren’t commercially popular? Time to re-consider.

So there you have it. We’re not quite Slack, but using your own product and tech in new ways was an enjoyable and rewarding experience for our team, and one that we hope to build on as new an interesting use cases arise out of need.

Ref: https://www.npr.org/2018/07/27/633164558/slack-flickr-stewart-butterfield