Walk into a 40-person SaaS company and you will find instrumentation everywhere: every trial tracked, every funnel step measured, every churn signal scored, the whole revenue machine visible on a dashboard someone actually looks at.
Walk into a 40-person consulting firm — often billing similar revenue — and you will find proposals assembled by copy-pasting the last proposal, scoping knowledge living in two partners' heads, delivery status collected by asking people in meetings, and margin per engagement computed quarterly, if ever, in a spreadsheet with a name like final_v3_ACTUAL.
Same decade. Same size. Five years apart in tooling — and I would argue that is the polite estimate.
The gap is structural, not cultural
The lazy explanation is that consultants are technophobes. It is wrong; these are smart people who advise other companies on modernization. The real explanation is feedback loops.
SaaS got tooled up because recurring revenue is unforgiving. Ship a bad onboarding flow and churn punishes you within the quarter, visibly, in the one number the board reads. That pressure funded an entire industry of instrumentation — CRMs, product analytics, revenue ops — because operational blindness showed up in the metric that mattered.
Consulting's project revenue hides the same failures. An underbid engagement doesn't show up as churn; it shows up as a team quietly working weekends, a margin figure nobody computes until the fiscal year closes, a senior partner absorbing the pain and calling it client service. The loss is real — often 20–30% of an engagement's economics — but it never lands in a single number at a moment when anyone could act on it. No visible pain, no tooling budget. For twenty years, that equilibrium held.
Why the truce just ended
Being uniformly behind was fine when everyone was behind. Your competitors also scoped by intuition and tracked delivery by asking around, so nobody lost a deal to tooling. It was a stable, industry-wide truce.
AI broke it, in a way that is easy to misread. The misreading: "AI writes proposals now, we should get it to write ours." The reality: AI is a compounding machine for systematized knowledge — and it returns roughly nothing on knowledge that lives in people's heads.
A firm whose past scopes, estimates, actuals, and outcomes exist as structured artifacts can point AI at them and get something genuinely new: proposals drafted from what actually happened on similar engagements, estimates checked against the firm's real bias, scope boundaries stress-tested before signature. A firm whose institutional knowledge is two partners' memory can point AI at nothing. The prompt is empty.
This is the part the industry hasn't priced in. The five-year tooling gap used to be a shared handicap. It is now the difference between firms that can compound and firms that can't — and unlike the SaaS tooling wave, which consulting could safely ignore, this one moves deal economics: who scopes faster, who bids tighter, who walks into a pitch with evidence instead of adjectives.
What catching up actually looks like
Not a transformation program. The firms getting this right are doing something almost embarrassingly narrow: they pick the one pipeline every deal flows through — enquiry, discovery, scope, estimate, proposal — and make it produce structured artifacts instead of one-off documents.
That single move pays three times. The immediate return is speed: proposal cycles drop from weeks to days because assembly stops being archaeology. The second return is accuracy: when scopes and actuals are comparable across engagements, underbidding becomes a measurable bias instead of a recurring surprise. The third return is the compounding one: every engagement now leaves behind data the next engagement can use — which is precisely the asset AI multiplies.
The judgment stays human. What a client actually needs, what the risk is worth, when to walk away — no tool touches that, and firms selling judgment should be insulted by anyone claiming otherwise. But judgment operating on instrumented reality beats judgment operating on memory, every time, and SaaS proved it a decade ago on their own revenue machine.
Consulting firms tell clients that what you don't measure, you can't improve. The five-year gap is what taking your own advice looks like when you haven't. The good news buried in the hot take: the gap is five years wide but about one pipeline deep — and closing it is now the cheapest it has ever been.
Isn't consulting fundamentally different from SaaS — bespoke, relationship-driven, resistant to systematization?+
The client-facing judgment is bespoke. The machinery around it — how proposals get assembled, how scope gets tracked, how delivery knowledge accumulates — is repetitive and highly systematizable. Firms conflate the two and end up systematizing neither.
Why did SaaS get tooling first?+
Recurring revenue punished operational blindness immediately — churn shows up in next month's number. Consulting's project revenue hides the same losses inside individual engagements, so the pressure to instrument never compounded. Different feedback loops, not different intelligence.
Does being behind actually cost anything if everyone in the industry is behind?+
It did not, until AI moved the baseline. When every competitor runs on documents and memory, nobody loses deals to tooling. That truce is over: firms that systematized their delivery knowledge can now compound it with AI, and firms that didn't have nothing for AI to work with.
What should a firm systematize first?+
The proposal-to-scope pipeline. It touches every deal, its inputs and outputs are already documents, and errors there — underbidding, ambiguous boundaries — are the most expensive per incident. It is also where AI assistance produces visible returns in weeks rather than quarters.
Related service
Want us to deploy this workflow for you?