# Trade-offs, the iron triangle and saying no

> Which lever of the iron triangle in project management to pull when work runs late, why adding people slows a project (Brooks's law), how to say no to stakeholders, from a founder to a colleague and a client, without losing them, and version 4 of CatChow's roadmap.

- Course: [Outcome-Based Product Roadmaps](https://productovi.com/courses/outcome-based-roadmaps.md)
- Module: Prioritizing with evidence
- Lesson 12/15, ~49 min
- Source: https://productovi.com/courses/outcome-based-roadmaps/trade-offs-and-saying-no

## The first monthly review

**Alex Tutor:** It's CatChow's first monthly roadmap review, and you're there for the product team with an update. Today is about trade-offs: what gives when work runs late, and how to say no.

**Sofia:** Lucy's migration test is done. It works, but moving everything takes until the end of February. Until then, the pilot stays at two clinics.

**Brandon:** That's a problem. PurrPoint's procurement manager loves the form. She wants all 14 clinics using it in January.

**Ian:** Then hire two contractors for the warehouse and finish in January. I'll approve the budget today.

**Sofia:** Everyone wants something sooner, and nothing moves without moving something else.

**Alex Tutor:** That's the **iron triangle** of project management, also called the triple constraint: scope, time and resources, with quality caught in the middle. The sides are linked: squeeze one, say with an earlier date, and another has to give.

![A triangle frame of three rods holds a soft ball in the middle; a hand pushes one corner inward and the ball bulges out of shape.](https://productovi.com/courses/outcome-based-roadmaps/inline/trade-offs-and-saying-no-illustration-2.webp)

*Squeeze one side, and something else has to give.*

**Alex Tutor:** And if nothing visibly gives, quality pays quietly: rushed work, lost orders, bugs you discover in March.

**Sofia:** So which lever is which here?

> **Exercise:** Match each side of the **iron triangle** to what pulling it would mean for CatChow's warehouse and the clinic pilot.

## The iron triangle: which lever to pull

**Sofia:** Quality is out: lost clinic orders are exactly the stock-outs we're trying to get rid of.

**Alex Tutor:** Right. And time alone doesn't help Brandon. What about scope: does everything have to move at once?

**Lucy:** Not really. We could move clinic orders first. They'd run on the new system by mid-January; subscriber orders would follow, and the whole move would finish at the end of March.

**Sofia:** A month later overall, but the clinics get what they need about six weeks sooner.

**Alex Tutor:** A trade-off worth making, because the clinic theme is what's waiting. Pick the lever by what matters most: here, a date for clinics, not a finish date for the whole move.

**Sofia:** And the pilot grows only once clinic orders are on the new system.

![Moving clinic orders first puts them on the new system by mid-January so all 14 clinics can start in February; the whole move ends in March, a month later than all at once.](https://productovi.com/courses/outcome-based-roadmaps/diagrams/m4-l3-staged-migration.png)

*Staging scope protects the date that matters.*

> **Exercise:** Which lever should Sofia pull for the warehouse and the pilot, and why?

## Brooks's law: why more people make it later

**Sofia:** Ian's offer is still on the table, though. Two contractors can't hurt, can they?

**Lucy:** They would. Nobody ever documented the old system. My two people would spend January teaching instead of moving data.

**Alex Tutor:** It's an old observation, often called **Brooks's law**: adding people to a late software project makes it later. Newcomers need training, and every extra person adds coordination.

![Two newcomers arrive at a busy workbench; the experienced worker puts down her tools to explain the machinery while the half-finished work waits.](https://productovi.com/courses/outcome-based-roadmaps/inline/trade-offs-and-saying-no-illustration-3.webp)

**Sofia:** So resources is the lever that looks easiest and works worst.

**Alex Tutor:** For this kind of work, this late, yes. Extra people help when they join early, or on work that splits cleanly into separate parts.

> **Exercise:** Adding two experienced contractors to the warehouse replacement now would finish it sooner.

