# 90 Day Pricing and Packaging Refresh

> A 90 day repricing project: research, modelling, tier design, internal approvals, billing migration, and launch, with owners and artifacts for every week.

Source: https://saas-marketing.net/playbooks/pricing-and-packaging-refresh/
Topic: SaaS Pricing
Type: playbook
Published: 2026-09-11
Last updated: 2026-09-11
Publisher: SaaS Marketing (saas-marketing.net)
License: CC BY 4.0. Quote or republish with attribution and a link to https://saas-marketing.net/playbooks/pricing-and-packaging-refresh/

## Short answer

A pricing and packaging refresh runs in four phases across 90 days. Days 1 to 20 are discovery: usage data, win loss pricing objections, a competitor price book and 12 to 20 customer interviews. Days 21 to 45 cover value metric selection, tier design and a three scenario revenue model. Days 46 to 60 are approvals across finance, sales leadership and legal. Days 61 to 90 cover billing build, entitlements, site updates and a monitored launch.

## Key takeaways

- Start with the value metric and the tier map, not the price points, or you will rebuild the model three times.
- Ship packaging changes before price changes so you can separate migration noise from price elasticity signal.
- One named owner with finance and sales sponsorship is the single best predictor that the project ships.
- Budget 20 working days for billing and entitlement work alone, because it is almost always the critical path.
- Model cannibalisation explicitly: assume a share of existing customers drop a tier when you add a cheaper one.
- Grandfather existing contracts for at least one renewal cycle unless finance has agreed to absorb the churn risk.

---

Most repricing projects die in week six, and they die the same way. Someone opens a spreadsheet, types a new number in the Pro tier cell, and the next four months are spent arguing about whether it should be 79 or 89. The number is the last decision, not the first.

What follows is the 90 day shape that works: discovery, design, approvals, build, launch. Each phase ends with an artifact and a named signer. If a phase ends without both, you have not finished it, you have just run out of calendar.

## Why start with packaging rather than price

Because packaging changes are reversible and price changes are not, politically. If you move a feature between tiers and it lands badly, you move it back in a sprint. If you raise prices and churn spikes, you cannot un-raise them without signalling that the first number was made up.

There is also a measurement argument. Ship packaging and price together and you get one blended result you cannot decompose. Ship packaging in Q1 and price in Q2 and you can see plan mix shift separately from price acceptance. That sequencing costs you a quarter of incremental revenue and buys you a decision you can defend to a board. Teams under severe runway pressure will take the blended hit, and that is a fair call. Everyone else should split it.

A pricing refresh that produces a beautiful tier map and never ships, because no one had authority to book engineering time. Get the eng commitment in writing during discovery, not during build.

The broader framing lives in our [SaaS Pricing Strategy](/saas-pricing/) hub. This page is the project plan.

## Days 1 to 20: discovery, and what the data pull has to include

Discovery answers one question: what are customers actually paying for, and what are they paying for that they do not use. Four workstreams run in parallel.

The usage pull comes first. For every account, export the last 12 months of the two or three candidate value metrics: seats provisioned versus seats active, records stored, events sent, workspaces created. You are looking for the metric that correlates with retention and expansion, not the one that is easiest to bill. A metric that grows while the customer's value does not is how you end up with the Datadog style bill shock stories that circulate on Hacker News every few months.

Second, pull pricing objections from win loss. Not the CRM dropdown, which says "price" for everything. The call recordings. Gong or similar will let you search for "too expensive", "budget", "quote", "per seat" across six months of lost deals, and the actual language is usually more specific: it is not that the price is high, it is that the buyer cannot predict next year's bill or cannot justify paying for 40 seats when 12 people use it.

Third, build a competitor price book. Public pages, captured as screenshots with dates, plus whatever your sales team has seen in competitive deals. Record list price, the value metric, the free tier boundary, annual discount and what sits behind "contact us".

Fourth, run 12 to 20 customer interviews. Split them across current customers on different tiers, recent closed won and recent closed lost.

**The discovery interview structure that gets past politeness**

Artifact at day 20: a discovery memo, ten pages maximum, with the usage analysis, the objection themes, the competitor price book and the interview synthesis. Signed off by the owner and the exec sponsor. No recommendations yet.

