Startups move fast, they often grow very fast, they can crash just as fast.

More than one thing can kill you. You can grow too fast, working capital spikes, costs can spike, cash collections slow to a crawl.

Daily habits of tracking key metrics are essential. It’s valuable to keep a high tempo for the business but more importantly to provide an early warning system so you can spot when you have a problem before it becomes a crisis.

Get someone to build you a Daily Report, a few suggestions

• Cash on hand
• Cash received
• New Income,
• Hours billed,
• New users signed up,
• Total daily users
• Monthly run rate
• Monthly burn rate
• Website Visitors
• Average Revenue Per Customer
• Average Order Value

There are a bunch of options you need to decide what the critical KPIs are for your business.

Keep that Dashboard updated every day, emailed to you first thing in the morning so no matter where you are or what you are doing you know exactly what is going on, even better if its a live dashboard you can view on a computer or an App.

A daily dashboard is an early warning sign for trouble, you don’t have to wait a month or a quarter to find out there is a problem and can jump on it and fix it long before its a crisis.

Image courtesy of

http://webapplayers.com/homer_admin-v1.7/

http://webapplayers.com/homer_admin-v1.7/

I originally wrote this post over a year ago but got new material a few months back so its a mix of four experiences.

There is a fantastic scene in the movie Casino where Robert De Niro is walking through the lobby and a maintenance man comes up to him and says “I’m sorry sir I have done everything I can to get the pool clean but it just isn’t working”.

De Niro replies

It’s not your fault it’s mine I should have sacked you a long time ago

Brutal but true.

A few months ago I had to make a hard call on a freelancer who continually disappointed me over a period of 5 months with late and buggy code, disappearing for days from the chat room, constant broken promises and did I mention lots of buggy code.

Finally after a foolish logic mistake which resulted in his code needlessly continuously transferring over 1Tb of data in a few days. Not a good look and had the instance not crashed it could have been 10tb or more.

To confuse matters he had previously delivered working but not elegant code that seemed to be reliably processing large volumes of data (the TB data transfer was millions of transactions, sadly with some faulty logic) which made me think that maybe I had to put up with the bad in order to get the particular skills which were rare.

I have made this mistake in the past, I originally started writing this post about 9 months ago when another team member on another project left an accidental easter egg (some hidden surprise bug or unintended functionality) in some code that nearly killed a project completely. I actually wrote part of this story while sitting in the same room as them having a heated discussion with another team member.

Every time I kept an employee for too long, I knew there was an issue but I thought they were working their way out of it (or being coached out of it), but alas my initial concerns usually are proved right, often in the most critical times.

Its a trap. I worked at Hewlett-Packard for 4 years, the HP Way is a strong cultural guide to running a great business, however while its ok to believe everyone wants to do good work (and both these people wanted to) its not ok to believe that everyone can do good work at the level you need.

Anyone who has had to hire more than a few people has screwed up a hiring decision. It happen’s and it’s difficult to know who is going to work and who isn’t.

There are a bunch of techniques you can use to reduce the odds, but short of being entirely referral based recruiting where a trusted member of the team or someone known to you recommends the candidate its tough to get it right 100% of the time.

Referral recruiting is a great strategy for fast growth companies where you are ramping up the team, apparently most of the team from Youtube all went to University together and they kept recruiting their mates to fill the engineering teams which allowed them to grow very quickly in the first few years.

The problem is not the realisation that they your team member is not the right fit, its the wishful thinking it will get better and then not taking immediate action when you know there is a problem.

As one of my bosses used to say “You know its time to replace the CEO when you are even asking that question”.

So you know they are wrong for the job, so what are you going to do?

Most people just let it go, hope they leave, sometime manage them out but generally it’s hard unless they are a complete screwup.

When you are a startup this problem is doublesided.

Your ability to attract people to work for you is naturally limited, you haven’t got much cash, you can’t pay large salaries, by definition you are a risky prospect, in many cases you have happy with someone who can code and has a pulse.

The real issue is that the younger the company is, the more holding onto bad players hurts. The first 10 are so important, each individual has a huge influence on the company and your survival.

My advice, do a small work trial initially. Especially in Australia with our restrictive labour laws you cant afford to let non performers hang around after the 3 month mark.

Secondly but critically you have to get rid of them the moment you know, it never gets any better, the things that annoy you only get worse, just do it.

Rules are the Scar Tissue of Past Mistakes

I was talking to a few new entrepreneurs about how to manage the process of employing freelancers to run their projects and startups and thought I would list down the mistakes I have made and the methods and tools I use to try to make this a tighter process as it is very easy to have this come off the rails.

I always preferred to have my own team members on staff as employees, but when it comes to software development it’s very difficult to bootstrap an application or business using software developers based in Sydney or other western cities, its especially impossible if you are self funded using your wages.

And so there are hundreds of thousands of would be entrepreneurs hiring freelancers via Freelancer.com to code up projects in the hope that one of the experiments might turn into a real startup.

