Business Technology

Monday 19 November 2018

Startup diary: Grab the chance to build a fresh culture which will sustain your new venture

Stock image
Stock image

Richard Rodger

A startup is not just an opportunity to bring a new product into the world, it is also an opportunity to design a new organisation. The design of your organisation, and the culture you establish in the early days, has a profound effect on the success of your company.

There are many paths to success, but far more paths lead to failure. The first rule of startups is 'don't die'. If you allow your culture to just happen, you are not controlling a variable in the survival equation. In the early days, especially if you have a large founding team, you can get by without deliberate decisions about your culture and the nature of your organisation, but you will regret it later. Your productivity will be affected as you try to retroactively repair unwanted behaviours that have crept in.

A company is an organisation designed for productive output, and a startup is a company designed for growth. When growth takes off, you won't have time for anything other than fighting fires. At the very start of your company, before growth really takes off, you have a unique opportunity to put in place a few basic elements of cultural design. Get these right, and they will sustain you through the growth phase.

This is my reasoning for investing some of my time on this aspect of the business. I'm in the minority on this decision. Most startup advisers, investors, and successful entrepreneurs, would strongly prioritise other things, like finding the right product-market fit, above all else. But over-emphasising some tactics over others feels like survivorship bias to me. Some businesses are just going to succeed anyway - it's a statistical certainty.

The business book 'Good to Great' by Jim Collins, published in 2001, is a good example. It attempted to identify what great companies did right, the idea being that you could copy them. But if you had invested in the 11 firms covered by the book, only two would have outperformed the S&P 500 index in the time since the book was published.

A much better strategy is to identify things that are very likely to kill you, and not do them. This gives you capacity to make mistakes on your way to finding and sustaining growth. A dysfunctional culture will definitely make you more likely to die.

I'm writing these articles as part of a cultural strategy of radical openness. We'll try to be as open with the world as we can be. But internally we also need to be open with each other.

Pick a colleague at random in your workplace. Do you know what their goals and tasks are? Do you know how they are being measured by their manager? Do you know what legal and compliance constraints they work under? Do you know their preferred style of communication? Are they an introvert or extrovert?

You may pick up a random selection of answers to these questions from your immediate colleagues by working with them over time, but wouldn't it be better if you just knew these facts about your co-workers already?

You've probably discovered most of them by accident anyway by first engaging in an unproductive way. Or think of the many ongoing communication failures in your workplace. Perhaps it would help to acknowledge that people actually are different from each other!

The management consultancy McKinsey & Company addresses this problem by making everyone take personality tests. Project teams are designed to balance personality types, so that they can be more productive.

I want to extend this idea, and make it less restrictive. In metsitaba, your goals, tasks and targets are public knowledge, so everybody understands the measures that drive your decisions in work. You also get to define what you need and expect from colleagues in terms of communication style, availability and working relationships. In particular, as an organisation dependent on creativity, understanding and accepting neuro-diversity is vital.

As a remote working company we have no choice but to share this information about each other in a structured way. Therefore, I am introducing a detailed 'playbook' for everybody in the company, to be shared with all colleagues. This is a commitment to internal openness that is an element of the cultural design of the company. As always, decisions have consequences, and I will let you know how it goes.

On the email newsletter front, this week we have 144 subscribers, and an open rate of 25pc. The production of the newsletter was delayed due to Halloween and I did not send it out until Friday, which I hope is the cause of the lower open rate (last week 32pc of recipients read the mail).

Normally I mail it out at around 6pm on Wednesday. I am still very much producing the newsletter by a manual process, and thus I am on the critical path. If anything affects my time, like a bank holiday, it has knock-on effects. It's time to start building a production process for the newsletter - building up the team's ability to deliver it without me. We need to turn the newsletter into a product that we can deliver reliably. This is something a startup needs to be good at, because you have to do it again and again. Next week I'll discuss the first steps to make this happen.

Just in case you missed it in the paragraph above, I did not solve the problem by working on the bank holiday. This is not my first startup, and in earlier times my standard problem solving technique was indeed 'just work harder'. I still believe in hard work, and you can be sure I'm putting in a lot more than 40 hours a week on this startup. But burnout is real. I've been there. It means months of bad decisions, poor personal relationships and a lack of real progress. What works for me now is a simple tactic: work as hard as you like during the week, but don't work at the weekend (or bank holidays). This gives you pacing and rhythm and preserves your decision-making capability, which is the most precious resource in any startup.

Consider also that if I had worked on the bank holiday, I would not have experienced the pain of a late newsletter delivery. The need to move newsletter production to a higher level of efficiency would not be so keenly felt. Working longer hours is a great way to avoid thinking and taking decisions. It's also a great way to burn through all your cash doing things that don't move you forward. Be very careful how you spend your time, because when the cash is gone, it's gone.

I have set a target of 500 subscribers by the end of the year. There are six working weeks left, which means I need to get 60 new subscribers a week to reach my target. I have switched away from online ads on reddit.com, which were a bust, to the tactic of sending personal emails to conference speakers. This really does not scale, and is slow work, but I need to let this tactic run its course too, so that I can get a reliable sense of how effective it is. I'll report back on the numbers for this next week.

The reason for setting the target at 500 subscribers to a newsletter for tech speakers was to validate the hypothesis that we should focus on speakers first for the minimum viable product (MVP). We are about to start building the MVP and cannot ignore the fact that this target now looks unlikely. It is already time for that most tedious of startup clichés, the 'pivot'? A startup should pivot to a new business model when it has no traction, when nobody is buying what it makes.

Like so many decisions in business, we must make this one without good information. Should the MVP remain focused on speakers?

Direct feedback from readers of the newsletter has been very positive. We have 144 subscribers with almost no promotion, apart from a very small online ad campaign, some infrequent irregular tweets, and a few LinkedIn posts. I have been doing a lot of other customer discovery work, mostly meetings with people in the event and tech industries, and have good indications of current business needs when it comes to speakers.

But we don't have a slam-dunk. More data is not going to magically appear. It is time to make a decision.

I'm going to sit down and analyse what I have, and design an MVP feature set based on the data at hand. Next week I'll walk you through that analysis, and tell you what we are going to build.

Indo Business

Business Newsletter

Read the leading stories from the world of Business.

Also in Business