.png)
1 hour ago
2
Press enter or click to view image in full size
Pricing in SaaS looks easy from the outside. Pick a number, put it on the website, and start charging.
In reality, it’s one of the trickiest decisions you’ll ever make. The way you charge doesn’t just impact revenue. It shapes the kind of customers you attract, how they behave, and ultimately what kind of company you end up building. Moreover, it’s very hard to change the price for an existing customer later on.
A quick disclaimer before I go further. Everhour was our very first product. We launched it back in 2015 with no investors behind us. A lot of decisions were made on intuition or based on advice that later turned out to be wrong. Some of the mistakes I’ll describe may sound obvious now, but they weren’t obvious to us back then. We learned by doing, and often by failing.
At Everhour, we’ve been experimenting with pricing for a decade. Free plans, different tiers, cheap versions, flat fees, grandfathering, discounts, minimum seats — you name it, we tried it. Some moves worked, some failed spectacularly, and a few became principles we’ll never break.
Here’s our story of what we experimented with, what worked, and what we’ll never repeat. And I’d be very happy to hear what worked and did not for you.
Starting with tiered pricing
When we turned on monetization in 2015, our first model was tiered pricing. It looked neat and “fair.”
Press enter or click to view image in full size
Inside a tier, customers were happy. They could add people without paying more, and the cost per user dropped as they grew.
The pain came when crossing tiers. A team with 15 users paid $39. Add one more user and suddenly the bill jumped to $119. Customers hated it. They refused to upgrade, asked for discounts, or left.
For us, it meant revenue didn’t grow with customer usage. We had happy customers inside tiers, but almost no one was crossing into the next one. That’s when we realized tiered pricing punishes growth instead of rewarding it. We decided to grandfather old accounts but move all new customers to a different model.
Switching to per-user billing
Next we tried per-user billing. This made more sense. As our customers grew, we grew with them. It felt fair, simple, and sustainable.
But then we hit a new problem. Solo users and very small teams. Because of how Everhour integrates with project management tools, even a two-person team could sync hundreds of projects and thousands of tasks. Our system was overloaded while these customers paid almost nothing.
Press enter or click to view image in full size
On top of that, they were often the toughest users. If something went wrong, they didn’t hold back. Complaints on social media, refund requests after short outages, aggressive support tickets. Ironically, the bigger customers were calmer, more patient, and more grateful.
So we introduced a minimum of 5 users. It was basically a usage fee: you could use fewer seats, but you’d still pay as if you had 5. That filtered out the freelancers and raised our average cheque. Some people complained we were “too expensive” for individuals, but the truth is Everhour was never designed for one or two people. Those who liked the product and could afford it kept paying.
Press enter or click to view image in full size
Later, we added a free plan for up to five users, but without integrations. That solved two problems at once: small teams could decide to stick around, and we got visibility in “Top X Free Tools” blog posts and resources. The free plan turned out to be less about monetization and more about marketing.
The Lite plan that went nowhere
At some point we thought we could save the freelancer crowd with a Lite plan. 2 users minimum, 10 maximum, cheaper seat price, but limited features and integrations — mainly Trello, Notion, and Todoist. Tools that are most popular among micro teams anyway.
Press enter or click to view image in full size
Our theory was that more small teams would sign up, and the volume would make up for the low price. Reality proved different. Support costs stayed high and conversions were flat. In the end, it wasn’t worth the hassle. We shut it down and again grandfathered old accounts.
Flat fee experiment with Trello
We also tried flat fee pricing in our Trello power-up we built as a side project. This power-up worked fully inside Trello’s ecosystem and covered only the most essential time-tracking needs — features that small teams actually used in our main product, without the extras.
Press enter or click to view image in full size
We priced it at $10 per month, unlimited users and projects, plus a limited free version. The logic was simple: keep the price low, remove friction, and grow by volume.
Again, it didn’t work. Out of more than 30,000 qualified users we only converted around 500 paying customers. Free users disappeared overnight once they hit a limit. The product was too simple to justify multiple tiers or upsells. Maybe with per-seat billing we could have earned more as some teams grew, but we chose not to complicate things and didn’t want to sink more time into it.
Why we always grandfathered
One thing we never compromised on was grandfathering. If you signed up at an old price, we never forced you to switch. You could stay on your plan indefinitely. The only exception was if you canceled and came back later — then you’d have to join the new plans. Even for accidental cancellations due to failed payments, we restored the old plan through support.
This policy built a lot of trust. Many customers still pay us at prices that don’t exist anymore, and they’ve stayed loyal for years. For them, it’s a no-brainer: they’re getting value at a price no competitor can offer today. For us, it’s been one of the best long-term retention strategies.
