Teardowns and before/after case studies·

The Trial-End Screen Fix Nobody Ships: A Working "Stay on Free" Link

The Trial-End Screen Fix Nobody Ships: A Working "Stay on Free" Link

Every SaaS teardown I do starts the same way: the founder points at the pricing page, I open the app, and the actual leak turns out to be somewhere else.

This one was a day-14 trial-end modal. Not the pricing page. Not the plan grid. The screen that appears when the trial clock hits zero and the product decides whether the user becomes a customer or a closed tab.

Details are anonymized and the numbers are directional — the mechanism is the point, not the exact percentage. If you want the same read on your own flow, we publish teardowns like this on the /blog regularly.

The flow under review: a day-14 trial-end modal

The product: a small B2B tool for ops teams, roughly $29–$79/month, no credit card at signup, 14-day trial.

When the trial expired, the user hit a full-screen modal inside the app. Two buttons. No close icon. No link. Behind it, their account still existed, but every route bounced back to the modal.

The team's read on the problem was "people don't see enough value in 14 days." Their fix was going to be a longer trial.

Their read was wrong. The screen was the problem.

What the original screen actually said

Here's the before, close to verbatim:

Your trial has expired Upgrade now to continue using [Product]. Contact sales for team plans. [ Upgrade now ] [ Talk to sales ]

Four things were broken, and none of them were the price:

  • It never said what happened to their data. The single biggest unspoken question at trial end is "do I lose the 14 days of work I just did?" The screen answered it with silence. Silence reads as "yes."
  • There was no exit. A modal with no dismiss, no back, and no link is a wall. Users don't argue with walls — they close the tab, and once the tab is closed, the decision is made for good.
  • The button hierarchy was a coin flip. Two filled buttons of equal weight means the product hasn't decided what it wants. "Talk to sales" on a $29 product is a dead end dressed up as an option.
  • It gave no guidance on what happens next. "Upgrade now" leads to a plan grid with three tiers and a billing toggle. That's a second decision stacked on a first one, at the exact moment the user is least patient.

The team was measuring trial-to-paid and seeing a number they didn't like. What they weren't measuring was how many people hit that modal and left the session within ten seconds. That number was the real story.

The change: one link and three lines of copy

The fix took under an hour. They didn't touch the pricing page, the trial length, or the plan tiers.

What changed:

  1. Added a real "Continue on Free" text link under the primary button — not a greyed-out button, not a "No thanks, I hate saving time" link. A working path that downgraded the account to a capped free tier: read access to everything they'd built, plus a small monthly allowance of new items.
  2. Rewrote the headline and body to answer the data question first, before asking for money.
  3. Pulled two real numbers from the account into the body so the screen wasn't generic.
  4. Deleted the second filled button. "Talk to sales" moved to the pricing page where it belongs.

Before:

Your trial has expired Upgrade now to continue using [Product]. Contact sales for team plans. [ Upgrade now ] [ Talk to sales ]

After:

Your trial ended. Your work is still here. In 14 days you created 4 boards and invited 2 teammates. All of it stays saved either way. Pick a plan to keep adding boards and teammates — or stay on Free and keep read access to everything, plus one new board a month. [ Upgrade — from $29/mo ] Continue on Free No card on file. Cancel anytime. We email you before anything is ever deleted.

The primary CTA got stronger in the rewrite, not weaker — a specific price beats a vague "Upgrade now," and the free link removed the reason to bail before reading it.

If you'd rather not guess which element of your trial-end screen is doing the damage, run a free audit on your trial-to-paid flow — it flags hierarchy, dead ends, and missing reassurance in a few minutes and hands you a prioritized list.

Why an escape hatch raises paid conversion

This is the part that makes founders nervous, so let's be precise about the mechanism. It isn't that free users pay more. It's that the wall was killing people who would have paid later.

  • Clarity beats persuasion at the decision point. Nielsen's visibility-of-system-status heuristic isn't about spinners — it's about the system telling the user what state things are in. "Your work is still here" is system status. Without it, the user is guessing, and guessing at trial end means leaving.
  • Trust collapses when there's no exit. A screen with no dismiss reads as a hostage situation, and small teams are the most allergic to that feeling. They've all been burned by a tool that held their data. An exit that actually works is a trust signal, and trust is what makes the primary button clickable.
  • Forced choice produces avoidance, not compliance. When the only options are "pay" or "lose everything," a large share of users pick a third option you didn't offer: close the tab and never come back. Removing the trap gets them past the screen and back into the product, where the upgrade prompts can be contextual instead of confrontational.
  • The free path gives you feedback you were blind to. Before the change, the team had no idea who ghosted. After, "Continue on Free" clicks became a segment they could watch, email, and re-target at the exact moment they hit the cap — which is a far better upgrade prompt than a modal.

The lift came from people who were already close to paying and got stopped by uncertainty. You were never going to convert the rest at that screen anyway.

The P0/P1/P2 playbook for your trial-end screen

P0 — do this week, no test needed:

  • Give the modal a real exit. If you don't have a free tier, use "Remind me in 7 days" (once) or "Export my data." A disabled or fake link is worse than no link.
  • State the data outcome in the headline. "Your work is still here" or "Nothing gets deleted" — not in the fine print, in the headline.
  • Collapse to one primary button and one plain text link. Kill every second filled button.
  • Remove "Talk to sales" from the trial-end modal unless your ACV genuinely requires it.

P1 — next two weeks:

  • Insert one line of real usage from the account (items created, teammates invited, whatever maps to value). Generic modals get generic responses.
  • Instrument the modal: views, primary clicks, secondary clicks, and exits within 10 seconds. That last one is your actual leak rate.
  • Add the price to the primary button. "Upgrade — from $29/mo" filters out the people who were never going to pay and de-risks the click for everyone else.

P2 — once the above is live:

  • Build a capped free tier so the downgrade is real, not a data export.
  • Trigger contextual upgrade prompts at the cap, not on a timer.
  • Segment the "Continue on Free" cohort and track 30/60/90-day conversion. Some of them will pay — you just moved the decision to a moment when they had a reason.

One warning on testing: if you're doing a few hundred trial expirations a month, an A/B test on this screen will take a year to reach significance. Do the heuristic review, ship the P0s, and watch the 10-second-exit rate instead.

What to measure before you call it a win

  • Trial-end modal → upgrade (primary conversion, the headline number)
  • Trial-end modal → session end within 10 seconds (the ghost rate — this should drop first)
  • "Continue on Free" clicks (new segment, new opportunity)
  • Support tickets containing "can't access" or "lost my data" (should go to near zero)
  • Free-cohort conversion at 30/60/90 days (the number that proves the escape hatch wasn't charity)

If the ghost rate doesn't move, your problem isn't the copy — it's that the exit still isn't real, or the data reassurance isn't where people look.

Stop treating trial end as a last chance to extract money. It's a status update with a purchase option attached, and the clearer you are about what survives, the more people click the button that pays you.


Start a free FlowAudit at /signup and you'll get a prioritized P0/P1/P2 fix list for your own trial-to-paid screen in minutes — no call, no consultant, no 40-slide deck. Just the specific things to change before your next cohort expires.