**Alex Tutor:** And the flip side: when do extra people actually help?

> **Exercise:** When is adding people to a project most likely to make it finish **sooner**?

## How to say no to stakeholders without losing them

**Brandon:** One more. A partner clinic director wants a stock tracker for every brand on her shelves, not just ours. Can we build that?

**Sofia:** A tracker for every brand is a different product. Only the CatChow part serves our key results.

**Alex Tutor:** So the answer is no to the tracker, but not to her. What's the problem underneath?

**Sofia:** Knowing what's running low before the shelf is empty. That's one of our clinic subthemes.

**Alex Tutor:** Then a good no has four parts: restate the need so she knows she was heard; say plainly what won't happen and why; offer what you can do for that need; say when things might change.

![A person kindly declines a huge box a visitor holds out and hands her a smaller gift instead; both are smiling.](https://productovi.com/courses/outcome-based-roadmaps/inline/trade-offs-and-saying-no-illustration-4.webp)

*Restate the need, say no plainly, offer what you can, and say what would change it.*

**Sofia:** The same four parts work for the founders and for Brandon.

**Alex Tutor:** They do. Refuse the request, not the person, and people keep coming to you with the next one.

> **Exercise:** Put the four parts of a good “no” in order.

## No to January, yes to February

**Sofia:** Now I have to tell Brandon that January won't happen.

**Alex Tutor:** Start from what he needs, not from what you're refusing. What does he actually need?

**Sofia:** Something firm to tell PurrPoint, and not to look as if he over-promised again.

**Alex Tutor:** Then give him the reason in PurrPoint's terms: expanding now risks the very stock-outs they want gone. Then a real next step, with a condition.

**Sofia:** Clinic orders move by mid-January. If the December review holds up, all 14 clinics start in February.

**Alex Tutor:** That's a no he can sell: a reason, a date with a condition, and nothing he'll have to take back.

> **Exercise:** Which message to Brandon is the best way to say no to January?

**Alex Tutor:** Now spot the gap in a real draft. Brandon wrote this to the clinic director before asking for help.

> **Exercise:** Brandon's draft: “I hear that you want to know what's running low before the shelf is empty. We won't build a tracker for every brand: it's a different product from ours. We'll look at clinic requests again at our June review.” Which part of a good no is **missing**?

## Loyalty points, again

**Ian:** The founders saw the RICE scores. They think the model is set up against loyalty points. They want points in Next, at least.

**Sofia:** It isn't rigged: there are a few comments behind points and eight person-months of work. But they aren't a bad idea either.

**Alex Tutor:** Then say **not now**, and make it a real one: the reason, and the condition under which the idea comes back. A vague “later” is just a slow no.

**Alex Tutor:** Here are two ways to write it. Which one is a real “not now”?

> **Exercise:** Two answers to the founders about loyalty points. Which one is a real **“not now”**?

**Sofia:** The condition: if churn is still above 8% at the June review, once the Now themes have had time to work, points come back, starting with a small test.

**Alex Tutor:** Good: checkable, dated and tied to the objective. Add what would bring points back sooner: evidence, such as cancellation surveys where people name the price.

**Sofia:** So a “not now” list, where each item has its reason and its way back.

**Alex Tutor:** Exactly. It also shows the founders their idea wasn't lost, only queued behind evidence.

> **Exercise:** Which of these belong in a good entry on a **“not now” list**? Select all that apply.

## A message for the founders

**Sofia:** Ian wants something about loyalty points that he can forward to the founders.

**Alex Tutor:** Use the four parts, and keep the tone warm: it's their company, and the idea may yet come back.

**Sofia:** The need, the reason in plain numbers, what we're doing about churn now, and the June condition.

**Alex Tutor:** And no RICE jargon. The founders care about subscribers, not formulas.

> **Exercise:** Write the message Ian can forward to the founders (3–5 sentences): loyalty points won't be in Next for now. Say why, what CatChow is doing about the same goal, and the condition under which points come back. No RICE jargon.