Multiple plans and upsells
Of course, we also tried the classic playbook of offering multiple plans: Free, Lite, Basic, Team, Enterprise. Everyone does it.
But in reality, almost all of our customers picked the same plan. The other plans barely mattered. What they did add was complexity, both in the product and in our messaging. And it always felt like we were trying to push people into paying more by holding back features.
We didn’t like that. Some SaaS companies thrive on upsells and aggressive tiering. For us, it felt fake. We preferred one honest plan with everything included. Customers could trust that the price they saw was the price they’d keep, without surprises or hidden traps.
Press enter or click to view image in full size
We also tried offering upsells to grandfathered accounts. As we added new features, we encouraged older customers to move to the current plan, even offering discounts for loyalty.
Press enter or click to view image in full size
A fairly small percentage switched. Their old plan gave them everything they needed, and psychologically the barrier of paying more was too strong. If the core product works, customers won’t pay extra just because you want them to.
Playing with discounts
We weren’t big on discounts, but we did experiment. Personally, I always believed discounts should help you earn more, not less.
Volume discounts for very large clients worked fine and helped us close deals. That felt fair.
What surprised us was urgency campaigns. At the end of certain months we’d run a promo like “Sign up in the next five days and get 20% off.” Conversions spiked a little, but the real magic was in retention. These customers stayed longer. They knew they had a deal and didn’t want to lose it.
But we never made it a regular thing. It felt unfair to existing customers who had signed up earlier at full price. We even saw some people trying to cancel and resubscribe just to get the discount, but we only applied it to new accounts. After a while, we decided to stop running these promos.
Seats instead of users
We started with billing per user. Add a user to the system and we billed you with the next billing cycle. Remove them and we prorated credit back to your balance.
It sounded logical, but in practice it was messy. Some customers canceled before the next billing cycle and we lost money on those deferred prorates. Others hated seeing too many small prorates stacking up into larger amounts — it felt like they were paying more than expected, and sometimes they didn’t even recognize the total when it showed up on their bank statement. For monthly subscriptions it wasn’t either good nor critical, but for annuals the sums were significant. Our losses on non-renewals were painful.
We fixed this by charging the prorate immediately when a user was added, instead of deferring it to the next cycle. But then we faced another wave of complaints: too many micro-payments and invoices. Customers often wanted to remove someone and instantly replace them with another user, or they added several people within just a few hours and didn’t want to receive multiple separate invoices.
So we moved to billing seats instead of users. Now a company can buy, say, 10 seats and the invoice stays the same each month. Or they can purchase an extra 3–5 seats in advance and later invite users to fill them — no prorates, no noise.
It’s not perfect. Some customers forget to remove unused seats and later complain we’ve been charging them unnecessarily, even though we make empty seats very visible in the interface and the quantity is always shown on invoices. Still, if they contact us, we usually refund the difference for the last payment.
Overall, though, this model works for the majority and has become the de facto standard across SaaS. And if you look at how bigger players like Asana or Monday handle it, they’re even stricter — you can only add seats in increments of five, and you can only remove them if the total remains divisible by five. We don’t impose such limits, which makes our approach more customer-friendly.
We often see companies starting a trial with 2 people and then, right after converting, buying 20-30 seats. They fill them gradually over the next few weeks. That wouldn’t have worked smoothly with per-user billing — especially in companies where every expense needs separate approval.
For the business, seats have been the right choice. It’s not flawless — nothing ever is — but it solved more problems than it created.
What we took away
After all these experiments, here’s what stuck with us.
- Tiered pricing might look fair and even make customers happy when they’re at the top end of a tier, but it always makes them angry the moment they have to cross into the next one.
- Per-user billing works great, but expect lots of tiny checks and the wrong audience to dominate. A minimum fee helps filter that out.
- Free plans can be useful, but mostly for marketing. Don’t count on upgrades — think about virality or visibility instead.
- Lite and flat-fee models sounded attractive, but in practice they were tough for growth.
- Grandfathering built trust and loyalty, but upselling grandfathered customers is hard.
- Multiple plans and aggressive upsells felt fake.
- Discounts worked only when tied to urgency or scale, and even then, you need to be careful not to upset your existing customers.
The biggest lesson? Pricing doesn’t just decide your revenue. It decides who your customers are, how they treat you, and what kind of company you build.
What we didn’t try
Some pricing models never appealed to us.
For example, charging based on customer revenue (like ChartMogul). It doesn’t really fit our case, but I just don’t like this model overall.
We also avoided usage-based pricing tied to API calls or account activity.
We never tried monetizing through ads or selling anonymized data — and we never plan to.
And unlike many SaaS companies, we never forced existing customers to accept price increases. I know some businesses calculate churn vs. uplift and decide it’s worth it. We’ve stayed away from that path.
