So-What Laddering

Thinking Tools2.1 Finding the Reasons to ResolveAlso known as So-What Test

From management consulting practice, as set out by Barbara Minto (The Pyramid Principle, 1978 edition) and Ethan M. Rasiel (The McKinsey Way, 1999).

Asks "so what?" of a finding, then of the answer, and again, climbing from observation to implication to an action the audience can take.

One finding climbed rung by rung to an action, and one that cannot climb

The director answers every finding with so what? At each of the analyst's turns, pick the answer that climbs a rung: from the fact, to the pattern, to what it means, to an action the director can take herself. The last finding cannot climb at all. Illustrative.

Analyst

Sales director

Analyst

Sales director

Select a turn to read it. At each analyst turn, pick the answer that climbs.The finding. An office-supplies distributor's analyst opens the quarterly review with the number.So what, 1. The director has heard the number. She wants to know why it matters.Rung 1. Pick the answer that adds something the number alone does not say.Pick a move to see the reply.The director nods and asks again.A restatement: the same fact, more dramatic, no higher up.The director leans in: which 30?Right: a pattern under the number, one rung up.The director asks what price has to do with it.A jump: an action with no reason under it yet.So what, 2. The pattern is clear. Is it a blip or a warning?Rung 2. Pick the implication for her business.Pick a move to see the reply.The director agrees, and asks again.Too general: true everywhere, so it says nothing about her region.The director asks what can be done before January.Right: an implication with a size and a date.So what, 3. The audience is the regional sales director. The action has to be hers to take.Rung 3. Pick the action within her own authority.Pick a move to see the reply.She points out the website belongs to another division.Outside her authority: a good idea addressed to the wrong person.She approves it and asks for the list by Friday.Right: the top rung, an action she can order today.She asks what that changes.Not an action: waiting a quarter loses the renewals.Another finding. The analyst has one more slide: the sales team's calls are slightly longer.So what, 4. Nobody can think of what the 12 seconds changes.Keep or drop. The finding is true. It leads to no implication for this audience.Pick a move to see the reply.The review ends ten minutes early.Right: a finding that cannot climb past the first rung only takes time.Two people ask about call scripts, and the meeting runs over.Kept: it takes attention from the finding that matters.

Right calls: 0 of 4Right calls: 1 of 4Right calls: 2 of 4Right calls: 3 of 4Right calls: 4 of 4

  1. The finding (Analyst). An office-supplies distributor's analyst opens the quarterly review with the number.
  2. So what, 1 (Sales director). The director has heard the number. She wants to know why it matters.
  3. Rung 1 (Analyst). Pick the answer that adds something the number alone does not say.
    • It is the worst third quarter in six years The director nods and asks again. A restatement: the same fact, more dramatic, no higher up.
    • Almost all of it is 30 accounts that moved to ordering online once a month The director leans in: which 30? Right: a pattern under the number, one rung up.
    • We should cut prices to win the volume back The director asks what price has to do with it. A jump: an action with no reason under it yet.
  4. So what, 2 (Sales director). The pattern is clear. Is it a blip or a warning?
  5. Rung 2 (Analyst). Pick the implication for her business.
    • Online ordering is a big trend across the industry The director agrees, and asks again. Too general: true everywhere, so it says nothing about her region.
    • 200 more accounts fit the same profile, and their contracts renew in the first quarter The director asks what can be done before January. Right: an implication with a size and a date.
  6. So what, 3 (Sales director). The audience is the regional sales director. The action has to be hers to take.
  7. Rung 3 (Analyst). Pick the action within her own authority.
    • Rebuild the company website so ordering is easier She points out the website belongs to another division. Outside her authority: a good idea addressed to the wrong person.
    • Put a rep on each of the 200 before renewal, with an online reorder list the rep sets up She approves it and asks for the list by Friday. Right: the top rung, an action she can order today.
    • Monitor the trend next quarter She asks what that changes. Not an action: waiting a quarter loses the renewals.
  8. Another finding (Analyst). The analyst has one more slide: the sales team's calls are slightly longer.
  9. So what, 4 (Sales director). Nobody can think of what the 12 seconds changes.
  10. Keep or drop (Analyst). The finding is true. It leads to no implication for this audience.
    • Drop it from the deck The review ends ten minutes early. Right: a finding that cannot climb past the first rung only takes time.
    • Keep it, in case someone asks Two people ask about call scripts, and the meeting runs over. Kept: it takes attention from the finding that matters.

