His idea is quite unique – in that it is meant to provoke – and therefore his business requires a deeper focus on the legal structure than what might otherwise be required.  It was the structure that was the subject of our conversation and as I’m not a legal expert, I had no hesitation in referring him to my excellent solicitor who is very adapt in setting up start ups and has helped me greatly in the past.

After our lengthy conversation about the benefits of a trust/company structure he went on his way, brimming with enthusiasm and ideas and armed with a lead worth it’s weight in gold – a referral.

But what makes a good referral?

This experience showed me clearly that you don’t need to be a subject matter expert in a field to provide a good referral.  I have very little idea in corporate trust structures, but that didn’t seem to matter.  My good friend Ms Polka is a mastermind of finding things on the internet.  I know without doubt, that if I ask her for the best way to do this, or the best person to seek for that, she will be pretty darn close to spot on.  I value her referrals more than most because I know she has put the effort in to research it for her self.  So is it trust?

We generally trust our friends to give us good referrals, and I think I know why.  When you’re referring something to a friend, you’re putting your own reputation behind the product or service you’re recommending.  If your friend has a bad experience, you absolve some of the responsibility.  So you’re less likely going to want to refer something to a friend without knowing absolutely that it will work for them.

I once recommended a service to a friend which at the time I was having good experience with.  However the business changed ownership and customer focus became a low priority.  I’m now too embarrassed to even mention it to them, as I know they’re having just as much problems as me.

So we know that a referral is worth it’s weight in gold and that if your mate tells you to use this product, you’ll be much more likely to give it a go.  So how can we influence this when promoting our own product or service?

Without question, you need to have an awesome product or service to start with.  Once you’ve got that squared away, you can look at ways to influence the referral.

There are a number of independent third party websites that are in the business of referrals.  They provide a means for people to recommend, or warn against, the use of a product or service.  There are industry specific sites which allow reviews to be written, as well as generic service orientated websites.

Of course, you could join the leagues of Affiliates that allow you to earn cash by putting up a link on your own site.  However, I don’t value these the same as referrals, no, a true referral needs to be personal.

The referral is a big part of the AutoCarLog growth strategy.  It’s a way for the business to grow quickly and cheaply, and I’ve called it “Mooching off your mates”.

I still need my customers to love using the service so they don’t get embarrassed.  The “mooch” system simply provides an incentive to speak out about it.  I’m gathering the hordes of otherwise contently quiet customers and turning them into trumpeting voices of reference.  How? By giving the product away.  If you sign up a few of your mates, you can use it for free.

I’m rewarding people for helping me out.  It still needs to be a good service, but I’m providing an easy means to share the love and giving a little bit back as thanks in return.

• Day one – The Idea
• Day two – The Setup
• Day three – The Pitch
• Day four – The Function
• Day five – The Graphics
• Day six – The Gateway
• Day seven – The Launch

Synopsis

One week ago I set myself a challenge to build a business in seven days and for less than $500.  Today at 1.17pm, AutoCarLog, the product of that challenge, was launched.

Over the last week I had come up with and researched an idea, registered a bank account, credit card gateway, setup secure website and domain name, developed a marketing strategy, built a fully functional back end and database, designed the interface with a focus on the user experience, and built my very own SMS gateway.

The launch itself consisted of uploading the signup page, sending an email to everyone who had pre-registered and posting an update on Twitter.  I must admit, as soon as it was launched, I found myself refreshing my database admin page every few seconds to see if and when I had my first customer.

After 15 minutes AutoCarlog had its first customer! (It was my mate, who had problems with his paperback logbook and inspired the idea.)

Motivation

There were two motivations for doing this challenge; to inspire others and to see if it could be done.

Can it be done?

The challenge was to build a business in seven days and for less than $500. So can it be done? Yes. Just.

To reduce costs AutoCarLog used my existing company structure, but if you’re starting from scratch you can operate under your own name.  If you didn’t want to operate under your own name you would need to factor in the cost of registering one ($110-$160 depending on state).

I also did all the software development, which saved further costs.  If you haven’t built a website before you could teach yourself (and sacrifice time) or hire a third party developer (and take a hit on cost).  Alternatively you could negotiate a partnership with a developer and graphic designer for a share of the business.  (I might even be interested!)

Another thing this challenge taught me was to think outside the box, particularly with the SMS gateway.  If money wasn’t a tight constraint, normally I would have signed up with an aggregator.  However this challenge forced me to look at alternative ways, which is how the idea to use a Mobile phone and PC was born and why I eventually built my own hardware SMS gateway.

