Cheap team vs senior team

Salesforce Consulting Agency Costs Calculator

Hourly rate is not the cost. A "cost-effective" team often looks cheaper on the proposal - then shifts planning, review, rework, coordination, and risk onto your internal people. That second bill shows up later, and it is usually bigger.

AI makes the traditional split ("seniors for hard work, juniors for easy Setup/Flow tasks") outdated: execution is increasingly cheap compute, while judgment, architecture, and review stay expensive.

Below is an interactive model of two staffing shapes with the same knobs - only the defaults differ. Agency invoice and true cost (including internal shadow QA) are both shown. Change any assumption; the results update live.

We discuss the whole topic - including the risks and hidden costs - in our post The Real Cost of Salesforce Consulting Services.

INTERACTIVE MODEL

Cost & time calculator

Two scenarios, the same set of knobs - only the defaults differ. Work is measured in abstract delivery points. Both teams get AI leverage; supervision, growing rework, and internal shadow-QA are priced into the true cost.

With these assumptions, Scenario B (senior-led) finishes about 7.3 months earlier (2.1× sooner), and its true cost is 38% of Scenario A (a saving of $1,094,584, 62%).

SCENARIO AAI 1.75×/1.25×
Cost-effective

2 sr @$110 · 8 jr @$30 · 2 PM

Time to complete
13.9months
True total cost
$1,757,742

agency $1,464,785 + 20% internal shadow

$127k/mo · 2,776 pts/mo · supervise 30%
SCENARIO BAI 1.75×/1.25×
Senior-led

4 sr @$150 · 0 jr @$30 · 0 PM

Time to complete
6.6months
True total cost
$663,158

agency $631,579 + 5% internal shadow

$101k/mo · 3,360 pts/mo · supervise 0%
B true cost vs A
38% of A

saves 62% · $1,094,584

And finishes
2.1× sooner

7.3 months earlier

Progress to done% of scope delivered · months →
014 months
Scenario B · senior-led
Scenario A · cost-effective
THE MODEL

How the calculation works

Scope becomes months; months become invoice cost; invoice plus your internal load becomes true cost. Six stages, and what each one is accounting for - the worked numbers for both scenarios follow below.

1
Effective scopewhat actually has to be built
project size + discovery gaps

The work is measured in abstract delivery points, so two differently-shaped teams can be compared on the same outcome. On top of the agreed size, a share of extra scope is added for what the original estimate missed - unasked questions, thin requirements, edge cases found during UAT. A team that consults up front discovers less of this later, so its gap allowance is smaller.

2
Gross pts / mohow much the team can produce
team × hours × productivity × AI - supervision

Each role converts billable hours into points at its own rate. Seniors produce more per hour on judgment-heavy work, but lose a slice of their delivery time to supervising - reviewing, unblocking, explaining - and that slice grows with the number of juniors per lead, while you still pay their full rate. AI multiplies output for both teams. Project managers cost money and produce no points.

3
Net pts / mohow much of it is new progress
gross capacity - rework tax

Not all capacity moves the project forward. A share is spent redoing work: fixing defects, reopening closed tickets, chasing regressions. On weak foundations that share compounds month over month as technical debt accumulates, so late months deliver less real progress than early ones - while the invoice stays the same size. A capped ceiling keeps the model from reaching zero output.

4
Durationhow long it takes
net progress accumulated until scope is covered

Net points are added month by month until they cover the effective scope. Because the rework tax can grow, this is not simple division - a team losing capacity over time stretches out non-linearly, and each additional month carries a full month of burn. This is where a capacity gap turns into a schedule gap.

5
Agency $what the vendor invoices
monthly burn × duration

Monthly burn is every person on the contract at their rate, whether or not they produce points. Multiplied by the duration above, this is the number on the proposal - and the only one most comparisons look at. A low rate card attached to a long timeline can easily cost more than a high rate card attached to a short one.

6
True $what your company really pays
agency invoice + internal shadow load

Work that lands on your own people never appears on the vendor invoice: re-testing delivered features, re-reviewing pull requests, unblocking pipelines, standing in for consultants on stakeholder calls. Priced as a share of agency burn, this is the second bill - and it is the figure the comparison uses to decide which scenario actually wins.