## Days 21 to 45: design the value metric, the tiers, then the numbers, in that order

Design starts with a single decision: what does the price scale on. Everything downstream depends on it. Seats work when your product is used by named humans doing individual work, which is why Figma, Notion and Linear all price that way. Consumption works when usage tracks value directly and the customer can forecast it, which is Twilio and Stripe. Hybrid models, a platform fee plus usage, are now common and are harder to explain on a price page than teams expect.

Once the metric is fixed, draw the tier map. Three paid tiers plus a free or trial entry is the default for a reason, and our guide to [Good Better Best Packaging for SaaS](/guides/good-better-best-packaging/) covers why the middle tier does most of the work. If you run a self serve motion, read [Pricing for Product Led Growth](/guides/plg-pricing-and-free-plan-design/) before you set the free plan boundary, because that boundary is a packaging decision that determines your entire funnel shape.

For each tier, write down the job the tier exists to do, the buyer persona, the gating logic and the upgrade trigger. If you cannot name the upgrade trigger, the tier boundary is decorative. The mechanics of drawing those lines are in [SaaS Packaging and Tiering](/guides/saas-packaging-and-tiering/), and the underlying mechanism has a definition page at [Feature Gating](/glossary/feature-gating/).

### The revenue model, with the assumptions on the surface

Build three scenarios. Not because you believe any of them, but because the spread is the conversation finance needs to have.

| Assumption | Conservative | Base | Aggressive |
| --- | --- | --- | --- |
| New business ASP change | +5% | +14% | +25% |
| Cannibalisation (accounts moving down a tier) | 18% | 11% | 5% |
| Incremental churn at renewal | 4 pts | 2 pts | 0.5 pts |
| Expansion from new usage metric | +3% NRR | +7% NRR | +12% NRR |
| Sales cycle impact | +12 days | +5 days | 0 |
| Net ARR effect, month 12 | -1% | +9% | +21% |

Two rules about this table. Every cell must have a named source: an interview, a benchmark, a prior launch, or an explicit guess labelled as one. And cannibalisation must be non-zero. Every model I have seen that assumed zero cannibalisation from a new lower tier was wrong within two quarters.

**11%** Base case share of existing accounts that move down a tier when a cheaper option appears

Artifact at day 45: the tier map, the price points, the revenue model workbook with the assumption tab visible, and a one page migration policy covering grandfathering.

## Days 46 to 60: approvals, and the four people who can stop this

Approvals fail when they happen in sequence. Run them in parallel with a hard date.

| Approver | What they need to see | Typical objection | Time to clear |
| --- | --- | --- | --- |
| CFO or finance lead | Three scenario model, cash timing, billing system cost | Cannibalisation looks optimistic | 5 to 8 days |
| Sales leadership | Quota and comp impact, deal desk rules, discount floors | Mid quarter disruption to in flight deals | 5 to 10 days |
| Legal | Contract language, notice periods, auto renewal terms | Existing MSAs may fix pricing for the term | 7 to 14 days |
| Board or exec team | One page narrative, scenario range, launch date | Why now, and what happens if it goes wrong | One meeting |

Legal is the one people underestimate. Enterprise MSAs frequently contain price protection clauses, and someone has to actually read them. On a 300 customer base, budget a week of paralegal or ops time to categorise contracts into can change now, can change at renewal, and cannot change.

The sales conversation is its own project. Reps in the middle of a quarter with a quote out at the old price will fight a launch date, correctly. Set a quote honour window, 30 days is standard, and publish it before the announcement rather than after the first escalation.

Presenting one number. Boards approve ranges and reject point estimates they cannot interrogate. Show the conservative case and say plainly what you would do if it lands there.

Artifact at day 60: signed approvals, the deal desk rulebook, and the customer communications plan. If you are raising prices on existing accounts, the sequencing of those messages has its own playbook at [Price Increase Rollout Playbook](/playbooks/saas-price-increase-rollout/).

## Days 61 to 80: build, where the schedule actually breaks

Build is billing, entitlements, CPQ, site, docs and support. It is almost always the critical path, and it is almost always underestimated because the pricing team does not own the systems.

**Build phase completion criteria**

Two things go wrong here reliably. Entitlements drift from the tier map because an engineer made a reasonable local decision three weeks after the spec was written. And proration behaviour surprises everyone, particularly on annual plans with mid term upgrades. Test with real account shapes, not synthetic ones.