## A no for a client

**Sofia:** And the clinic director's stock tracker? Brandon needs words he can use.

**Alex Tutor:** Write it the way Brandon would say it to her. Remember that she's a client: she can always buy from someone else.

**Sofia:** Couldn't Brandon just say it's on the roadmap for later? It sounds kinder.

**Alex Tutor:** Kinder today, worse in spring, when she asks where it is. A clear no with a real offer is the kind version.

**Sofia:** What can we offer her, though? Her clinic isn't in the pilot.

**Alex Tutor:** Once the form works for PurrPoint, it can reach other partner clinics, and knowing what's running low is part of the clinic theme. Offer her an early place in that.

> **Exercise:** Write what Brandon can send to the partner clinic director (3–4 sentences): CatChow won't build a stock tracker for every brand, but it takes her underlying need seriously. Don't promise a date.

## Roadmap, version 4

**Alex Tutor:** Time to put the module together. Version 4 adds what was missing: how the decisions were made. Criteria, scores, the order and what's not now.

**Sofia:** The reminder emails went live on the 19th. Too early to see churn move, but they're out.

**Alex Tutor:** Good. Their effect will show in the cancellation reasons first. Now, the capacity for next quarter.

**Sofia:** Lucy gave me twelve person-months from January to March, five of them for the warehouse.

**Dataset (CSV):** [CatChow candidate solutions, scored in November](https://productovi.com/datasets/outcome-based-roadmaps/trade-offs-and-saying-no/catchow-candidates.csv)

Nine candidates from CatChow's roadmap with the RICE inputs Sofia and Lucy agreed in mid-November. Reach is per quarter; effort is in person-months for everyone involved. Synthetic data made for this course.

- `id`: Candidate ID
- `candidate`: Candidate solution
- `theme`: The roadmap theme it serves (none = no theme)
- `objective`: 1 = grow subscription revenue by keeping subscribers longer; 2 = grow the vet clinic channel
- `reach_per_quarter`: How many customers it touches in a quarter
- `reach_unit`: Who is counted in reach
- `impact`: Effect on each customer towards the objective's key results: 3, 2, 1, 0.5 or 0.25
- `confidence`: How sure we are about reach and impact: 1.0, 0.8 or 0.5
- `effort_person_months`: Total work for everyone involved, in person-months

**Alex Tutor:** Then the order has to fit into seven. Check every “not now” for a reason and a way back, and date the disclaimer today.

> **Exercise:** Write version 4 of CatChow's roadmap: the ranking.

## What to remember

**Sofia:** Brandon has February with a reason, Lucy's team keeps its focus, and the founders know what would bring points back. Nobody got a flat no.

**Alex Tutor:** Every yes costs something. When the cost is visible, people can accept the no.

> **Key ideas**
>
> - The **iron triangle**: scope, time and resources, with quality in the middle. When work runs late, one side has to give; if none does, quality pays quietly.
> - Pull the lever that protects what matters most. Cutting or staging scope often saves the date that counts; adding people late usually makes software work later (**Brooks's law**).
> - Every yes is paid for somewhere. Make trade-offs visible: what moves, what waits, and why.
> - A good no restates the need, says plainly what won't happen and why, offers what you can do, and says when it could change. Refuse the request, not the person.
> - Keep a **“not now” list**: each item with its reason and a checkable condition for coming back. A vague “later” is a slow no.

**Alex Tutor:** That wraps up module 4. In module 5 the roadmap has to live: winning buy-in, presenting it to the board, sales, engineering and PurrPoint, and the rhythm and rules that keep it current.

- Previous lesson: [Scoring frameworks: impact effort matrix, RICE and more](https://productovi.com/courses/outcome-based-roadmaps/scoring-frameworks.md)
- Next lesson: [Alignment and buy-in with stakeholders](https://productovi.com/courses/outcome-based-roadmaps/alignment-and-buy-in.md)

---
Productovi Learning. Free, no sign-up needed to read.