Both scenarios use the same knobs - only the defaults differ. True cost includes internal dual-QA and coordination load.

THE MATH, IN PLAIN ENGLISH

How we got these numbers

The same six stages, worked out in full for each scenario. Every figure recomputes live from the sliders.

Scenario A · Cost-effective

1 · Effective scope
Start with 20,000 points and add 20% for discovery gaps → 24,000 points to actually deliver.
2 · Gross pts / mo
Seniors/leads: 2 × 160h × 3/h × (1 - 30%) × AI 1.75× = 1,176 pts
Juniors: 8 × 160h × 1/h × AI 1.25× = 1,600 pts
PMs: 2 people - cost money, add 0 points.
Gross capacity ≈ 2,776 pts/month.
3 · Net pts / mo
Month 1 rework 15% means only 85% of gross becomes net-new scope.
Each later month rework rises 4 pp until 50% - tech debt eats capacity while burn stays flat.
4 · Duration
Add net points month by month until we cover 24,000 → about 13.9 months.
5 · Agency $
Agency burn: (2×$110 + 8×$30 + 2×$100) × 160h = $105,600/mo
Agency total ≈ $105,600/mo × 13.9 = $1,464,785
6 · True $
Internal shadow 20% → True total ≈ $1,757,742.
Compounding rework, month by month
Month 1 - rework 15% - net 2,360 pts
Month 2 - rework 19% - net 2,249 pts
Month 3 - rework 23% - net 2,138 pts
Month 14 - rework 50% - net 1,388 pts

Scenario B · Senior-led

1 · Effective scope
Start with 20,000 points and add 5% for discovery gaps → 21,000 points to actually deliver.
2 · Gross pts / mo
Seniors/leads: 4 × 160h × 3/h × (1 - 0%) × AI 1.75× = 3,360 pts
Juniors: 0 × 160h × 1/h × AI 1.25× = 0 pts
PMs: 0 people - cost money, add 0 points.
Gross capacity ≈ 3,360 pts/month.
3 · Net pts / mo
Month 1 rework 5% means only 95% of gross becomes net-new scope.
Rework stays flat at 5% (controlled debt).
4 · Duration
Add net points month by month until we cover 21,000 → about 6.6 months.
5 · Agency $
Agency burn: (4×$150 + 0×$30 + 0×$100) × 160h = $96,000/mo
Agency total ≈ $96,000/mo × 6.6 = $631,579
6 · True $
Internal shadow 5% → True total ≈ $663,158.
On True cost, Scenario B is 38% of Scenario A (saves 62%). It finishes about 7.3 months earlier - 2.1× sooner. If that feels wrong, change the assumptions - the model is only as honest as your inputs.
WORKED EXAMPLE · ONE ARCHITECTURAL DECISION

"Just use record-triggered flows - it's faster and cheaper"

No questions asked about the post-MVP roadmap, other teams, or where the process is heading. Follow the cost curve of that single call over the life of the project.

01
Kickoff
DECISION

The agency recommends record-triggered flows for the custom sales logic. Faster and cheaper to configure, they say. No questions about the post-MVP phase or adjacent teams.

02
Initial build
MVP

Large record-triggered flows are created for Opportunities, Line Items, Quotes and Orders. Requirements are met 1:1.

03
Second product type
+SCOPE

A new product line needs different logic. Extra branches are bolted onto the existing flows.

04
New customer segment
+SCOPE

Another segment, another set of branches. The flows now take a minute to scroll through.

05
Ramp deals
+SCOPE

Ramp-deal support added. Every change now risks breaking an existing path or hitting SOQL / CPU limits.

06
Async workaround
PATCH

Async paths are introduced to dodge limits. Row-lock exceptions appear - and retries are not possible in flow.

07
Bugfix spiral
STALL

Bug fixes consume roughly 80% of development time. New features stall on the backlog.

08
The rewrite
VERDICT

Verdict: migrate everything to Apex triggers. Six months, five people, no new customisations until it is done.

The bill for one skipped question

A 6-month, 5-person rewrite - plus a feature freeze while it happens. None of it appeared on the original quote. A senior team would have asked about the roadmap on day one and picked an architecture that scales.

WHERE THE HIDDEN COST HIDES

Fourteen line items that never make the quote.

These are cost patterns you can observe on a program - not insults. