It's generally accepted that startups should start charging as soon as possible. But for SaaS web apps the question is often whether to use a Freemium or Free Trial business model.

What is 'Freemium'?

Freemium is a model often used by B2C services such as Twitter, Flickr and Evernote, where the user gets a Free version of the product by default and then is asked to upgrade when they want extra features beyond what the free plan offers. The signup rate is usually very high and the ratio of free users to paying users is usually very high (obviously).

The advantage to using Freemium is it gets a large user base relatively quickly and theoretically acts as a marketing tool because users will tell their friends. The disadvantage is that the free users can cost in support costs and bandwidth. Also there is the ethical problem that paying users are effectively paying for the free users who will probably never pay.

There are some great examples of Freemium, e.g. Skype, because you can use it to ring your friends and family totally free and wouldn't expect to pay to make a call like this on the Internet, but when it comes to having your own SkypeIn number or being able to ring mobile phones from your Skype account you would be happy to pay. Also CurdBee and Invoice Bubble have a good Freemium model that lets you switch off advertising and get a few additional features for a low monthly price. So it works well in lots of cases.

What is a 'Free Trial'?

The Free Trial is used often by SaaS web apps. They offer a trial of their product for 14 or 30 days and after the trial is over your account is suspended and you are told you need to pay to continue using the service. The idea is that in those 14 days you will get full use of the product and try it out, and in this time their job is to capture your attention, educate you on the product and entice you to upgrade.

The conversion rates tend to be lower for free trials, but the service knows that if they get a happy user from this free trial then they will get a new paying subscriber and won't have to support any free users. It tends to lead to more money in the bank at an earlier stage, but there is a marketing disadvantage and the growth will be slower.

Personally I don't really like having a timer telling me how long I've got until I decide to buy a product, and I've often found with my Startups that people will sign up to a service, then not come back and try it until a couple of months later. Then they have to delete their account and sign up again using a different email (if they are keen enough to try it out again).

The Debate

I'm not saying one is better than the other, and in my experience with this I am still very much learning, and I'll detail my experience later. I think choosing the model largely depends on the type of service being offered.

For example, a company who use a Free Trial is Wistia who I rate and recommend. I knew about them because I actually featured them in my web apps blog a while ago. So when I actually needed video hosting I went to their site knowing fully well that I would be signing up to a free trial and I had 14 days to see if the product was right for me. At this time I'm still on their trial but pretty sure I will upgrade at the end of it.

So it works well for them because I made a decision in those 14 days and they will get my business within 2 weeks of me signing up, as opposed to getting my business after giving me months of free video hosting (if they were using Freemium). They may not have as many users as if they used Freemium, but they know who their customers are and they can put full resources in to supporting them, plus they known that all their server costs are covered because everyone who uses the service pays.

However, in other cases a free trial can be a turn off.

If a Twitter application like TweetDeck used a 14 day trial I wouldn't sign up for it and I'm sure many thousands also wouldn't. However they are in a market where people expect to download something like this for free and not have to pay. I actually don't know what TweetDeck's business model is but I'm sure it's something to do with marketing their other products - perhaps someone could inform us on this?

My experience with Freemium and Free Trials

In the case of Project Bubble I was advised to start charging as soon as the service was out of Beta, so I did so pretty early on once we had traction and a fairly good product. My first business model was Freemium because I wanted to see the same growth that I had experienced in the three months prior when it was free (we had 5000 users in 3 months!). I also wanted to honour the existing members and make sure I didn't upset anyone, so basically we added more features and encouraged our power users to upgrade to get these features. New users were coming in and on average taking a few months to upgrade (that was when they hit the limitations of the free account). I was happy with this in the early stages because I was working on another startup and had money in the bank to pay the bills.

However I soon realized that the business model was not returning enough revenue to pay for me to work full time on it, let alone fund marketing, iPhone apps or employ support staff. So I decided to bring in a free trial because I thought that by doing this all of a sudden all our free users would suddenly just see the light and upgrade, and also new users would just think it was the norm and upgrade after 14 days of using such an amazing product without a second thought.

I was wrong. What actually happened is I made quite a lot of users fairly annoyed (even though I tried to 'grandfather' them as best as I could), experienced a massive conversion rate decline for new visitors, and had a lot of work to do changing the language of the site away from 'Free'. I also wasn't capturing the users who were signing up. I should have been emailing them, asking them what they thought, showing tutorial videos to them and educating them.