Inspiration

The best thing about this challenge was hearing about the ideas it inspired others.  It is such a great feeling to be able to help out other people and here is some of their comments I received throughout the week:

From Manny (on semi-blog.com)

I found the info you provided VERY useful for my own website, which will need an SMS gateway! Thanks for the links you posted!

From Royce Fullerton (on semi-blog.com)

Thanks for sharing your experience in such transparency and detail Seb. I look forward to your new posts everyday. I am planning on doing something similar in size of scope when I have enough time to devote to it.

From Marney Perna (on Anthill Online)

OK I am now very interested. I agree with the many comments re not having sufficient skills as this is my one single most frustrating aspect of creating online products. Look forward to reading the next 6 days…. fabulous and thanks.

From Marfi (on ReadWriteWeb)

Wow! Great story, I wish him good look, he is an excellent example to everyone complaining and hating when the real obstacle is… [their] own lack of motivation

From TheNextCorner on Twitter

TheNextCorner @sebeckmas You already inspired me…!

Where to from now?

Whilst the AutoCarLog service is fully operational, the business is still in its infancy.  Like a new born, AutoCarLog requires a lot of attention to grow and I have no intention of sitting still.

In the next few weeks I have meetings with the Managing Directors of several Tax Accountancy franchises and will be selling ‘prepaid’ accounts to National stationary suppliers, hopefully to sit on the shelf next to the humble $4.69 paperback.

AutoCarLog will be expanding internationally and I have already been approached by people wanting to discuss licensing the service overseas.

Finally, I have a large PR campaign planned to promote the service in the offline space.

So I hope I’ve been able to inspire you to give your great idea a chance.  This series of blogs has shown it is possible to limit your exposure to just seven days and $500, which is a challenge you could easily set for yourself.  If you do, please let me know – I would love to hear about it!

Now time to get some sleep.

A very tired Sebastien Eckersley-Maslin at the end of the Challenge

With a dramatic click of the 'upload' button, AutoCarLog was launched

Today marks one week after I set myself the challenge of building a business in seven days and for less than $500.

After just one week, my business AutoCarLog is now online and fully operational, check it out!

A full detailed review of the launch will be posted soon, delayed to match Anthill Australia publishing schedule (noting that tomorrow is Australia Day here too).  I can tell you that within 15 minutes AutoCarLog had it’s first customer!

• Day one – The Idea
• Day two – The Setup
• Day three – The Pitch
• Day four – The Function
• Day five – The Graphics
• Day six – The Gateway

Nothing is particularly hard if you divide it into small jobs – Henry Ford

AutoCarLog was quickly becoming a fully functional system to manage vehicle mileage logbooks online.  Now I needed a way for my customers to update their log books using SMS, and for this I needed an SMS gateway.

What is a gateway?

An SMS gateway is a device that takes an SMS from the mobile phone network and spits it onto the internet.  There are surprisingly quite a few ways to go about this.

The easiest way is to use a third party.  As previously posted, in Australia there are quite a few companies who will manage the gateway and send every incoming SMS to your website.  They can also arrange to charge the mobile user a premium for this privilege (called ‘Premium SMS’).   These services have a start up fee and an ongoing monthly cost, which can be as high as $1,300 a month for six digit premium ‘19’ numbers.

With my $500 budget, I couldn’t afford to go down this path for AutoCarLog so I had to come up with another cheaper way.

Using a mobile phone

The core of a SMS gateway is the GPRS module, an electronic chip which can talk to the mobile phone network and receive SMS of an attached SIM card.  You can find these electronic chips everywhere – as every mobile phone has one.

So I can receive SMS on my mobile, but how do I get them to my website?  Well firstly, you need to get the SMS into a computer.

Most modern mobiles come with the ability to connect to a computer via a serial or USB cable.  The mobile phone companies will then install a suite of software on your computer to manage contacts, read SMS, and update SIM card details.  But what you might not know is, it is possible to send “AT commands” to a connected mobile by hand using a terminal program and read the messages directly.

The good news is, this idea isn’t new.  Here’s a tutorial from the Nokia website on how to connect to their mobiles, and here’s a great source about connecting mobiles to computers.  Look up the AT command set for your mobile, as although it’s common, it isn’t a standard.

The flow of information from the customer to the server

The program

Sending and receiving messages from your computer by hand is all very well and good, but you’re not going to want to sit at your computer 24/7 and type AT commands every time you check for a new message, no, you need a program to do this for you.

Any programming language that supports serial ports will do.  DotNET, C, C++, Python and the old Visual Basic can all do it.

Your program shouldn’t be too complicated, in essence it will:

• Check for new messages using AT+CMGL (Nokia),
• Read a message in using AT+CMGR (Nokia),
• Parse the message to retrieve the phone number, date and content, then
• Upload it to a website using a HTTP request.

