The cost of account sharing

Account sharing for B2B publishers – part 1

Account sharing is incredibly prevalent across the whole digital subscriptions space. Maybe you let a mate of yours use one of your Netflix family profiles. Perhaps there’s a SaaS tool you find really useful and you shared your login to it with a colleague because it was easier than getting a corporate licence. “Password123” anyone? Whatever the transgression, few of us can honestly deny we misuse our subscription accounts from time-to-time.

There are three parties in this situation. For this post I’ll refer to them as:

  • “the service” – which is the vendor offering the subscription;
  • “the subscriber” – who is the user that actually pays the subscription; and,
  • “the piggyback user” – who is using the subscriber’s account to access the service but never actually paid for the right to do so.

To begin to quantify the problem, somewhere between 30% and 40% of all subscribers to music streaming services share their login with at least one person, according to a survey of over 12,000 people carried out in 2017. That’s huge.

So, what’s the impact of account misuse? For B2Cs account sharing can be seen (if you squint) as a good thing. It’s word-of-mouth sales. The piggyback user is building the service into their habits and routines and, at some point, will probably need to actually sign up; if they don’t they weren’t getting value out of the service anyway.

That advantage doesn’t necessarily carry over to B2B, however. In general, a B2B subscription is for a substantially higher fee and the customer base is much smaller. A smaller market means that the word-of-mouth growth from piggyback users won’t have the scale to become viral (there are some obvious exceptions which are widely horizontal, like Slack). This is compounded as the cost of the piggyback user, the loss of potential revenue, is much greater.

I asked Tony Skeggs, CTIO of Private Equity International, for his view:

“[The cost of account misuse] is not an insignificant amount. Enough to justify setting up a claims team to track client behaviour and … claim back revenue.”

I’m sure that’s a position many B2B publishers share.

Now, every business is different and it’s dangerous to generalize (particularly with domain-focused B2B publishers) so there are two questions you should be asking yourself: “how big is the problem for me?” and “what can be done about it?”. The first question is the focus of this post, the second will be covered in part 2.

How big is the problem?

As with most difficult questions, this is multi-facetted. You need to know, at least:

  • How many users are regularly sharing their account?
  • What proportion of the piggyback users would otherwise convert into customers?
  • What would it have cost to identify the piggyback users as net-new leads?

The last point, in particular, draws attention to the balance here: account sharing is not bad in an absolute sense, it’s just probably responsible for an overall loss, in the case of most B2B publishers.

How many users are regularly sharing their account?

To the first question, it is tricky to accurately estimate the number of users who share accounts. However you can see tell-tale signs of account sharing if you have a decent BI tool in place or some access to server logs. This really needs a statistician to analyse – and is beyond the scope of this post- but if you want to properly understand how to identify account misuse you should read up on the Gaussian Mixture Model and similar clustering techniques.

In a nutshell, you will look for patterns in accounts that have an unusually high number of concurrent sessions or access the service from more IP addresses or geographies than normal users.

What proportion of piggyback users could be converted to customers?

This question is also non-trivial. You do know these people actually consume your product, though, so it’s a fairly safe bet that they would convert at least as well as an MQL – and you probably have that ratio already.

If you want to get  a better feel for this you are going to have to run some experiments. You’ll need to be creative and this will take investment. One tactic, for example, is to identify the 1000 accounts which use the most IP addresses then target them with a promotion to “invite a colleague and they get 1 month free”. If you can track the signups from that campaign you have a list of leads that are likely to have previously been piggyback users – then you just need to sell to them and track your success!

What would it have cost to identify the piggyback users as net-new leads?

This should be a fairly easy calculation, as if they weren’t piggyback users they would be marketing targets. What’s your marketing spend per MQL? This should be a known stat.

Overall cost of account sharing

Once you have answered the questions above you can have a good stab at calculating the cost of account sharing.

The lost potential revenue is the approximate number of piggyback users x the conversion rate x your ARPU.

The cost of realizing that revenue is your cost per MQL x the approximate number of piggyback users.

The difference is your potential return.

However, depending upon the capabilities of your tech stack, there might be more investment needed to get there. You would need to bake that into your model. The capabilities needed are explored in the next part but some solutions (like Zephr) provide many of these out-of-the-box, so you may not need additional investment.


Working out the business case for tackling account sharing might be difficult for you to take on or it might be so obvious you don’t even need to back it with data. The chances are that all B2B publishers (and all subscription businesses) suffer due to misuse of accounts to some degree of another.

There are a few steps you can take to limit account sharing and I’ll explore those in the next post in this series.

Leave a Reply

Your e-mail address will not be published. Required fields are marked *