Guide
Deal stages for software development agencies: a practical template
Most pipeline templates were built for SaaS. Software agency deals move differently. Here is a six-stage framework built around how custom development actually gets sold, with exit criteria and time-in-stage benchmarks.
Why default pipeline stages do not fit agency sales
The pipeline stage templates that come pre-loaded in most CRMs were built for product sales. Lead, qualified, demo, proposal, closed won, closed lost. Six stages with tidy names and the assumption that a prospect decides fairly quickly after a demo whether to buy.
Software agency deals do not work that way. They involve scope that is not fully defined at the start, evaluation processes that include technical due diligence, buying committees with different concerns at different levels, and a handoff to delivery that the client will be watching closely.
The result is either a pipeline that looks clean on a review slide but does not reflect real deal health, or reps who stop updating stages because the options available do not match what they actually know about the deal.
How software agency deals actually move
Custom software projects have a scope problem. The client often knows they need something but not exactly what. The first few conversations are about discovery as much as selling. The agency is learning the problem at the same time as they are being evaluated as a potential partner.
This is also why "qualified" means something different in agency sales. In SaaS, qualified usually means the prospect fits the customer profile and has budget. In agency sales, qualified often means the agency understands the scope well enough to write a credible proposal, and the prospect trusts the agency enough to take that proposal seriously. Getting there takes genuine investment from both sides.
Then there is the buying committee dimension. Larger engagements typically involve a technical evaluator, a budget owner, a project sponsor, and sometimes legal or procurement. Each of them can slow the process without the deal being dead. A deal in Evaluation for six weeks might be moving at procurement, stalled at technical review, or waiting on a board decision. The stage label does not tell you which.
A six-stage template for agency pipelines
Outreach
The prospect is aware of you and contact has been made, but no conversation about a specific engagement has happened yet. This stage is about getting a real meeting.
Exit: A discovery call scheduled.
Discovery
You are learning the client's problem. They are evaluating whether you understand it. No formal proposal has been discussed. Rushing this stage is one of the main reasons proposals miss the mark.
Exit: Enough scope clarity to write a credible proposal; a sense of timeline, budget range, and who is in the decision.
Proposal in progress
You are preparing the proposal or statement of work. The prospect is waiting. This stage should be short. A proposal that takes three weeks to produce from a good discovery call is often over-engineered.
Exit: Proposal sent.
Evaluation
The prospect has the proposal and is reviewing it, potentially with multiple stakeholders. This is where deals most often stall and where time-in-stage management matters most.
Exit: A verbal decision, a negotiation request, or a confirmed next step with a date.
Contract and close
The deal is agreed in principle. You are working through contract review, procurement, or legal sign-off. This stage often takes longer than expected, especially for larger engagements.
Exit: Signed agreement.
Handoff
The transition from sales to delivery. Introductions made, kickoff call scheduled, project structure confirmed. This stage is often skipped, which is a mistake. How handoffs are handled determines the first impression delivery gets to make.
Exit: Project formally started with delivery team.
What to track at each stage
Stage labels alone are not enough. A deal sitting in Evaluation for 14 days with a confirmed next call is very different from a deal in Evaluation for 14 days with no recent contact. The label is the same. The situation is not.
- Next action and date. What is the specific next thing that needs to happen, and when? This is more useful than stage alone because it captures forward motion rather than just current position. A deal with no next action and no expected close date should trigger a review.
- Decision-maker clarity. Who is actually making this decision? Who else is involved? Understanding the buying committee structure early reduces the chance of a late-stage surprise where someone you have not met has veto power.
- Last meaningful interaction. When did you last have a real conversation with someone at this prospect? Not an email, a real conversation. This separates deals that are actively moving from ones that just look like they are.
- Estimated close date. Not always accurate, but useful for spotting when a deal has missed its own timeline. A deal with a close date two months in the past that has not moved stages is a problem worth naming.
Time-in-stage benchmarks
These vary by project size and industry, but they give you a reference for spotting when something is moving too slowly for a typical mid-size software agency engagement.
When a deal has been in Evaluation for six weeks with no documented next step and no recent contact, that warrants a direct call. Not a follow-up email.
On the relational weight of proposal stage
Mariana spent years working with B2B service agencies across Portugal and Brazil on sales operations before moving into product. One thing she observes consistently across Iberian and Latin American markets: the stage of a negotiation is not just an administrative label. It carries relational weight.
"In Portuguese business culture, sending a proposal is not just a step in a process," she says. "It is an implicit signal that you take this relationship seriously enough to invest real time in. Agencies that rush the proposal just to move the stage forward often miss what the client is actually looking for at that moment, which is evidence that the agency is genuinely invested."
She also notes the inverse: a proposal that arrives late, with errors, or that clearly does not reflect what was discussed in discovery, signals something about how the agency will run the engagement. Clients read proposals the same way they read a first meeting. The content matters, but so does the care.
The mechanics of deal stages matter. But the stage of a relationship is not the same thing as the deal stage in your CRM. Both deserve attention.
Common mistakes with agency deal stage setup
- Too many stages. Eight stages with granular sub-statuses sounds thorough. In practice it means reps don't know where to put a deal and start skipping updates. Five to seven stages is usually the right range for a team without a dedicated sales ops function.
- No exit criteria. A stage without a clear definition of what moves a deal forward becomes a holding area. If a deal can stay in Evaluation indefinitely without triggering a review, it will.
- Conflating stage with probability. Win probability is a separate input from stage. A deal in Evaluation with a verbal yes has a completely different probability from a deal in Evaluation with one conversation and a downloaded proposal. Track them separately.
- Skipping the handoff stage. Closed won is not the end of the deal. It is the beginning of a different kind of work. Treating it as the pipeline endpoint leaves out the transition that determines whether the client will come back and refer you.
How Lumenbase handles deal stages
Lumenbase lets you configure deal stages with labels and exit criteria that fit how your agency actually sells rather than forcing a generic template.
- Custom deal stages. Configure stages with labels and exit criteria that fit how your agency actually sells, rather than a generic template that was built for a different kind of sales motion.
- Stage velocity tracking. Shows how long deals have been at each stage. When something looks stuck, it surfaces before it becomes a problem, not after the prospect has already made a decision.
- The Feed. Flags deals with no recent activity regardless of their stage label, catching the specific failure mode where a deal is parked at the right stage but nothing is actually happening.
- Next action fields. Track the concrete next step for each deal. Pipeline reviews stay actionable instead of turning into a conversation about stage label definitions.
- Lumo. Drafts follow-up messages and pre-call briefing notes from the deal's captured context, including what was discussed in discovery, what the proposal covered, and how long the deal has been at the current stage.
Who this is for
Software development agencies and custom software consultancies who have either outgrown a spreadsheet or found that their current CRM's default pipeline does not match how they actually sell. This framework applies to agencies of 5 to 200 people with deal cycles of one to twelve months. It does not require a dedicated sales ops function to implement, though it helps to have someone who owns the pipeline review cadence.