Fortunately there are heaps of examples online of how to go about this, including completed code. You should be able to get something together with basic programming skills, and it wouldn’t cost much to outsource the development of one.

The hardware way

I didn’t use a mobile for AutoCarLog for two reasons.  Firstly, believe it or not, I didn’t have a spare one lying around!  And secondly, (and the real reason) I can never resist the opportunity to get out the soldering iron and build a piece of hardware.  So instead of using a mobile, I built a stand-alone gateway using an AVR microprocessor and the GPRS module found in SAGEM phones.  You can see my prototype below.

AutoCarLog SMS gateway prototype

My device was built by combining modules from previous projects. I already had a system to send twitter updates from a pot plant (“water me please!”) so posting information to AutoCarLog – which is just a different website – should be easy.

Well it wasn’t.  The twitter example used a class built by someone else to talk to the twitter servers and I had to create one from scratch for AutoCarLog.  Also, I spent several hours wondering why my device crashed after 20 seconds, only to find it was because I missed a memory mapping tilde (~) in one line of code.

Eventually I got it to work.  I had a system to check for new messages using the AT commands in one hour, a system to parse the message for phone and content in six hours and a system to send these messages to the AutoCarLog website in seven hours (which would have been four if it wasn’t for the tilde).  I added a small LCD screen to tell me what it was doing and spent another six or so hours refining and testing the modem.  My girlfriend K didn’t see me that day.

Uploading to the server

Uploading to a server is as simple as opening a website.  Open up google in your browser and do a search for something.  See how the URL has now changed from www.google.com to something much bigger? You have just sent google some data (your search in this case) which is contained in the URL.  This is called a GET request.

For AutoCarLog the process would as such:

• Modem receives SMS message (custom hardware, but could be a standard mobile phone)
• Modem sends message to a computer (AVR MicroProcessor for me, but this could be a desktop PC)
• Function resolves phone number and message content
• Function sends an HTTP request to a website with number and message encoded in URL
• Website receives URL, decodes the message and sends to database

For example if 0413227564 sends the message “20321 b work planning meeting” to AutoCarLog, the modem would request the following URL:

/upload/?number=0413227564&message=20321%20b%20work%20planning%20meeting containing the two GET value pairs:
number = 0413227564 and
message = 20321 b work planning meeting (%20 is URL speak for a space).

My page on the server would then decode the message text and in the above example, it would add a business trip with the odometer reading 20,321 and comment “work planning meeting” to the logbook which the number 0413227564 was registered with.

Limitations

It is worth noting that these methods will only work if you’re receiving SMS at a slower rate than you can read them, which is about one per second.  However, if your business is receiving 60 messages a minute you could probably afford to go through a third party or invest in a commercial modem bank!

Final word

Building a modem sounds like a huge task, but when you break it down into smaller steps it becomes manageable.  An SMS gateway is just a modem talking to a computer.  A modem can be a simple mobile phone with a USB cable, or it can be a complex embedded hardware device.  The trick is to tackle the smaller pieces first.

Home made SMS gateway sitting on my book shelf at home.

• Day one – The Idea
• Day two – The Setup
• Day three – The Pitch
• Day four – The Function
• Day five – The Graphics

Day five, (well technically day six as it’s now 12.49am as I sit to write this blog!)

The logo

The first thing I set about designing was the logo as the logo is my brand.  It is the key element which ties together my website, business cards, letter heads, flyers, posters, brochures, stickers, car wraps, building facades…. so it is important to spend a bit of time to get something I like.  (Changing the branding after being in operation is painful and confuses the hell out of everyone.)

Looking at my mood board you can see where the inspiration for the logo came from, a combination of the odometer photo, the Touareg Speedo and the DayOneFitness logo.

The fifth generation of the AutoCarLog logo

The website

When designing your website you need to plan ahead.  How many pages? What is the aim or call to action for each page? What keywords are you going to target for SEO?  Will I use Adobe Flash or HTML?

Each page should stick to a single topic (e.g. benefits) and have at most seven points.  Any more and your reader will be confused or get bored and miss your many key points.

For AutoCarLog I wrote down every thing I wanted to have online using sticky notes.  I then grouped the sticky notes on to individual paper pages in similar topics.  After a bit of shuffling, I had my website:

• Homepage. The landing page will briefly state the key benefits and drive the user to the sign up page 1st and the remainder of the website 2nd.  The landing page content has to be interesting to capture your audience’s attention, testimonials are very good at doing this.
• Tour.  I wanted people to see the how easy AutoCarLog was to use.  Eventually this would be a video demo.
• Benefits.  A summary emphasising the key benefits of AutoCarLog.
• Signup.  A page to enter in account and payment information and sign up.
• Privacy.  It is important to clearly articulate your privacy policy if you’re collecting personal information including what you intend to do with it.
• Login. A simple page for returning customers to login to the admin section.
• FAQ. A long list of questions likely to be asked by potential customers, with the answer always being a positive phrase.

