Get your project off to the best start

Launching in small steps was how I started a project in August 2012. I talked about Amino Kapital and the difference between inviting a few users in and starting every marketing channel at the same time.

Launching in small steps began with a working version

I had identified a need through enquiries about capital. We then made a first version people could use, even though not every feature had been built yet.

I tested it myself and asked others to try it. Along the way, we found gaps in things such as copy and emails, and received questions from users that helped us develop the service.

I invited a smaller group first

Amino Plus members got early access on special terms. I explained there might be bugs and tried to give them something in return for taking part in the first phase.

Gradually, we opened it to more people. I described how I would use Facebook, Twitter and newsletters in stages and watch whether the service still worked.

A campaign taught me something about load

I also described a historical advertising offer I had shared in many places at once. I found the landing page became very slow and visitors weren’t getting through as I had hoped.

That was my experience of what happened; my guesses about the server and the number of lost sign-ups weren’t a technical investigation.

So I wanted to test, fix and expand gradually. The services, discounts, participant numbers and advertising offers in the video and original text are information from 2012.

Launching in small steps rather than with one big bang

Launching in small steps was the approach I had used for Amino Kapital. A conversation with Søren about his monitoring project Sitebeak prompted me in August 2012 to explain why I didn’t just launch press coverage, ads and newsletters all at once.

We had a small shared office with various consultants and start-ups. They helped with search engines, among other things, and we exchanged ideas about business development. There was also someone to have lunch with. Søren said he had got something out of our conversation about launching gradually.

Many people keep building their project until they think it’s finished. Then they go all out with press releases, Google AdWords and every mailing list they have. I preferred to learn in several stages before turning everything up.

Start with a need and a small solution

With Amino Kapital, I had identified a need because I was getting four, five or six enquiries a week from people who needed money. I corrected myself in the video: it was per week, not per day. Some, for example, had a signed order but needed DKK 100,000 to complete it and couldn’t get help from the bank.

We then made the first version. Søren had done something similar with Sitebeak. He wanted to develop a larger monitoring product but started with one thing: if your website went down, you could get an email.

You entered your email address and website address, and the site was monitored. I said the idea wasn’t unique and others offered that sort of thing too. The first version was free. Søren had plans for more, including alerts if a website was hacked and its content replaced. It didn’t all need to be ready before the first version went out.

The time I went too hard myself

I had a recent example of what could happen when I didn’t follow my own method. Amino had become a Google partner, and we had an agreement at the time for a voucher with DKK 750 of advertising for people who hadn’t previously advertised with Google AdWords.

I shared the offer on my Facebook account with nearly 11,000 people, on Twitter with around 10,000, and in Amino’s forum, where I mentioned around 25,000 visitors a day. All at the same time.

Then, in my experience, the page people were being sent to went down or became extremely slow. I provocatively said Google had gone down. I was talking about the Danish campaign page, not the collapse of Google’s entire search engine.

I guessed I had sent a thousand or two thousand people there and experienced response times of up to seven minutes. I joked about whether the page was running on a little server next door or on a calculator. That was my frustration and my guesses at the explanation, not a technical investigation of Google’s operations.

You don’t simply get that first attention back

People are busy. If I ask them to visit a page and get something, and the page hangs, they may not think the fault lies with the destination. They might just as easily think I’m an amateur who hasn’t checked what I’m recommending. I exaggerated it with a whole string of insults.

We got around eighty sign-ups. My guess was that we could have got two or three hundred if it had worked. If I then asked the same people to try again, some probably would, but far from all. I imagined perhaps twenty or thirty more. Those were my assessments of a missed opportunity.

I should have shared the offer in one place first and checked whether people were happy and whether it worked. Wait half an hour, an hour or until the next day, and then use Facebook. If it’s slow, find out why and get it fixed before sending even more people in from Amino. That way, the load and the problems could have been spotted along the way.

My first version had to work without having everything

I called the first version of Amino Kapital version 0.9. In my mind, that meant it was more than a rough prototype: it could actually be used, but there were still spelling mistakes and missing features.

For example, users needed an email when their subscription expired after sixty days, with the option to continue. That hadn’t been built yet, but of course it needed to be in place before the first two months were up.

There wasn’t a welcome email at first either. I discovered that by signing up myself: why hadn’t an email arrived? Because we hadn’t made it yet. I compared the missing details to chrome, trim, a radio and air conditioning. The basic product had to run, and then we built on it.

At that point, I had sent around thirty or forty corrections to our team in Manila, including spelling mistakes and wording. Many came from users asking why the site couldn’t do something or pointing out an error. That created a development phase where users helped shape the product.

The most loyal got something in return for being first

First, I asked people in the office to look at it and test it. We made a few more fixes. Then I invited Amino Plus members, our paying members, in. Instead of the price of DKK 499 at the time, they could join for DKK 10.

I mentioned around 1,550 Plus members. If two or three hundred found the service interesting, that could give us a manageable group to learn from. The huge discount also made it reasonable to ask them to put up with a few bugs.

In those first days, I also offered extra time to early customers. I said two extra months in the video. I knew problems could arise and that there weren’t enough investors at the beginning yet. I didn’t think it was right simply to charge full price for an empty site when people came to find someone to talk to about capital.

It was a way of softening the start. Loyal members got a benefit and might also feel like giving us a little extra help. I joked that Plus members were little kings and I scattered roses and violets on the street before them. They should have a good time. Other users should be treated well too, but paying members got special benefits.

Open up to more people and watch what happens between each step

When I felt it was working, I opened it to other Amino users at the full price at the time. By then, according to my account, we had brought in around twenty investors, so there was actually something to show customers.

Then I used Facebook, followed by Twitter. Was it still working? Were people reporting problems? Only when I felt we had the rocket under control did I send it out in Amino’s newsletter to around 32,000 entrepreneurs. Then things could move fast.

I was also working on attracting a steady stream of visitors through advertising. But paid traffic shouldn’t start too early. When you pay for visitors, you want to see that they also do what you want on the site.

The press should find activity, not an empty database

I initially talked about press releases, then corrected myself: in this case, a journalist from Politiken had contacted me after seeing the project on Twitter. I described him as a really nice man. Coverage then appeared in several places, including TV 2’s website, Berlingske, Jyllands-Posten, Børsen and Politiken.

I didn’t hold a big press conference. Nor would I invite the press myself before things were under control. Someone might say a project was already five or six days old, but I rarely heard that argument. I thought it was much worse if a journalist arrived to find an empty database.

Get search, measurement and the next wave under control

Along the way, I worked on search engines and looked at which searches were coming from Google. Should we focus more on terms about investors or loans? What produced the best results? We also needed to measure what visitors did when we started paying for traffic.

Finally, I wanted to use my own newsletter, with around 20,000 loyal people. I imagined an approach split between investors and people seeking capital, since both groups were on the list. That would be another big wave once everything else was working.

That was the gradual approach Søren had taken inspiration from. Sitebeak could start by telling you when your website went down and add more features later. My point was to release something useful, learn, fix it and then expand. The services, discounts and figures in this story all belong to the events of 2012. Read also Iværksætter til jeg dør for my experiences of getting started and developing a business.

Read also about getting the product right before marketing it.

Here’s my way of going for it: And as promised: http://sitebeak.com
http://googledanmark.com/amino750