Tactical insights for first-time founders to outsmart the burn, the churn & the breakdown.

Hey Founder,

Is there a deal you signed years ago that you wish you could go back and change?

There’s a customer you needed badly once, so you said yes to everything: unlimited scope, a flat fee, ninety-day terms.

You made those concessions under pressure because, back then, you wanted the deal more than the deal wanted you.

Now their usage has tripled, but your price hasn’t moved since 2023. And you’ve been too grateful (or maybe too avoidant) to touch it.

Somehow, one of your most important customers has also become one of your worst commercial deals.

If that same contract landed on your desk today, would you still sign it?

This issue is about the concessions you’re still carrying: how to find them, calculate what they’re costing you, and reset the deal without losing the customer.

Let’s dive in.

The deal that saved you and the deal that’s bleeding you are often the same one.

A customer gave you a chance when you needed it, and you were grateful enough to give them a concession. Maybe that was the whole pitch.

That concession wasn’t a discount. It was a loan.

And you’ve been paying interest ever since: in unpriced support hours, scope you never charged for, and cash that arrives ninety days late.

They’re a great logo. They’re also why an implementation scoped for six weeks ran five months, and why two people on your team have a standing meeting nobody bills for.

You’ve thought about raising the price or capping the scope. But every quarter you wait, the gap between what they cost and what they pay gets wider.

Parseur, an email-parsing tool, still charges some early customers $9 a month on a discontinued plan that now lists at $129. That’s a 14x gap on accounts that were critical to the company’s survival.

For Parseur, that’s deliberate. Grandfathering long-term customers is policy, and they still get new features. For many founders, though, legacy pricing isn’t a decision. It’s a default.

There’s nothing wrong with rewarding an early customer if you know the cost and decide the relationship earns it. The problem is when you’ve never priced the concession or asked why it still exists.

And in SaaS, the expensive concession often isn’t the price. It’s the terms:

You want to repackage, move part of the bill to usage, maybe add an AI tier. But that means reopening six conversations you’d rather avoid. So the pricing stays untouched, and one old contract ends up capping what you can charge.

AI add-ons, usage tiers, new packaging, support plans that cost extra. SaaS prices rose 16.4% in the twelve months to June 2026, roughly four times US CPI at 4.2% (Zylo). 

So you’re not introducing a foreign idea, you’re joining a conversation customers are already having across their stack.

Microsoft raised prices across its 365 suites on July 1, bundling in AI and security features. Atlassian stopped selling new Data Center licenses to new customers in March and pointed them toward Cloud.

The useful part is how Microsoft handled existing customers: the new price kicks in at renewal, no mid-cycle surprise, the renewal does the work.

That’s Margin Move 4, run by a trillion-dollar company.

Bessemer puts AI applications at 50–60% gross margins versus 80–90% for traditional SaaS. Every AI customer brings variable costs: inference, model calls, data processing, human review and support.

That creates an uncomfortable dynamic: the more value a customer gets, the more they may cost you to serve.

If a legacy account is still paying a flat fee while usage has doubled or tripled, the deal gets worse as the customer gets more value.

Revisit the commercial model before the account becomes too important to touch. 

The promise was never “this price, frozen forever.” It was “we’ll keep working together.”

And a price that no longer reflects the value, usage, or cost of delivery eventually makes that harder, not easier.

Leaving old terms untouched can look like loyalty, but sometimes it’s just avoiding a difficult conversation with a customer you like. Sometimes it’s genuine generosity. Either way, the relationship only works if the deal stays fair for both sides.

So the conversation isn’t: “We’re going back on our word.”

It’s: “The version of this relationship we created in 2023 no longer matches the value, usage, or cost of delivery in 2026. Let’s update it fairly.”

Pull your ten largest customers. For each, write down:

Then ask: If this exact deal walked in today, would you sign it? Every “no” is a legacy concession.

Add the monthly gap across those accounts and multiply by twelve. That’s what the old terms are costing you annually. It’s often the size of a hire.

Flag two things:

Decide which customers you’d be willing to lose. Then choose the one or two you’ll deliberately keep on legacy terms because the relationship genuinely earns it.

That’s the difference between carrying a loan and choosing to forgive one.

For everyone else, pick the renewal where the concession ends. Every loan needs a term. Give this one one.

You don’t always need to raise the headline price. First identify what’s actually broken.

Re-scope fixes margin. Define what’s included - hours, seats, API calls, environments, support - and charge beyond it. 

Re-price fixes revenue. Size the increase before you make it:

Start with the cap. Restoring boundaries is often easier than asking for more money, and it can improve the economics before you spend goodwill. 

And whenever possible, trade rather than take. Pair the ask with something valuable: a longer rate lock, a feature they want, priority support or an extra seat. They can say yes without feeling like they simply lost. 

Don’t spring new pricing mid-cycle. Attach it to the renewal and give customers time:

Then make the message do four things:

A founder asked me:

Lead with the commit. It’s contracted revenue and the number investors will expect to see. Switching to consumption mid-raise doesn’t look conservative. It looks smaller.

But show three lines together:

Yours is 0.7.

Healthy usage-based books can sit closer to 90–95%, because the commit should land just below what customers actually use, giving them room to overrun and expand.

At 70%, your commitments are landing above consumption. If the cohort data confirms it, say what’s happening: you’re selling commitments customers aren’t burning through.

And don’t make investors calculate what actually renews. Show them:

Renewal-adjusted commit = committed ARR × (1 − expected churn) × contraction factor

Lead with the contract. Then show the usage underneath it and explain the gap before they have to ask.

Send it my way, just hit reply.

Resources worth bookmarking: 

Before you go…

You don’t have to choose between protecting the customer and protecting the business. Generosity and fairness can coexist.

A renegotiation is a chance to reset the relationship around the value you deliver today, rather than the circumstances you were operating under three years ago.

And when you fix the worst deal on your books, you don’t just recover margin. You recover support hours and capacity you can reinvest in the customers who value what you’re building, including that one.

That’s your real moat.

See you next Thursday,

— Mariya

Hit reply and let me know. I read every single one (for real).

1. Early stage? Start with one of my practical finance guides.

2. Stuck on something specific? Book a 1:1 and we’ll work through it together.

3. Past $2M ARR? Work with me and the team at FounderFirst to run your CFO function.

Hey, I’m Mariya, a startup CFO and founder of FounderFirst. After 10 years working alongside founders at early and growth-stage startups, I know how tough it is to make the right calls when resources are tight and the stakes are high. I started this newsletter to share the practical playbook I wish every founder had from day one, packed with lessons I’ve learned (and mistakes I’ve made) helping teams scale.

Mariya Valeva

Find me on LinkedIn