This makes an already difficult process (software development) extremely difficult. Add to that cultural, language and time zone differences and the opportunity for mischief and misadventure is especially high.

I have been hiring remote freelancers for various projects for myself and other companies since before Freelancer.com existed, must be close to 12 years now.

Over the years I have had probably 20 from India, Nepal, Pakistan, Russia, Philippines, USA and the UK.

It’s tempting to think that they require less due diligence than an employee, but nothing could be further from the truth.

Given your time in both the interview and daily face to face conversation is likely to be short compared to a normal employee it makes it critical to carefully manage the relationship, the description and delegation and completion of tasks.

Often a lot of this work is done via chat sessions, especially if you are working a day job, you won’t get to talk to the freelancer very often, this makes it very difficult to get work done.

Im still learning and still making mistakes but here is a few tips that have worked well for me.

Treat it like a Real Employment Process

Initial Chat sessions to shortlist a few candidates and follow it up with proper Skype interviews, use the same sort of time and effort you would for a real employee, its easy to think that you don’t have to but trust me you will waste a lot of time and rehire numerous times unless you do.

Hire Individuals

I hire individuals or teams of two only, in my opinion hiring an outsourcing firm is a recipe for disaster.

If you think its hard to get what you want dealing with a freelancer of a different nationality, culture and time zone that speaks English as a second language, try getting what you want done with 1-2 layers of project manager and sales people in between.

Not only will be it much more expensive to carry this overhead but its very likely your personal work overhead is likely to be much higher as well. Typically these firms want you to create a detailed specification much like a traditional waterfall life-cycle and then will lock you down to the specification and then when you have to change it (which you will) they end up billing you hourly anyway.

The truth of the matter is that you have a vision about what you want, but in the style of the Lean Startup there is a lot to be learnt.

In some cases you may be trying to create a technical solution which might use known parts but is doing some technically difficult.

Mistakes will be made, lessons will be learnt, its best to find individuals who are flexible and happy to work on an hourly basis because any fixed plan will not survive contact with either a customer or testing.

Ground Rules

Start with a set of non negotiable ground rules. Explain that you cannot employ contractors who do not continuously apply these and sack anyone who doesn’t. My rules are as follows

• Contractors will load Slack to their desktop and mobile device and will update the team daily and ask/answer questions as required
• Contractors will commit their work to Github on a regular basis. New feature or bug fix requires a commit
• Branches should be merged regularly
• Commit messages should be detailed and cover all changes to the code for that commit
• Contractors will use Trello to pick up their tasks and update them, ask any questions or make notes on the card
• Code should be deployable, meaning that you should be able to clone the code from Github master and have it run on your server (this is hard to get right but its essential if you are going to build a functional development team).
• Try to be in a similar time zone +/- 6 hrs, anything more means you will never be able to chat with them during your waking hours which will throw your development back days everytime they have a question and you are asleep.

Start with a Small Test

Don’t take your contractor on full time or make them a member of the team without a test.

Its a pain the in ass but you need to try to find a discrete task that is large enough to give you an idea of their work ethic and quality and how well they will communicate and follow your rules.

On-boarding

You need to get together a set of documents and a checklist of things to be done to get a new developer on board. At a minimum you need the following

• A diagram which describes your tech stack
• A description of key processes and the resources required to deliver them
• A checklist of systems you need to give them access to (and revoke when they leave)
• A Skype call to describe your vision and why you want to solve the problem and how they are going to help you deliver that (you want to get them excited about the project, its not enough that they are simply filling in time, they need to be engaged).

Systems

You can use what you prefer, a lot of freelancers are resistant to using systems, they are often don’t see the need, if they won’t play the game sack them immediately, it never gets better.

We use Trello for task management and assignment, Slack for Comms and Github for code version control.

Trello is pretty easy to use, we usually have 4 columns

• Ideas (not in Roadmap) to capture all the good ideas that you are not ready to work on yet,
• Roadmap what you are committed to building
• MVP the tasks required to build the Minimal Viable Product
• Doing
• Done

A note about Github, so many developers will tell you they can use Git and Github but they have no idea. You need to work this out asap as they will screw your development process if they don’t have a clue.

You can’t just pick Git up in an hour, its a steep learning curve, they shouldn’t be learning on your time if this is your requirement. Ask them to branch your repo and make some changes, commit them and make a pull request and see what happens, you either get a blank look or you get some messages from Github saying its happened.

I use Jira for tasks in my day job and its good for a large team with a complex development environment but its a lot of overhead for a small team and I find it a bit convoluted and the flow is not as easy as Trello, also have used Asana which is ok, again just prefer Trello’s boards and flow.

Slack is fantastic but you need to know when to pick up the phone or video conference and call someone, if you find yourself labouring and starting to get the shits with the person on the other end its time to call.

