· 5 min read · northitech
The pyramid is the problem, not the people
Most failed engagements were staffed by capable people. They failed because the person who understood the problem was not the person who built the solution.
A consultancy pyramid is a pricing model dressed as an org chart. One partner, two managers, eight consultants. The partner sells, the managers manage, the consultants build. The client meets the partner and gets the consultants. Everyone involved is competent; the structure still loses the plot somewhere between the proposal and the code.
Where the budget goes
The proposal is written by someone who has read the client’s situation closely and formed a view. That view lives in their head. It is transmitted to the delivery team as a slide deck and a kick-off call, which is to say it is not transmitted. The delivery team then re-discovers the problem, slowly, while being billed as if they already understood it.
Ask anyone who has sat inside a large programme which variable predicted the overrun. The answer tends to be the same: whether the person who scoped the work was on the team that built it. When they were, the work came in near plan. When they were not, the number on the proposal turned out to be a fiction everyone had agreed to believe.
Why firms keep the pyramid
Margin. A partner’s judgement billed at partner rates is expensive; a partner’s judgement sold at partner rates and delivered at consultant rates is a business model. The pyramid is not an accident of growth; it is the product.
What we do instead
There is no pyramid because there is no layer to put one in. The person who writes the proposal writes the code. The team is the two founders, fixed for the engagement, and the substitution clause is written into the statement of work: nobody joins or leaves without the client’s written consent.
This caps our size. We cannot take on a fifty-engineer programme and do not pretend otherwise; a global integrator is the right answer for that. We are the right answer for scoped work where the judgement matters more than the headcount, and where the client would rather pay for two people who understand the problem than fourteen who are learning it on their budget.