Additionally, on every page would be a large call to action button in a prominent location to encourage people to sign up.

Why now

I would just like to quickly explain why I’m doing the graphic design now, after the functional code.  Yesterday I explained that I aim to build sites which are easy to use and have a great user experience.

I build a fully functional website in the default HTML style and then when it is complete, I design the graphics to fit the functions and enhance the experience.  The other way around means your functions are restricted by the design, and doing it concurrently leads to many design iterations.

The end result is my sites are easy to use and look great too.

Design ‘rules’

Artists will argue that there are no rules in art.  I don’t believe this is true for functional website graphic design (as art is that which only exists for itself, is it not?).  Fortunately for me, rules do exist which can be applied and followed.

The key rules I considered when designing the AutoCarLog website were:

• The Flow. On the public pages this is a measure of how the user is driven around the site to the call to action.  By contrast on the customer back-end pages, it is a measure of how easy the user can drive the site.
• Fonts. Pick one or two at most.  If you’re going with two, then have one for headings and the other for content and be consistent.  I used Geneva for the body and Georgia for the navigation and headings.
• Colours. The colour is very important, I generally get my colours from a photo I like (using the eye drop tool on the Mini in my mood board).  Pick one or two colours with matching contrasts and an additional colour to use for your call to action.

• Layout and whitespace.  Use whitespace well, I quite like the look of open sites but make sure your hard hitting content is above the fold and can be seen by most browsers.  Remember most users don’t like scrolling.

Call to action

The call to action is the most important part of your page.  It is what you use to achieve the aim of your page in the precious few seconds someone is looking at it.  If your aim is for people to contact you for example, then your call to action could simply be large prominent text “call us now for a free quote on 1800xxxxxxx”.  The aim for AutoCarLog was to get people to sign up on the spot, a button to the signup page.

The colour of your call to action is very important as it subconsciously influences how people feel.  The colour needs to stand out against the background and should provoke an emotion, excitement and desire.  Orange, yellow and red are good colours to use.  I used red as it fit in well with the other colours and was based on the Spreadly layout which I liked.  See how it stands out and almost demands you click on it!

AutoCarLog call to action button

Templates

Before I got too carried away, I jumped on google and looked up ‘beautiful website templates’ in case someone had already created something similar to what I had in mind which I could purchase, use, or adapt.

Using templates are a great way to save a heap of time and if you use a free one, please ensure you credit the original designer!

For AutoCarLog there was nothing which I could use easily, so a fresh site had to be created.

The mock up

The first step to create the site was on paper.  In my notebook I made a very rough sketch of how I wanted the site laid out, using inspiration from my mood board.  Then I designed it in photoshop.  (There are many great and easy to follow tutorials explaining how to do this.)  Photoshop (or similar graphics package) is used as it allows you to quickly design the site with a holistic view.

After designing the look in photoshop, I extracted elements which couldn’t be done easily in HTML and saved these as images, then using guides I worked out where my page divisions (div) would be and coded a ‘blank template’.  I abstracted the top (navigation) and bottom (footer) into separate files and voila! AutoCarLog had a face!

AutoCarLog has a face!

Outsourcing website

Just like outsourcing development it is quite easy to outsource the graphic design.  See yesterday’s post about how to go about outsourcing the design.  Also, if you’re a handy developer who is just stuck for a bit of inspiration, consider getting a photoshop website designed, which you can cut up and turn in to code yourself.

Well it’s now 2.41am and I’d better get some rest before creating the SMS gateway!

• Day one – The Idea
• Day two – The Setup
• Day three – The Pitch
• Day four – The Function

Even without knowledge of how a website works, you can manage the construction of one.  All you need is three key elements: your mood board, an understanding of the data and a user experience map.

Scope

The first thing was to understand the scope or context of the problem.  Using crude stick figures in my notebook I scribbled down the two ways users would interact with AutoCarLog and made note of the blocks which were outside of my control.

User > website > database

User > mobile > SMS gateway > database (subject for Saturday’s Blog)

In order to maximize my efficiency I needed to understand the relationships between each block.  These are; the user interaction with the website (user experience) and the website interaction with the database (database design).

If you’re going to hire someone to do your development, a good understanding of these two elements will mean you can save time (and money) by giving them a head start.

