Two consultancies can have the same workforce headcount this year and end up in very different places by 2028: one built its plan around changes it could see coming, the other filled gaps as clients reported them. The difference is not usually reported up as a hiring failure, but as a bid the firm didn’t win, a team assembled under time pressure, or a client relationship that cools after rushed delivery.

The Swedish resourcing platform Cinode describes the underlying problem: utilization tells a firm how many consultants are busy, not whether it can staff its next deal. Its research describes firms with people sitting on the bench at the same moment they were turning down winnable work, because nobody connected the two facts.

Competency data is drawn from three places that go stale, according to Cinode: CVs updated only at hiring, spreadsheets only one person maintains (until they don’t), and relevant experience nobody logs formally.

Sales pipeline sits in a CRM, delivery capacity in a resourcing tool or a spreadsheet, and skills data in an HR system, and the three are rarely correlated. That information usually exists somewhere in the firm; but it is not gathered and interpreted regularly.

This article covers four disciplines that bridge that gap: what specialism the pipeline needs and when, what level of seniority that work calls for, where the right people are, and which employment model can close each shortfall. A final section covers what an unresolved gap costs, which is a number most practices never track.

Forecasting a Demand Curve Running in Two Directions

All specialism forecasting is currently subject to a time-limited fact: the 2027 deadline has created a genuine, temporary spike in demand for migration-relevant skills. A firm that sets its permanent headcount purely around today’s scarcity is building capacity for a demand curve that will fall away once most ECC migrations are finished.

A less obvious problem can be found underneath that one. SAP’s own Business AI release notes through the first half of 2026 confirm Joule now handles custom-code migration, ABAP test generation, and a chunk of finance and supply-chain verification work: capability that, until recently, meant routine junior and mid-level delivery hours. That’s the demand-reducing half of the picture.

SAPinsider’s AI Adoption and Maturity research: 91 percent of organizations use AI at some level, but only 17 percent have it built into core workflows. That 74-point gap points to unfilled demand for people who can configure and govern the AI layer, not just switch it on.

The same specialism map is shrinking in one place and stretching thin in another, and that is a different forecasting problem from a market that’s simply getting hotter or cooler.

The forecasting method most practices default to misses this. Treating a deal as either signed or not, right up until it closes, tells a firm what it might sell, not whether it can deliver it. Weighting each pipeline opportunity by its probability of closing, then mapping that weighted figure against capacity by specialism, seniority, and location together, gets much closer to a usable forecast.

The method has limitations. It breaks down fast once probability scores go stale, or once sales and delivery use different definitions of “committed.” A rolling monthly forecast checked against actual capacity acknowledges that drift faster than an annual planning cycle.

Also, the market itself isn’t one uniform thing. DSAG’s 2026 investment report, drawn from its own membership, finds only 37 percent of customers targeting the 2027 deadline itself, with close to half pushing their migration to 2030. On-premise S/4HANA still leads investment intent at 42 percent, ahead of private cloud at 22 percent and public cloud at just 6 percent.

Acute scarcity in particular specialisms, and a market moving more cautiously, are both true at the same time. A plan that treats “the market” as a single signal misreads the situation.

Forecasting the Seniority Mix

Knowing which specialism to plan for answers half the question. The other half is what level of seniority that work needs, and this is where most workforce plans fall short.

Commentary on the SAP ecosystem through late 2025 and into 2026 describes AI-assisted delivery compressing the junior-heavy base that consulting pyramids have run on for decades: the research synthesis, first-pass configuration, and draft output that used to occupy several years of junior time.

The commentary describing it is consistent about the mechanism, and points to real, current pressure on graduate hiring and entry-level pay: but it isn’t backed by large-sample data yet.

A parallel paper, presented at a recent NBER labor-studies session, confirms serious research attention on the same question, though its specific findings couldn’t be independently verified. Put plainly: the direction is real and visible in hiring and pay signals but it isn’t fully measured yet, to the extent that it can provide useful guidance.

A firm forecasting total headcount without forecasting seniority mix runs into a predictable failure: it overbuilds junior capacity that AI increasingly absorbs, and underbuilds the mid-level and senior people who still have to review and sign off on the AI’s output.

The mismatch doesn’t usually become clear as a hiring shortfall; it can be seen as promotion-rate pressure, because the plan leaves fewer rungs of seniority than the work now needs. Setting a named target ratio for seniority mix, and checking it against actual promotion and attrition data as often as headcount targets get checked, catches a pyramid narrowing at the base while there’s still time to act on it.

Where the People Are, and What Language They Work In

The third input a workforce plan needs is location, and it gets treated more casually than it deserves. 

A specialism forecast that’s accurate for the business as a whole, but wrong at the country level, still fails: the right skill sitting in the wrong location isn’t deployable capacity for the client across the table. This is not a problem that can be resolved by simply moving staff to the right country (an extremely arduous process) or by allocating consultants to work remotely.  

