Pricing Software You Used to Sell as a Service

TL;DR Pricing software you used to sell as a service is a licensing decision before it is a price decision. The unit comes first, because a product priced off a services rate card inherits the hour, and the hour falls as the firm works faster. Then the gate between advisory and product, a packaging decision most firms set by accident. Then the shape of the price, and the migration of the clients who already have the tool. Firms pointed at a sale have one more reason to do the work early: instrumented pricing transfers with the asset.

A services rate card cannot price a product

Your firm prices labor well. Rate cards, scoped statements of work, a blended rate that carries the risk of a long engagement: those instincts were built over years of practice and they are sound for people and hours. Then a model your team built to solve one client’s problem went out inside three engagements, and a fourth client asked what it costs on its own. Nothing in the rate card answers that.

The product was built once. It now runs for every client who licenses it, and the cost of serving the tenth client is nothing like the cost of serving the first. A price reasoned from hours treats the product as if it were delivered by hand each time. The situations we see most often are laid out on the services firms page; the argument underneath them follows.

Why the hour is the wrong unit for a product

The hour measures the firm’s input. A product’s value sits on the client’s side of the table, in the scenarios run, the cases processed, the reports produced without a consultant in the loop. When the price attaches to the input, the two move in opposite directions. The firm works faster, the hours fall, and the price falls with them while the client receives the same output.

Why automation forces the unit question early

Automation moves the question forward. It absorbs work a consultant used to do by hand, so what the firm delivers comes unhooked from the hours it bills. A firm in that position has two paths: price the product on its own terms, or leave it inside the engagement fee and carry the cost. Most firms take the second path without noticing they took it, because nobody made the decision. The unit question arrives sooner than the firm planned, and it arrives with an installed base already expecting the tool for free.

The test to run before anything else: if the product’s price would fall when your consultants work faster, the unit is still the hour.

Across the services firms we have worked with, the shape repeats. The tool augments a service that is billed by the hour, so the better the tool works, the fewer hours the firm bills. The whole value proposition sits on the services side. The software has no carve-out and no way to be substantiated as a line item of its own, so the pitch slides back to doing the work better, faster and cheaper. Better, faster and cheaper is fewer hours, and fewer hours is less revenue, so the tool that made the firm more capable reads to its clients as a discount.

The compensation plan holds it there. In most services firms the partners who own the client relationships are paid on services revenue. The tool goes into the proposal as a way to win the engagement and comes out priced at zero. Hardware companies run the same pattern in reverse: the box carries the number and the services around it are given away to close the sale. In both cases whatever nobody is paid to sell never earns a value proposition of its own, and for a services firm that is the software.

Pricing software you used to sell as a service starts with the unit

Licensing comes first because it settles the value metric: the unit the software is sold by, and the unit every packaging and pricing decision downstream inherits. The licensing model is two decisions, the unit and the grant. For a productized method the candidate units are concrete: a scenario run, a case processed, a client account, a report produced, a seat for the analyst who runs it. Which one fits is read from the firm’s own deal record and differs for every firm; the frame that produces the short list is the same everywhere.

What the client counts when it reports the work happened

Ask what the client counts when it tells its own board the work happened. A client that ran the model to plan headcount counts scenarios, or the sites it planned; a client that used the tool to clear a backlog counts cases. That count sits on the client’s side of the boundary, it grows when the client receives more value, and the client can estimate it before signing. A unit with those properties is a candidate value metric. The hour has none of them.

Professional services pricing under AI argues where the injection point sits for the services themselves, at the boundary where the firm’s work product crosses into the client’s operations. Here the method has already become software, and the unit question is about the software.

When is the hour still hiding inside the unit?

A fixed fee reverse-engineered from partner hour estimates is still hour-based pricing with the arithmetic hidden. So is a subscription set by dividing last year’s engagement fee by twelve. The tell is in the renewal conversation. If the client asks how many hours the team spent, or why a smaller scope did not produce a smaller fee, the price is still being read as time. The unit has to be one the client can reason about without a timesheet.

The five shapes the transition takes

Across the services firms we work with, the transition arrives in five recognizable shapes. Most firms carry more than one at once. Each raises a question the rate card cannot answer.

Shape 1: the fixed-fee tool

The firm built a model to solve one client’s problem, it lived in a spreadsheet for years, and now the firm charges a flat fee for it. The fee was set to feel fair in the room beside the engagement that produced it. That is a reasonable way to price the first sale, and it carries no information about the tenth. The question a fixed fee raises is what it leaves uncaptured as the client’s use grows, and whether the fee scales at all.

Shape 2: the tool you gave away

The tool sat inside the service fee, or it was free because free won the work. A tool that went out at zero has no price on record, so there is no position to defend when the next client asks for one. The question is which tools belong on a price list at all, and how to introduce a price without souring a relationship built on free. Read the buyer’s side first: from the client’s chair, a tool with no product team behind it is custom code they now maintain.