An example

An analyst at an office-supplies distributor built a twenty-slide quarterly review for the Midwest sales director. It opened on a fact: third-quarter sales fell 8 percent. Her manager read the draft and wrote one question beside every slide: so what?

She climbed. The 8 percent turned out to be almost entirely thirty accounts that had moved to ordering online once a month, where no rep saw them. That was a pattern. The next so what gave the implication: two hundred more accounts had the same profile, and their contracts renewed in the first quarter. The last so what had to end in something the director herself could order, so the website rebuild came off the list, since another division owned it. What remained was to assign a rep to each of the two hundred before renewal, with an online reorder list the rep would set up. Eleven slides failed the test at the first rung and were cut. The review ran twelve minutes and ended with a decision.

Why it works

Analysts know why their findings matter, so they stop at the finding and expect the audience to finish the thought. The audience rarely does. A true number with no implication gets a nod and no action, and the meeting moves on.

Asking so what of each answer forces the climb the analyst skipped: from fact to pattern, from pattern to consequence, from consequence to a move. Fixing the top rung to an action within the audience's own authority keeps the ladder from ending in advice addressed to someone else. The test also works as a filter. A finding that cannot climb even one rung is noise for this audience, however true, and cutting it gives the findings that do climb more room.

The thinking behind it

Barbara Minto Barbara Minto, the first woman MBA hired by McKinsey, taught its consultants to write with the main point first and the supporting points grouped beneath it, and published the method as The Pyramid Principle (revised second edition by 1978). The so-what test is the habit her structure demands: every point in a pyramid must say something the reader needs, and a point that only repeats data has no place at the top.

Ethan M. Rasiel Ethan M. Rasiel, a former McKinsey associate, described in The McKinsey Way (1999) how the firm's consultants structure problems: start from a hypothesis, break the problem into an issue tree, and test each branch with facts. His practitioner's account, with Minto's book, is the written source for the so-what test, which McKinsey and similar firms teach as part of the craft.

How to run it

  1. State the finding precisely. Write it with its number, place, and period, so the next rung has something exact to stand on.
  2. Name the audience and their authority. Decide who will hear it and what they can actually decide or order.
  3. Ask so what, and write the answer. The answer must add something the finding does not say, such as where the change comes from.
  4. Ask again of that answer. Climb from pattern to consequence, giving a size and a date where you can.
  5. Stop at an action the audience owns. The top rung is something they can do this week, not a recommendation for another department.
  6. Drop what cannot climb. Cut any finding that stalls at the first rung, however much work it took.
  7. Lead with the top rung. Open the presentation with the action and its reason, and keep the climb underneath as support.

Where it goes wrong

Each rung restates the last. An answer such as the worst quarter in six years sounds like progress and adds nothing. If the answer could be written without new information, it is the same rung.

The ladder jumps. An analyst goes from a number straight to an action, skipping the pattern and the implication, and the audience cannot see why the action follows. Rebuild the missing rungs or the action will be argued with.

The ladder ends at someone else's door. An action that belongs to another team or a higher level is a suggestion, and the audience can only pass it on. If the action is out of their reach, change the audience or find the part they own.

When to use something else

  • Pyramid Principle. When several findings that each climb need ordering into one argument with a single point at the top.
  • SCQA. When the job is framing the story of the finding for a reader, from situation to answer.
  • Five Whys. When the finding is a problem and the climb needed runs down to its cause, not up to an action.

Where it sits in SOLVE eX

Go to the source

  • The Pyramid Principle, Minto International, revised second edition; the first edition is undated in library catalogs, 1978
  • The McKinsey Way, 1999

Back to the catalog