What is a flat rate bounded by?

Something, always. Common boundaries are a provisioned instance size, a storage or bandwidth allowance, a user count, a matter count, or a fair-use ceiling with published overage terms. A rate advertised with no stated boundary has one anyway, written somewhere less prominent, and finding it is the first thing to do with any quote.

Does a flat rate mean unlimited data?

Rarely, and the claim is worth reading closely. Unlimited normally attaches to one dimension, most often users, while another dimension carries the real ceiling. What matters is which dimension is capped, what the published rate is for crossing it, and whether crossing it renegotiates the agreement or simply adds a line.

When does a flat rate cost more than a metered one?

Below the crossover, which is a volume rather than a date. A small matter that closes quickly pays a full period under a flat rate and a small metered bill under a per-gigabyte one, so the fixed model loses. The cost calculator computes where that crossover sits on your own figures, and the per-GB cost guide works through the arithmetic behind it.

Flat-rate pricing in eDiscovery: one meter, which is time

Flat-rate pricing charges a fixed amount for a fixed period, whatever the volume. The meter is time rather than data or headcount, which means the bill is known in advance and moves only when the term does. That is the entire proposition: not that the total is lower, but that the total is knowable before the matter reveals how big it is.

How does flat-rate pricing work?

One number, one period, and the variables that move every other model held outside the calculation.

Under per-GB pricing the bill tracks how much data exists, so it moves whenever the other side produces. Under per-user pricing it tracks how many people have access, so it moves whenever the review team grows. A flat rate severs both connections. Data volume and headcount become operational decisions rather than billing events, which is why the model changes behavior as much as it changes arithmetic: nobody has to ask what a further custodian costs.

The trade is that the provider absorbs the variance it used to pass on. It prices that risk into the fixed number, which means light matters subsidize heavy ones. A firm whose matters are consistently small is paying for headroom it never uses, and should say so rather than assume the fixed model is automatically the disciplined choice.

Every flat rate is therefore bounded. A rate with no boundary is not a pricing model, it is an unpriced liability, and no provider carries one indefinitely. The boundary is the substance of the offer.

What does flat-rate pricing look like in practice?

The boundary sits on one of a small number of dimensions, and which one it is determines what kind of surprise is possible.

Where a flat rate is commonly bounded, and what crossing the boundary triggers
Boundary	What it caps	What happens at the cap
Provisioned instance	Compute, memory and storage of the environment	An upgrade tier, quoted separately
Storage allowance	Total data held at once	A published overage rate, or a bundle
Bandwidth allowance	Data moved in and out over a period	A published overage rate
User count	How many people hold access	A per-seat charge above the included number
Fair use	Usage judged against contracted scope	Renegotiation, which is the least predictable outcome

The last row is the one to press on. A numeric cap with a published rate is still forecastable, because crossing it produces a number you can calculate in advance. A qualitative cap produces a conversation, and a conversation is exactly what the model was bought to avoid.

The market is moving in this direction unevenly. In the Winter 2026 eDiscovery Pricing Survey, four of the 53 respondents reported alternative pricing models for remote collection, which the analysis reads as some providers moving toward flat-fee or subscription arrangements. Four of 53 is a very small base and is worth reading as a signal rather than a proportion.

How does flat-rate pricing show up on an eDiscovery invoice?

As one line, which is the point of it.

There is no per-gigabyte hosting charge that grows each month, no per-seat total that moves with the review team, and no separate production or export fee, because those are the meters the model removes. What can still appear alongside it are charges that sit outside the platform entirely: forensic collection, contract review labor, and anything the agreement lists as an overage above a stated allowance.

That makes the invoice easy to read and shifts the difficulty to the contract, which is where it belongs. The questions are what period the rate covers, what the boundaries are on each dimension, what the published rate is for crossing each one, what a renewal looks like, and what it costs to export everything and leave. The Fee Field Guide decodes the metered fees this model replaces, and the pricing page publishes a rate of this shape in full.

Sources

ComplexDiscovery, A Complete Analysis of the Winter 2026 eDiscovery Pricing Survey, conducted with EDRM across December 2025 through February 2026 on a cohort of 53 respondents, for the movement toward flat-fee and subscription arrangements. The survey documents that these models are in use; it does not define them.
The Sedona Conference, Glossary: eDiscovery & Digital Information Management, for the surrounding vocabulary.
Fee categories this model replaces: the Fee Field Guide dataset.

Flat-rate pricing