Keep it simple

Before getting stuck into some design, I took a step back and asked if I could build the service I wanted using (and perhaps modifying) an off-the-shelf-product.  For AutoCarLog? No. There were too many unique features which would mean that it would take longer to modify an existing system then it would to develop a new one, however that’s not always the case.

Take a look at your context diagram.  If you only have three elements “user > website > database”, then chances are you can use an off-the-shelf product.

There are many template-able free products you could use such as wordpress and Drupal.  Have a look at these, and the available plugins before you dive into building your own, no point in re-inventing the wheel!

User Experience

Drawing my user experience

The user experience is something I value highly in every product and service I produce.  One of my products must be intuitive and simple yet offer all the information my customers could ever want.  A good user experience reaches deeper than pretty graphics (which I haven’t even considered yet) and should be done first.

I used a mind map, with the user in the center and explored all the ways I wanted them to interact with the website.  They needed to signup a new account, add new trips to their logbook, manage their account and view and download their completed logbook at the end of the year.

From the mind map, I wrote down a series of network diagrams explaining how a user would fullfill a particular function.  All up I had about 30 of these, here’s the one to add a new mobile number to their account:

Login > click on senders > enter new mobile number > click save

Data in data out

Almost every dynamic website you come across has a database behind the scenes. Twitter, Facebook, eBay, my blog, even Anthill Online has a database driving the content and Google takes it to the extreme with over 1 million database servers.

If you’re going to use someone else to do your code, having a basic understanding will help you discuss the development with them, particularly if you use an online development source.

At the bottom is a photo of the page I used to design the AutoCarLog database.  The blue shows the initial concept, the red is the actual database I created.  From that initial concept I was able to explore the relationships between the data and ultimately build the database.

Language

I’m using Apache, PHP, MySQL for AutoCarLog as I’ve used it extensively in the past and it’s free.  I can also use my existing code modules used in other websites for the common functions (like logging in).

This triad of open source offerings work very well together and are well supported by a massive community with plenty of examples of code online.  There is very little you can’t do with them.

Choosing this combination is also slightly cheaper than some alternatives as the hosting cost won’t be as high (it uses Linux) and there are more developers available – you don’t need to understand the language to take advantage of that fact!

Outsource

AutoCarLog mood board

I now had everything I needed to build my site myself or to get someone to do it for me.  For AutoCarLog I had the skills to do the development myself, but that’s not always the case.  So where to find that someone?

If you have the budget, consider a local business.  There are a lot of them around, so it pays to do your research to get a good quote.  Give them your three key elements so they can give you an accurate estimate.

A more hands on (and cheaper) approach is to use an online marketplace such as guru or elance.  Here you publish your requirements and budget and others bid for your work.  Be aware, you need to rule your new employee with an iron fist to ensure you get want you want on time, but in most cases it’s cheap and it’s quick.

Lastly, the thing you need to remember when outsourcing a site is that there are two components to the it: the functional guts and the pretty graphics.  (This blog so far has only considered the functional guts.) So if you’re hiring an individual consider their specialty may be biased towards the functional guts (developer) or the pretty graphics (graphic designer).  Make sure you enquire as to their skills in both areas.

Give it a go

Web development isn’t as daunting as you might think.  There are heaps of tutorials and examples to follow and if you’re a naturally structured person who likes writing things down in lists, then you should be able to pick up code.

I taught myself everything I know about web development by following online guides and just giving things ago.  After just two months I had a basic contact management tool (which I still use).  After just two years I had built an enterprise application driving a business with 30 staff that I sold to my largest competitor, as it was superior to what they had been using for years!

But for me, I’ll do the pretty tomorrow, now on to the functional guts!

Designing the database

• Day one – The Idea
• Day two – The Setup
• Day three – The Pitch

Never miss an opportunity. A blackboard in the men’s urinal at the local.

The Strategy

I had limited time and very limited funds, so how was I going to market AutoCarLog?  This question provoked quite a bit of discussion on the forums I was posting in.  The answer?  I couldn’t.  With my limited resources and time spent in development I had to get other people to do it for me, and importantly, I had to get them to do it for free.

The Pitch

“If you can’t explain it simply, you don’t understand it well enough.” – Albert Einstein

I couldn’t do anything without the ‘Pitch’.  The pitch is a one liner that wins hands down in the ring at Pitch Club.  It’s how you sell your product or service to someone in an elevator when you only have 20 seconds of their uncomfortable attention.  A pitch must address the value proposition by answering “what is it and what’s in it for me?”

AutoCarLog is the easiest way to maximise your taxable claim for car expenses, by logging every trip by mobile phone and automatically calculating your claimable amount.

Not perfect, but I could deliver it with passion and conviction and it explained what the service was and how they would benefit.

