How do you compare SaaS competitor prices when billing units differ?

A competitor charging per seat and another charging per outcome cannot be compared by placing their headline prices side by side. Start with one customer workload, identify the lowest plan that supports it, and record each vendor's billing rules. Calculate a comparable cost only where the public evidence supports it.

For a founder researching five competitors, the useful result is a comparison that explains what makes each offer cheaper or more expensive under explicit assumptions. A single price ranking can hide those assumptions.

The examples below illustrate billing mechanics across different software categories. Slack, Notion and Intercom are not presented as substitutes or as a competitive shortlist for your business. Sources were accessed on 2026-10-18.

Define the customer workload first

Write a short scenario before collecting prices. Specify the customer type, number of people needing full access, required features, monthly usage and billing commitment. Use the same scenario for every competitor.

For example, a hypothetical support-software scenario might require five agents, 1,000 monthly conversations, a shared inbox and monthly billing. If a product bills for resolved conversations, total conversation volume alone is insufficient: the number of billable resolutions remains a separate assumption.

Choose requirements that reflect the audience you intend to serve. A plan designed for an individual is not a useful benchmark for a team that needs administrative controls. Likewise, an enterprise plan may include capabilities that your intended customer does not need.

Separate the billing unit from the headline price

For each competitor, collect these fields in a worksheet:

| Field | What to record |
| --- | --- |
| Plan and eligibility | Lowest publicly documented plan satisfying the scenario |
| Billing unit | Member, active member, workspace, event, outcome or another defined unit |
| Commitment | Monthly or annual term; payment schedule if published |
| Included allowance | Quantity, reset period and whether pooled or per user |
| Excess usage | Published overage rate, upgrade requirement or hard limit |
| Required extras | Add-ons needed to meet the scenario |
| Evidence | Exact page URL, access date and unresolved questions |

The definition of a unit can matter as much as its price. Slack's fair billing policy applies to subscriptions and add-ons bought through its website using a credit card or self-serve invoicing. It describes billing for active members and prorated credits for inactivity; those credits are non-refundable. This establishes a specific billing rule, not a rule that should be assumed for every enterprise agreement. Slack's fair billing policy (accessed 2026-10-18).

In your worksheet, distinguish the number of people in a customer's organization from the number the vendor would bill. If you cannot establish the latter, leave it unresolved.

Treat included features and included consumption separately

A feature can be available on a plan while its consumption incurs additional charges. Intercom's pricing FAQ describes full-seat charges for its helpdesk plans alongside usage charges for Fin and some messaging channels. It separately describes Fin used with an existing helpdesk as having no seat costs but minimum commitments. Those are different purchase configurations and should occupy separate comparison rows. Intercom Pricing FAQs (accessed 2026-10-18).

Record three different questions: Is the capability available? What consumption is included? What triggers an additional charge? A checkmark answering the first question does not answer the other two.

Also record the scope of a limit. Notion's pricing page describes unlimited file uploads on paid plans while specifying an approximately 5 GB maximum per file. Its page also lists different page-history periods across plans. These establish that an unlimited quantity and a limit on individual items or retention can coexist. Notion Pricing (accessed 2026-10-18).

For your own comparison, replace broad labels such as unlimited storage or advanced reporting with the exact published allowance or capability. If the wording remains ambiguous, preserve that uncertainty.

Calculate comparable costs only after the rules are clear

A starting model is: recurring base charge, plus billable seats, plus chargeable usage, plus required add-ons. Apply each vendor's actual rules for included allowances, tiers, minimum commitments and caps. Do not assume every price follows a simple linear formula.

Here is a hypothetical example using invented prices solely to demonstrate the calculation:

| Assumption or result | Hypothetical Plan A | Hypothetical Plan B |
| --- | --- | --- |
| Base pricing | $30 per seat per month | $100 per workspace per month |
| Five seats | $150 | Included |
| Included events per month | 1,000 per workspace | 500 per workspace |
| Additional events | $0.10 each | $0.20 each |
| Total at 800 events | $150 | $160 |
| Total at 2,000 events | $250 | $400 |

These are not vendor quotes or forecasts. They assume identical event definitions, sufficient features, monthly billing and no other charges. The exercise shows why the lower base price need not produce the lower scenario cost.

Run a small, expected and larger workload through your worksheet. Change one assumption at a time so you can identify whether team size, usage or a required upgrade changes the result. If vendors count different events, do not equate them without an explicit mapping.

For annual offers displayed as a monthly equivalent, retain both that equivalent and the stated commitment. Do not describe an annual commitment as a cancellable monthly plan. Record currency and whether tax treatment is stated; avoid combining different currencies into one ranking without a dated conversion method.

Keep unknowns visible

Use distinct evidence labels: published, calculated from published terms, assumption, and not established. A calculated figure should point back to both its source inputs and its workload assumptions.

If a competitor requires a quote, record quote required rather than zero. If an overage rate is absent from the pages reviewed, say not found in reviewed pages. That does not establish that overages are free or that the vendor never publishes the rate elsewhere.

For a bounded five-site research pass, a useful page selection is the homepage, pricing page and the most relevant billing or limits page for each competitor. Record conflicts between those pages rather than silently selecting whichever wording makes a cleaner comparison. A narrow review can produce useful evidence while still leaving questions unanswered.

Turn the comparison into a positioning hypothesis

Connect each pricing observation to the customer scenario. For example: a flat workspace fee may be easier for a growing team to budget for, if usage allowances cover its workload. That is a hypothesis to test, not proof of buyer preference.

Public pages establish what competitors publish about their offers. They do not establish their realized selling prices, customer satisfaction, margins or willingness to pay for your product. Use the comparison to identify specific questions about your packaging rather than treating it as evidence that you should copy a competitor's price.

Before acting, check that all five rows use the same workload, that annual and monthly commitments remain distinguishable, and that every numeric result is either sourced or explicitly hypothetical. A useful comparison should make its limitations easy to inspect.

Get a bounded five-site research report

The Five-site competitor research report costs $75 USD and covers exactly five public competitor websites, with up to three public pages per site. It provides a source-linked comparison of positioning and published pricing in a Markdown report and CSV, with independent automated quality review and one factual correction. The delivery window is 120 hours.

Provide five distinct public HTTPS competitor URLs and public, nonconfidential context about your audience and what you want to compare. The report provides a documented public-information baseline; it does not include private data or guarantee business outcomes. You can use the method above to apply your own workload assumptions to that baseline.