TL;DR: Some tax software vendors use "concurrent user" pricing, limiting how many staff can be logged in simultaneously. This looks cheap upfront, but firms consistently underestimate how many concurrent licences they actually need, especially during peak filing periods like January. The math shows that even a 2-licence deal for a 10-person team has a ~47% chance of hitting its limit at any given moment, and that jumps to near-certain during busy seasons. By the time you realise you need more licences, switching software isn't realistic. Always model your actual usage before buying. Or consider AbraTax that doesn't limit concurrent users.
While doing our user research, talking to tax agents and performing competitor analysis, we’ve discovered some software vendors selling tax filing software license their products using concurrent users as part of their pricing calculation.
A concurrent user is effectively how many people can be logged into the software product at the same time. Exceed the limit and another user in the organisation can’t log in and use it until their colleague has finished.
At first glance, these deals can look very attractive. From lessons in pricing it’s also quite clever.
A product offer might include just one or two concurrent users and quote pricing on a per-client basis that appears significantly cheaper than competitors. Even when software providers don't advertise it this way, we’ve witnessed accountants naturally calculating the effective cost per client when comparing products.
The trick is that it's surprisingly easy to underestimate how many concurrent licences you'll actually need.
The birthday paradox
The famous birthday paradox demonstrates how unintuitive probability can be.
Most people assume you need a very large group before two people are likely to share a birthday. After all, there are 365 possible birthdays.
In fact, a group of just 23 people already has around a 50% chance that two people share a birthday.
The birthday paradox isn't directly about software licensing, but it highlights an important lesson: our intuition is often poor when judging probabilities.
Applying the same thinking to concurrent users
We can build a simple mathematical model for software usage.
Feel free to skip this section to the results.
Using a binomial distribution, we can calculate the probability that more people need the software than the number of concurrent licences available.
Suppose:
- n = number of employees
- p = probability that any one employee needs the software at a randomly selected moment. For example, this could be to submit a return, perform a calculation or simply look up client information.
- c = number of concurrent-user licences
- X = number of employees needing the software simultaneously
The results are surprising
Varying the team size and concurrent user licenses purchase we can arrive at the following outcomes, showing the propbably of not having enough licenses:
For example, assume the following:
- each employee independently has a 25% chance of needing the software at any moment;
- the firm has 10 employees;
- only 2 concurrent licences have been purchased.
The model shows there is approximately a 47% chance that demand exceeds two licences at any randomly chosen point in time.
It gets much worse during filing season
Of course, software usage isn't constant throughout the year.
Anyone working in tax, especially personal tax, knows workloads are heavily focused around filing deadlines. Our own filing data clearly shows this pattern.
During January, for example, it would be reasonable to assume employees are using tax software far more frequently.
If we increase the probability of an employee needing the software from 25% to 75%, the picture changes dramatically.
Now the probability of exceeding the available concurrent licences becomes almost certain for most team sizes.
A team of ten people may effectively require close to ten concurrent licences to avoid bottlenecks.
For some vendors, that can mean paying nearly double the headline price.
Why this pricing model works
The problem is that firms often compare products using the advertised starting price without considering how many licences they'll actually need in practice.
Only after implementation - when staff begin using the system during busy periods - does the pressure to purchase additional concurrent licences become obvious.
At that point, changing software is rarely an attractive option. Procurement has already been completed, data migrated, staff trained and processes established.
The very nature of personal tax with the inherant time contrants as end clients deliver their records late etc means pressure is great.The upgrade becomes the easiest and the often the only choice available.
The takeaway
When comparing software vendors, don't just compare the advertised price.
Ask:
- How many people are likely to need the software simultaneously?
- What happens during January and other peak periods?
- How much would additional concurrent licences cost?
A product that appears cheapest on paper may become one of the most expensive once your team starts using it in earnest.
Always model your expected usage before choosing a concurrent-user licensing model.
Download our concurrent user licence capacity calculator (Excel) to estimate the probability that your team will exceed its available concurrent licences based on team size and expected software usage.
At AbraTax we don't restrict concurrent users. We offer both pay as you go style plans or users can purchase an annual subscription plan.