Bulk buying power

A great lesson learned is that it is far easier to sell to one buying many than many buying one.  “But AutoCarLog is for the individual” I hear you say, “who would want to buy it in bulk?” How about Tax accountants?  Could I convince them to refer my service for me?

Armed with my awesome pitch I called the major ones up.  After just six phone calls I managed to get a meeting with the Managing Director of one, and an invitation to email a proposal to another.  I made sure I spoke with the decision makers and they were all supportive of the idea.

The deal was they would refer their clients struggling with their car expenses to use my product and in return I would sign over half the signup fee.  I had in effect 1,000 voices around Australia promoting AutoCarLog. Brilliant.

All it took was a couple of phone calls so don’t be afraid to give it a shot!

Word of mouth

Word of mouth advertising is one of the strongest forms of marketing and it’s free.  But it is also the hardest to drive, the best you can do is influence it.

The best way I can think of to influence word of mouth is to have an awesome product and target the bottom line.  I decided to offer AutoCarLog completely free if people referred the service to their friends.  Sign three people up and pay nothing.  No catches, no gimmicks, just a simple referral system.  It worked for Amway.

Price point

Sell for too much and few will buy, sell for too little and it will take a lot of sales to recoup costs.  Setting the right price was my next challenge.  I conducted a snappy market research by asking friends, accountants, my bank manager and everyone I spoke to on the phone asking, them two questions “how much and how often”?

I settled on $39 for six months, $69 for one year ($5.75 a month), which in the words of my web savvy accountant was “chicken s*** cheap”.

Value adds?

Finally, I had (what I thought) was a great idea to send all AutoCarLog’s customers a small business card size sticker with a quick ‘how to guide’ to stick in their car somewhere – a reminder to update their log and explain how. To save costs I could print the first few out myself for less than $10.  When I was pitching this idea to K she drew me back to an old episode of the ‘The Apprentice USA’ (Season 3, Episode 15) where contestant Tana spent hours putting Dazzlers (little sparkly jewels) on her limited edition t-shirts.  She believed they would bring in a higher price.  The truth is she would have been better off spending that time better marketing her product (without the dazzlers) and might not have lost that round as a result.

So whilst I love the idea of a sticker, it’s not going to bring in more customers.  I need to focus on the core service and spend the time marketing that.  (For now at least, I can always print them later!)

Right, now on to some nitty gritty development.

Designing the user experience

A good user experience is very important to me in sites and products I design. However a good user experience doesn’t come by itself.

For every new project I start I walk through a user experience design process where I use mind maps to flesh out how the user will interact with the site.  Then I let people use the site and product (as a ‘beta’ release) and provide feedback and use their feedback to refine the design.  Ultimately (after several iterations) I end up with a product that is intuitive and easy to use.

With AutoCarLog I didn’t have the luxury of going through a heap of user testing, but I still felt that spending time designing it was worth while.

As I was doing the development, a model will increase my development efficiency.  If you’re using a third party, then a user model will help them understand what you want.  In both cases decreasing the time taken and reducing the cost of development.

• Day one – The Idea
• Day two – The Setup

I had six days left, four of which were business days – so it was critical that I set everything up today to allow me to focus on development and marketing.

Structure

The first thing to look at was what business structure I wanted to operate under.  The best way to do it is unique to your situation, so it’s best to ask your accountant and solicitor.  At a minimum you should be looking at registering a business name and ABN, for more complicated structures involving companies and trusts you should seek advice – however if you know what you’re doing, Cleardocs can get you up and running in under an hour.  Other things to consider are trademarks, if you need to lease a post box or virtual office space, if you want a 1300 phone number and if there are any licenses you need to apply for. For this project, I decided to operate under my existing software company, thus saving the cost of registering a business name (about $110).

Billing

Although AutoCarLog will make extensive use of SMS, I didn’t want it to be a premium service (one which the customer pays a ‘premium’ to send or receive SMS).  There had to be in incentive for my customers to send in lots of trips – and charging for each of them individually was one way to ensure no-one would use the service.  Besides, a business owner with 100 trips a week is hardly going to want to spend $100 keeping their log book up-to-date, remember, I’m up against a $4.69 paperback here!

What about by credit card? Customers could sign-up for a period of time, have unlimited access and post unlimited trips during this time.  I could also integrate the credit card gateway into the site and automate the whole process.

There are quite a few third party services which offer just a product, (eWay, PayPal and PayWay are just a few) and there are basically two ways of accepting payments.

• After clicking ‘buy’ my customer would be redirected to the credit card payment site and then redirected back to AutoCarLog,
• Customers enter in their credit card details on a secure page hosted on AutoCarLog and the payment is done in the background via an API.