Shape 3: the bespoke build that keeps repeating

Every build started as a one-off. Then three of them turned out to be the same build under different client names. The repeated parts are a product hiding inside engagement work. The parts that were different stay custom and stay priced as engagement work. Deciding which is which is a packaging decision, and it usually arrives before anyone in the firm has called it one.

Shape 4: the advisory gate

Advisory work drifts toward scope and outcomes while the software wants a recurring price on a unit that grows. Where the gate sits between the two decides how much of each revenue line survives the other. Most firms discover they set it by accident, on the day the first client received the tool inside an engagement.

Shape 5: the recurring layer nobody priced

Hosting, maintenance, support, the release shipped to keep the tool running. That is recurring revenue the firm has never carried, and it tends to disappear into a services line until someone goes looking for the margin. The question is whether it is bundled into the product’s price or sold beside it, and on what unit.

We see the hidden version of this in the spreadsheets services firms quote from. Hosting and support are costed into the professional-services build-up and blended into the hourly rate, so the recurring layer never appears as a line of its own. It works against the firm twice. The client sees an hourly rate that reads higher than the competitor’s, because it carries a product cost the competitor’s rate does not. And the firm never learns what the recurring layer earns, because nothing in the quote separates it from the hours.

Where the gate between advisory and product sits

The gate is a packaging decision. What ships inside the software, what stays an engagement, and what the team still builds bespoke for a client who asks. Editions are the instrument, and which packaging shape fits a given firm is a packaging question. What no framework settles for you is where your own gate sits, because that depends on which revenue line you are protecting and whether it was ever at risk.

The precedent the first free client sets

An unstated gate does not stay unset. It is set by the first client who received the tool inside an engagement, and every later client inherits that precedent. The tool is part of the service, and a price for it reads as a fee increase. A gate decided deliberately can be explained to a client, while one set by precedent can only be renegotiated, one relationship at a time.

Does the product cannibalize the practice?

Booz Allen Hamilton faced both sides of this. Software the firm built out of its own consulting expertise was going to displace work it was being paid services fees to manage. The product had to earn back more than it took away. And a capability everyone had filed as an optional add-on turned out to be central enough to the platform’s value that it belonged in the base package. Both were packaging decisions, made before a price was attached to anything.

The cannibalization question has a test. Name the services revenue the product would displace, then ask whether that revenue was defensible on its own. Where it was leaving anyway, the product is replacing revenue the firm was about to lose. Where it was defensible, the gate has to be drawn so the product and the practice do not compete for the same purchase. If your firm cannot answer the second question from its own deal record, talk to an expert before the next client asks the price.

Moving the installed base onto product pricing

The clients who already have the tool are where the transition is won or lost. They received it inside an engagement, at a price of zero or a fee that felt fair at the time. They have every reason to read a new product price as a fee increase for the same relationship.

Why a cutover date is a churn plan

A blanket cutover treats every account the same, and no two accounts have the same history with the tool. The pattern we see hold is a transition modeled one account at a time: who pays more, who pays less, and what each receives that it did not have before. The general mechanics of moving existing customers to new pricing apply; what differs here is that the product arrived inside a relationship that already had a fee. Some accounts move at the next renewal. Others move across two cycles. The plan reflects each relationship rather than a date on a calendar.

What the price attaches to for an existing client

The migration is usually solved by changing what the client receives, so the price attaches to something new. A version the client did not have, a capability that was never in the engagement, support the firm never formally offered. The client pays for a capability it can name. Where the price attaches to nothing new, the conversation reverts to the engagement fee and the firm ends up defending a timesheet.

One firm we worked with began as a managed service wrapped around its own product, and its value proposition was the service. As the product matured on direct customer feedback, some customers started using it directly and terminated the managed-service portion of the contract. They moved one account at a time, each because the product now carried something the service had never delivered. The firm today sells a managed service and direct access to the product, the way any software company would. Services firms can become extraordinary software companies for that reason. They build products that already work, and they understand the role of services and the profitability of running them, which a software company has to learn.

Why productized pricing raises a services firm’s enterprise value

Many services firms in this position are pointed at a sale: a founder exit, a roll-up, a strategic buyer who wants the product more than the practice. A buyer pays for a business it can model forward. Revenue that renews by default, margin exposure that is bounded in the licensing model, and a pricebook that explains the prices the firm realized are what a multiple capitalizes. A firm that arrives at diligence with the architecture living in a founder’s head spends the window defending it.

That is the reason to do the work before the tool has a hundred clients on a hundred different arrangements. Instrumented, repeatable pricing transfers with the asset, while arrangements that live in the partner who made them have to be explained account by account in the data room.

Where a services firm starts

The sized first step is small. Run the free Pricing Architecture Assessment on the product you are building; the services pricing is a separate question. Then bring the result, the deal record, and the partner who owns the client relationships to a conversation. We work licensing, packaging, and pricing for that one product with your partners and your product lead in the same room, and your team operates the result on LevelSetter afterward. Talk to an expert and describe the tool, who has it today, and what they paid.

FAQs