Tool choice matters less than most vendors suggest, but it is not nothing. If you are on Stripe Billing with a simple seat model, this phase is short. If you have a homegrown entitlement service and a Salesforce CPQ instance nobody understands, double your estimate. We cover the category in [SaaS Pricing and Billing Tools](/tools/saas-pricing-software/).

## Days 81 to 90: launch and the numbers you watch daily

Launch day is dull if the previous 80 days went well. What matters is the monitoring window.

Watch these daily for 30 days: plan mix on new signups, trial to paid conversion by plan, quote to close rate, average discount, downgrade requests, and support ticket volume tagged pricing. Compare each against the pre launch baseline you captured in week one, which is why the discovery data pull mattered.

Set a rollback trigger before launch. Something concrete: if self serve conversion falls more than 20 percent against baseline for ten consecutive days, we revert the free plan boundary. Writing that down in advance is what separates a measured launch from a panic.

Expect a two to four week dip in self serve conversion as the funnel re-sorts, then recovery. If it has not recovered by week six, the problem is the tier boundary, not the price.

Run a formal readiness review before you flip anything, using the [Pricing Change Readiness Checklist](/checklists/pricing-change-readiness/). Teams that want to compress this whole sequence into a structured two week effort can work through the [SaaS Pricing Sprint](/courses/saas-pricing-sprint/).

## What this costs, honestly

A 90 day refresh at a 50 to 150 person SaaS company consumes roughly 0.5 FTE of the owner's time for the full quarter, 15 to 25 engineering days, a week of legal or paralegal time, two days of finance modelling and a full day of sales enablement. If you buy external help for research and modelling, 25,000 to 60,000 dollars is the common range.

The failure mode worth naming: companies that run this process, produce a good tier map, and then discover the engineering team cannot deliver entitlements for two quarters. You end up announcing packaging you cannot enforce, which is worse than not announcing at all. Scope the build before you scope the ambition.

## What to do next week

Pick the owner and get the exec sponsorship in writing. Then run the usage export, because it takes longer than anyone expects and it determines the value metric decision that everything else hangs from. Book the 20 customer interviews in week one so they land inside the discovery window rather than dragging into design.

## Frequently asked questions

### How long does a SaaS repricing project actually take?

Ninety days is realistic for a company under roughly 200 employees with a single product line. Larger companies with multiple products, regional price books and a CPQ layer usually need four to six months. The variable is not analysis time, it is billing system work and legal review of existing contract language.

### Should we change packaging and prices at the same time?

Prefer packaging first, prices second, separated by one quarter. Doing both at once means you cannot tell whether a conversion drop came from the new tier boundaries or the higher number. If the board demands one event, at least instrument the funnel so plan selection and price acceptance are measured separately.

### Who should own a pricing and packaging refresh?

A single named owner, usually a product marketing lead or a head of pricing where one exists, with an executive sponsor in finance and one in sales. Committees produce decks. The owner needs authority to book time from engineering, RevOps and legal, which means the sponsorship has to be explicit at the start.

### How many customer interviews do we need before repricing?

Twelve to twenty conversations across segments is usually enough to hit saturation on willingness to pay language and packaging confusion. Split them: current customers on two tiers, recent closed won, recent closed lost. Pair the qualitative work with usage data, because what customers say they value and what they use diverge often.

### What is a value metric and how do we pick one?

The value metric is the unit your price scales on: seats, contacts, events, workspaces, API calls, gigabytes. Pick the one that grows as the customer gets more value, is predictable enough to budget, and is visible in your product today. If you cannot instrument it within a quarter, it is the wrong choice for this refresh.

### Do we grandfather existing customers?

Usually yes for one renewal cycle, with a clear end date communicated at launch. Indefinite grandfathering creates a permanent legacy price book that costs more to maintain than the revenue it protects. Give sales an explicit early migration incentive so accounts move before the deadline instead of at it.

### What breaks most often during a pricing launch?

Entitlements. The price page and the sales deck ship on time, then customers on the new plan find they can still access a gated feature, or cannot access one they paid for. Test every plan and feature combination in staging, including downgrade paths, before anything goes live.