The bank loved my pitch

Both have their advantages and corresponding set of fees.  I decided with PayWay as it allowed me the greatest flexibility, was reasonably priced and I could understand their API. So first thing in the morning, I went to the bank to setup the AutoCarLog trading account and made the call to PayWay. Done. $220. Contracts sent via email, I’ll have them in a few minutes.

Whilst at the bank, I pitched my idea to the manager.  She loved it! She even suggested I contact the bank’s email newsletter team as they might be interested running with the story.  Just goes to show – you never know where leads might come from, so pitch your idea to everyone!

SMS Gateway

I wanted my customers to be able to update their logbooks directly from within their cars at the end of each trip, using SMS.  This meant I needed a ‘gateway’, something that allowed each SMS message to be sent to a computer, processed and added to their logbook.

The easiest way is to set up an account with a third party gateway.  But this can be expensive – upfront I’d be looking at over $1,000 for a six-digit number, between $100 and $500 for an eight-digit number and around $100 for a ’standard mobile’ number.  Add to that a charge per incoming message and add a monthly account fee.  Another option is to have your own SIM card and buy a SMS Modem, however they start at $350 and go all the way up to the $1,000’s.  Both these options were too expensive.

Alternatively, I could always build my own!  There are two ways to do this, firstly I could use an existing mobile phone with USB adapter and write some software on a host computer, or I could build a dedicated standalone modem.

The cheapest option would be to use a spare (or ‘borrowed’!) mobile handset you have lying around, but alas I didn’t and I wasn’t going to cannibalise my iPhone.  So I decided on building a dedicated modem, leveraging components I previously put together for a client to update their scrolling LED displays by SMS – thus saving a heap of time.  The microprocessor, GPRS chip and circuit boards came to $172.50 and the only ongoing fee was the SIM card subscription.  With that settled, I left the modem building alone until the end of the week.

The number

Now that I’ve decided to manage my own SMS gateway, I needed a number, and not just any number.  I needed a number that was memorable.  I went to my whiteboard (which still had the brain storming words on it) and using my phone I wrote out a list of phonewords in my notebook, here’s just a few:

• 04×2 886 227 (04x AUTOCAR)
• 04×2 886 564 (04x AUTOLOG)
• 04xx 227 564 (04xx CARLOG)
• 04×5 642 665 (04x LOGBOOK)

Registering 0413CARLOG

Armed with this list, I headed to the local telco to try and register a new SIM.  I handed the list over to the girl behind the counter asked to register one of these numbers.  She looked at me blankly then pointed to the list and said that these weren’t numbers, they were letters.  A quick demonstration of how a phoneword works later and I was informed I could have my 3rd choice: 0413CARLOG.  However – it was going to cost me $22!  I had not budgeted for this and I knew things were tight – but, the number was good… I asked if this could be waived, as I was on a challenge! Alas no, my challenge was not their problem.  I hesitated and went ahead anyway.  $22 once off for the number plus $5 a month for the SIM card.  0413CARLOG.

Hosting

If you’re going to be starting an online company you’re going to need a domain name and some hosting. I had already registered www.autocarlog.com so all I needed was somewhere to host it.  You can of course sign up for a free google pages account but you’ll never have the flexibility and control of your own hosting, and for anything that requires databases and server side scripting it’s a no brainer.  So for AutoCarLog I registered an additional domain with my existing host for a cost of $4.40 a month.

As I was using a credit card gateway, I needed to have an SSL certificate, which in turn needed a static IP address.  $3.30 a month for the static IP, $66 per year for the SSL.  Luckily, I just saved $22 from my budget which I needed for the number registration!

Costs

After coming home I tallied up my score:

• Web hosting: $4.40
• Static IP: $3.30
• Domain name: $13
• SSL certificate: $66
• SMS modem: $172.50
• Bank account: $6.50
• SIM card: $5
• Custom number: $22

All up, I’m looking at $496.80 up front with $15.90 (bank, hosting and SIM card) due in a month.  Not bad for a day’s work! Tomorrow I’ll look at marketing, but tonight I continue on with the development.

Phonewords for AutoCarLog

• Day one – The Idea

Everyone has a one good million dollar idea, but not everyone acts on them.  There is always an excuse, always a reason why not.  The most common I hear is “I don’t have time” and “I don’t have the money”.  Bollocks! I want to show that with just a little capital and a little bit of time anyone can turn an idea into a business – a start up.  This is the first blog in a series where I attempt to create a start up business in just seven days and for under $500.

Day one – The idea

The first thing is to come up with an idea.  The idea has to be simple to develop, quick to market and have wide appeal.  Due to the short time frame the idea must also be within my capability. A premium SMS service instantly comes to mind, however as these have a high initial setup cost they would blow my $500 budget, so no. But SMS might still have merit.