What I found was people weren't really upgrading as I expected and the product was still geared to the market who had previously found the service absolutely wonderful - freelancers and those used to getting stuff for free (such as those in the open source community etc). So I had to make a decision to either change back to Freemium or proceed and make the service more suitable to the free trial model and do better at capturing leads.

I couldn't afford to continue like this. At the time I was also trying Google Adwords out and paying a lot of money to get users to the site. Without capital I knew I had to go back to Freemium and use that as the primary marketing tool once again, hopefully experiencing the growth I used to have in the early days.

Note: once you switch out of Beta (totally free) and start charging you will never get the same signup rate, so don't expect to go back and get those results like I did because you won't.

A Hybrid

Now Project Bubble has a hybrid of the two. The tone of the site explains that Project Bubble can be 'tried out' for free (without mentioning the amount of days) and that there is a free account if users just want something basic. However the tone clearly states that Project Bubble is a professional service catered to a small business market (design agencies etc) and it costs money! Once inside users get a free trial of everything for 14 days, but after that they go on to a free account with limitations and advertising, plus a notice to say that the trial has ended and they should probably upgrade (even though they can continue to use the service if they want).

I'm not saying we've got it 100% right, but I think we have a good balance now and we're capturing the right people who are serious about finding a great project management tool for their small business. What's more important is capturing leads and actually building a relationship with the users who signup through email, blog or UserVoice. I'm hoping we'll see some good results now we've finally decided on our model.

In the process of working all this out (14 months) I've learn a lot of valuable lessons, which I'll share.

The Lessons I Learnt

So as I said above, it's important to know your market first before deciding which model to use (e.g. B2B or B2C) and knowing what your users would want. Also another important principle as mentioned above is charge as early as you can so you're making it clear from the early stages what you are about.

• Don't annoy or confuse your users
  Although I mentioned earlier I annoyed users by bringing in the Free Trial after already using Freemium, hardly any users were actually that annoyed. Most were more confused because they saw a landing page saying their trial had expired and they thought it was free. So I quickly grandfathered them, giving lots of free accounts out to Beta testers and offering a reduced subscription to others. With hindsight I would probably have made things a lot clearer earlier on, but sometimes with Lean Startups you don't always know exactly what your next step is.
• Communicate and inform changes to your users
  So linked to the last lesson, make sure you inform your users each step of the way. Tell them why you're making the changes and how it affects them. Be friendly, but also firm - remember you are the CEO (probably) and you have to make decisions. If you are just starting out then make a roadmap and plan out how long your beta program will be, how long you are going to honour and 'grandfather' your beta testers for, and when you will release your chosen pricing model.
• Don't change things too often
  I'm a good one to talk, because my startups are often ultra lean! But I always made sure I gave things a good go and tried them out before changing things too quickly. We had a beta program for 3 months, then we tried the Freemium model out for about 3 months before trying the Free Trial system. After 2 months we switched to the Freemium Hybrid model which we have today. The changes were also not too drastic and I hope I explained things on the blog clearly, if not with personal contact with our users through email.
• Get feedback
  I tried to stay in contact with our users by using PollDaddy to get their thoughts and opinions. I would ask them things like 'how much they would be happy paying' (don't ask that question because the responses you won't like!), and other questions like 'are you happy with the features currently or do you want more'? Stay in contact with your users and make them feel like they are part of the family. I think a lot of Project Bubble's success is down to the wonderful community of users we have who I try to stay in contact with as much as possible.
• Don't undervalue your startup
  Your app is probably worth a lot more than you think! A lot of web app founders suffer from undervaluing their startup. I certainly did when I started out. If it wasn't for a good friend telling me "Your app is as good as Basecamp - start charging!!" (still don't really believe it by the way), then the app would probably still be free. Look at your competitors apps and pricing, get feedback from your users and if they are generally happy and willing to pay then charge the going rate.
• Understand your market and where your startup fits in
  Does your service help with productivity or invoicing? Then you probably shouldn't be giving away too much for free because your users will be making a lot of money off you for nothing. Is your service in the Social sphere, e.g. a Twitter app? If so then your users will likely not sign up if you mention a 'free trial', they'll expect it for free but you can charge them for add-ons which they'd likely be glad to pay for. Know your market and know that your chosen model will attract different types of users with different expectations.

Some of those tips might be quite general, but I've tried to focus them on the Freemium vs Free Trial debate. I hope they help you decide which model to use.

Let us know which one your using and what you've learnt in the process.