Language is what a simple headcount-and-office-location model misses.

Most client-facing SAP roles across the German-speaking market still require fluency in German, even when English-speaking delivery roles exist in parallel on the same program. A firm planning coverage purely around office locations, or the cheapest delivery hub, overlooks a real constraint: who can actually be put in front of a given client.

The correct use of language is not simply a question of simple communication of the facts. It affects perception of professional competence, and the ability to form effective working relationships.

One large SAP staffing and delivery partner’s own account of resourcing 250 SAP professionals across Germany, Austria, and Switzerland points to a better planning unit: treat neighboring markets sharing a language and a regulatory culture as one footprint, not three separate hiring problems on a map.

That reframes coverage as a specific, checkable answer to who can walk into a given client’s building and work in the language that client operates in: a more relevant test than a current geographical location.

Why Freelance Capacity Isn’t as Flexible as It Looks

Closing a gap like this today comes down to four levers: hire permanently, build the skill internally, borrow it through a contractor, or let AI absorb part of the task. Permanent hiring is the default lever, but under current scarcity, it’s also the slowest and most expensive option for a spike with a known end date.

Contractor capacity moves faster, and these firms need it as a stability mechanism, not a stopgap reached for once a gap opens.

In the German-speaking market specifically, using that lever well means treating it as a compliance question before a cost question. Scheinselbständigkeit (Germany’s “false self-employment” status) is defined under Section 7(1) of Social Code Book IV, in which a contractor working like embedded staff can be reclassified as an employee, retroactive to their start date.

The hiring firm then owes employer and employee social-insurance contributions going back four years (thirty years if intent to evade is found) plus a 12 percent annual penalty, a risk that can run to roughly 40 percent of the contractor’s total pay.

A separate law, the Arbeitnehmerüberlassungsgesetz, requires a licensed arrangement wherever the engagement functions as disguised temporary staffing rather than genuine contracted delivery. A firm builds that structure before the gap appears, keeps contractor capacity workable rather than a liability.

Before reaching for either lever, a plan is worth checking against a question most skip: is this gap real market scarcity, or did the firm create it?

A 2021 study by Hoffman and Tadelis in the Journal of Political Economy found that moving an employee from one of the worst-rated people managers to one of the best cuts their attrition risk by roughly 60 percent. The effect concentrates on a firm’s best, most relied-on people, and has no link to the manager’s own technical or billing performance.

By that finding, a firm that promotes and staffs purely on delivery merit is likely to lose its best consultants to its own best billers, then pay a premium to hire or borrow back the capacity it drove out. That makes retention the cheapest of the four levers, and the one skipped most often.

What Firms Can Calculate

An unresolved gap in the first three disciplines tends to surface as one of three problems, according to research from the skills-analysis firm muchskills.

The first is practice thinning: a specialism with three experts today has one left in eighteen months if nobody develops into the role behind them. The other two are a single point of failure, where one person holds a skill the whole practice depends on, and bench misallocation, where available consultants don’t match the pipeline.

Each of these becomes problematic as a bid the firm declines or weakens: which is an opportunity cost nobody logs, because the firm simply doesn’t win the work and moves on. Any bid won anyway, using a stretched or mismatched team, turns into delivery risk instead, and usually a few weeks into the engagement, when fixing it costs the most.

The opposite problem, is too many consultants on the bench. Saibon’s own bench-cost framework puts the current industry-baseline utilization rate at 68.9 percent against a 75 percent target, and running that gap through a hundred-consultant firm billing €1,000 a day produces an annual revenue shortfall above €1.3 million.

Firms that let bench time run past three weeks see 40 percent higher turnover among the people sitting on it. No comparable formula turns up anywhere in the available research for what a lost bid costs.

A firm that starts logging bids declined or lost specifically to staffing inability, even informally, turns that invisible cost into a tracked one. That single change does more for a workforce plan than any of the four disciplines alone, because it finally makes the other three accountable to a number instead of a feeling.

None of these four disciplines pays off as a one-time exercise. A specialism forecast drawn up this quarter tells a firm less with every month that passes, and the same holds for a seniority ratio, a coverage map, or a contractor roster set once and left alone.

The clearest predictor of which of these firms keeps winning work through this cycle is whether workforce planning runs on a calendar or only shows up during a crisis. Run carefully, addressing all four counts together on a regular basis, it determines how much of the pipeline a firm can actually deliver, not simply how many people are on the payroll.

IgniteSAP works with SAP consultancies on exactly this kind of resourcing problem, helping practices build the permanent bench and contractor pipeline a plan like this depends on. If that support would help your own workforce planning, get in touch with us on LinkedIn.