I spent a few hours yesterday afternoon with the 2012 class of StartMate companies in our Sydney incubator. As someone who’s spent their career working in business to business technology, I was excited to see 3 of the 8 companies focusing very strongly on solving real pain points for businesses. Smart teams, real problems, solid technology – full of win.
While their target markets – property, film and IT management – were all very different, the common challenge faced by all enterprised focused startups who want to get big was the same - distribution. I know this only too well, as this is the thing we spend more time thinking about withAffinityLive right now than anything else.
One of the tempting ways to try and ramp up distribution is freemium. The theory goes, if I make my product free, I’ll remove a big barrier to adoption, word of mouth will have a stronger effect in driving user volume, and we’ll be able to show VCs a nice hockey stick graph that goes up and to the right.
My concern is that for almost all of the SaaS companies that target businesses, those who try freemium realize it was a mistake, hopefully before they kill their business. Freemium is almost always a very very bad strategy for selling to businesses.
In planning our AffinityLive sales and marketing model, I spent a lot of time talking to people who’ve been there and done that. Most of the conversations were very off the record, but one source I always point entrepreneurs to is the experience of the guys from Chargify. They originally had a freemium model, and the reasons they abandoned it – and the backlash they sustained in the process – were blogged about extensively and very honestly by founder David Hauser. It should be compulsory reading for any entrepreneur targeting the business market.
There’s a lot that has been written by people with a lot more experience in the matter, but just telling you not to do something as trendy as freemium isn’t going to cut it. We all want to see our products used, and most engineers start solving a problem for ulturistic reasons, which explains why freemium has so much appeal.
So, rather than just saying don’t do it, I’ve listed the four attributes that can make freemium a successful strategy for your business focused online service. You don’t need all 4, but you’ll want to have most of them baked into your product and business model to have any hope of making freemium work.
As I see it, here are the criteria to required to make freemium work for business (ie, non-consumer) facing product:
- You need to have a land and expand distribution model. This means you use freemium as the thin end of the wedge and use the early adopter as your sales team, evangelising it to colleagues to eventually get the business to buy in financially for features like security, administrative control. Box does this very well, as does Yammer. Note though that these two business put a lot of money into their enterprise sales teams doing traditional enterprise sales things – small and medium guys are crumbs comparatively from what I understand.
- You need the ability to deliver benefit independently of core business processes. If you can solve a problem that doesn’t require any connection to the rest of the business – ie, no need to access any customer, accounting, HR, production or other data – then this can work a treat. This is why tools like Balsamiq or Prezzi for individuals that are used primarily in a workplace context work as a freemium product if you have the other attributes. However, being an occassionally used tool that isn’t tied into the core business processes comes with its own risks around poor revenue and high churn. Higher revenues generally require a stronger connection in the core business processes (otherwise you risk becoming a feature and see high churn), which is difficult to get if you’re delivering benefits to the individual without touching the business.
- Natural increased use model that pushes users inextricably to paid. Freemium is a marketing strategy, so you want to make sure whatever your product does, you have a conversion lever that “just happens” as users use the product normally. Dropbox have this nailed – you use it, you keep using it, and one day they say “you need to pay now” – it is a cumulative use lever. Remember the Milk have a different lever focused around urgency, which itself is tied to the utility you get from the product – their mobile apps only sync once a day unless you pay, at which point it will sync in real time – which means when you care about the app, you’ll pay for it. Find your driver, but make sure that users who actually use the app will be compelled to pay.
- Volume is more important than revenue. This only works for some business focused companies (see Wave Accounting and Spiceworksfor two examples of firms betting everyting on this model), and means you’re making money in an indirect way, either through advertising or transactional revenue. Assistly implemented it too, but only to drive their acquisition price to SalesForce higher when they already knew they would either get acquired or raise a big round (their decision wasn’t based on revenue). While venture backed companies have the luxury of not needing revenue to survive today, they need to be able to show how they’re going to see big revenue later to justify the investment risk, and pursuing this strategy to make yourself appealing to investors is pretty damn risky. The key is to make sure you’re sureof your revenue model; the number of ideas I’ve seen where folks say “and we’ll have all this data and that has to be valuable to someone” is about as much of a strategy as the Underpants Gnomes.