Daily Habits

You need to establish a time to chat via slack or Skype

You need to chase daily

You need to make them commit to Github regular

You need to make them show you the working product regularly because what happens is they don’t commit, they build whatever they think you wanted or whatever they think they can get away with, but its not what you actually wanted, weeks have gone by and you are pissed off, your money is gone, they are tired and you need to reboot the project. You need to see working code regularly and change course as required.

Hope this is helpful, if you have any great suggestions I would be happy to share them with the readers.

Too often I hear or receive pitches from entrepreneurs that waffle on endlessly without actually saying anything. If I don’t understand what you do after 30 seconds of burble I start to lose interest, Im sure most people are the same but I will actually tell you.

Blah blah blah, technology, blah, cloud, blah, new paradigm, blah, trendy business model, blah blah, push notifications, big data enabled, blah, IoT connected, blockchain powered blah blah, it does my head in.

Get to the point!!!

Its really simple (but hard to get right) just tell us who you are and what problem you are trying to solve and what customers you want to help and why.

Im Fred Musk from Flintstone Motors. We make a new type of car that’s powered with an engine.

Mums and Dads are tired of running with their feet to make their car go, they can’t carry the whole family and they can’t travel long distances.

They want a self powered car, but so far no one has been able to make one commercially until now.

This week we are launching the new Tyrannosaurus P85D.

It doesn’t need you to run with your feet to make it go.

It uses a new technology called an engine to make it go (which ironically burns dinosaurs).

Its much faster than a normal Flintstone car and your feet don’t get worn out.

And you arrive at your destination feeling refreshed.

We will replace regular cars because people are tired of running everywhere and its much more comfortable and can carry all the family across the country without stopping.

I encourage you to get your elevator pitch down to 5 floors not 40 floors, use the rest to answer questions.

Thanks to Hanna Barbara and Youtube

Following on from yesterdays topic of knowing your startup’s Runway raises the topic of how long it takes to either raise capital or sell a business.

I have mentored entrepreneurs for a few years now, I was a Entrepreneurs Organisation member (which I highly recommend) and spent a lot of time discussing and watching entrepreneurs trying to raise capital or sell their business.

I now mentor at a few of the local incubators as well so I get to talk to a lot of new startups every month.

Until you have done it or seen it done you have no idea how long it take to raise capital or sell a business and how stressful it is.

Many times I watched entrepreneurs operate with the mistaken belief they can close their fundraising or sell their business within a few months.

The truth is most startups are woefully underprepared or unsuitable for either raising capital or sale and it might take 6-12 months to get them prepared (thats a topic for another day).

Unless you have a Unicorn (billion dollar company) and everyone is fighting for your deal or you are in the unfortunate position where you are forced to fire sale your business due to insolvency or the 4 Ds, (Death, Disease, Divorce or Distress), most business sales or capital raising’s are going to take you 3 months minimum and depending on how bad your story is it could be much longer.

So realistically in the case of a capital raising you should start fund raising 5-12 months from the end of your runway (the range is deliberately wide so the time doesn’t sound prescriptive, different businesses in different locations will have vastly different times to raise capital, if you are a life sciences business in Australia or a hot SAAS company in the Valley the range may be even wider).

Given a lot of seed rounds usually only provide 6-18 months of runway you can draw your own conclusions but essentially from the day you take your first round you should realise you are on the campaign trail continuously, it can’t be a last minute thing.

I recently had a young entrepreneur come and see me who had been running his startup for a year on the smell of an oily rag and had got to the point where everyone was financially exhausted and they were desperate to get funding.

He only had about 8 weeks of runway left when he spoke to me and I indicated he was probably not going to make it in time and while he had built a great product and was starting to get some customers it wasn’t clear he had really found a working product market fit.

My advice to him was the only way he was going to make it was to extend his runway himself.

As one of my old HP Managers said to us during a downturn “The only way you will get yourselves out of this is to sell your way out”.

I told him he needed to get out and sell the product directly to his end users (it was a SAAS product) and in my opinion he should be focused on nothing but sales for the next 90 days.

If he was successful this would result in two things, first his runway would get longer from the revenue gained and secondly he made his company an investable prospect by demonstrating customers were wiling to pay.

In the end he split his time and only put a day a week into sales, managed to bring in some revenue but not enough and 8 weeks later he had to abandon the capital raise (despite coming close and getting commitments) and has put the business on the back burner and gone to work for a high profile startup.

I’m sad I was proven right, I thought the business had a lot of promise, I don’t bring it up to be right, I want people to understand that you can’t wait until the last 8 weeks to start raising capital or selling your business.

The second lesson, if your runway is short and you have a product which is close to market ready that will produce revenue from launch (SAAS, SME, Enterprise or a consumer subscription product), in my opinion you should spend your time on sales calls instead of investor calls.

Closing the sale will help close the investors and if you are lucky and skilful maybe reduce the need for them at all.

In most cases 3 months would be a fast result to get a term sheet, DD, all the paperwork done, waivers signed and money in the bank, expect it to take longer, in some cases much longer.