I recall an idea that came whilst chatting to a good friend of mine a few months ago.  Working from his car, he was repeatedly frustrated by having to write down his mileages in a logbook, a logbook which would then be misplaced, lost and chewed by the dog. Logs would be jotted down on the back of receipts, discarded wrappers or scrap paper.  Ultimately he wasn’t recording every business trip and he was therefore unable to claim the most he could come tax time.  A quick calculation on the back of a laptop revealed that missing just 1 trip in 10 would cost him $1,000 in lost tax claims! He suggested a log book could be done electronically and I thought this would be a great idea to use for this challenge.

A service to maintain a vehicle mileage logbook online and update it anywhere via SMS.

It was simple to develop, I had experience with building SMS and online business systems.  It was quick to market, the benefits could be explained quickly, promoted online and bought on the spot.  And it had wide appeal – 7 million Australians claim work related expenses each year on their tax returns.

The marketplace

Although I already had one customer, my friend who kept losing his logbooks, I needed confirmation that others would like the idea.  I jumped on facebook and twitter and asked the question.  I floated the idea to friends, family, my bank manager, my accountant – any one who would listen.  I even asked randoms waiting in queues  – though I often received a cold response to this.

From this people were very warm to the idea of managing their logbooks online and having the ability to send SMS.  Lots of people didn’t like premium SMS and a few suggested that they would like to send an email rather than an SMS to save money from their smart phones.  My accountant put me onto various ATO websites and noted that costs using the service could even be considered a tax deduction. People were generally happy to pay up to $10 a month for such a service.

The competition

Ok, time to see what else is available.  I jumped online and tried every search term combination I could think of and came up with a list of existing businesses which compete directly or indirectly with the AutoCarLog concept.

• Officeworks. At $4.69 this paperback logbook is the cheapest and most popular product I would be competing against!
• TrackInABox is a product that does automatic logbooks using GPS tracking. Pricey and targeted at truck drivers.
• AutoCentral is one of many fleet management software systems targeting large organisations.
• There are a heap of iPhone apps which do log books, but none seem to be targeted to Australian Tax savings.
• SideBuddy is proof the AutoCarLog concept is viable and was the only one I could find like this.

There were of course others, but the above list shows the variety of tools and services which currently exist to manage vehicle mileages including one very similar concept.  Although there was competition, I never see an existing service or product as a reason to back out. I see them both as a challenge to do better and reassurance that the concept is valid and the marketplace is already being educated.

The name

The name is important, it’s on your business card, it’s on your website, it’s the first impression people get with your product, it sets the scene.  Initially I started thinking of inventing a word, such as Carnoodle or Odonaba using the cheekily simple web2.0 name generator, but I wanted the name to draw business in by itself.  So I needed something which was memorable, quick and described itself.  Not easy.  I jumped on my whiteboard and wrote down words that describe the service, ‘logbook’, ‘car’, ‘mileage’, ‘automatic’, ’sms’, ‘online’, ‘tax’ and eventually settled on the combination ‘AutoCarLog’.

The name “AutoCarLog” is a combination of words that are instantly memorable and informative. The added advantages are that it is short and therefore twitter friendly (not needing to use a url shrinking service) and top level domains .com and .com.au were available.

The plan

The most important part of a business is of course the plan.  It doesn’t have to be a full blown business plan suitable for seeking funding from the most conservative of banks, but a plan will help me prioritise tasks for my six remaining days and work out where to spend my $500.

When I came up with the challenge, I figured $500 was more than enough, it turns out it was just enough! My quick budget is as follows:

• domain name registration – $13 a year
• web hosting – $5 a month
• SSL certificate – $88 a year
• credit card payment system – $220 plus $22 a month
• Bank account – $6.50 a month
• SMS modem – $140
• SIM card – $5 a month

Total upfront – $499.50 and ongoing – $66.50 a month

As there were things I could only do during business hours (visiting the bank) I split each day into two.  During the day I would have to cover registrations, marketing and promotions and during the evenings and weekends I could focus on development and design.

• Day one. Research & Planning
• Day two. Registrations
• Day three. Marketing & Development
• Day four. Development
• Day five. Graphic design & Development
• Day six. SMS gateway & Development
• Day seven. Launch!

Giving me 7 hours research, 7 registrations, 7 marketing, 42 development, 7 graphic design, 14 sms gateway and 7 for the launch – 91 total!

Each morning I will blog about the previous day. I will also cover other topics which don’t fit into the daily summaries at my blog semi-blog and you can follow my progress on twitter.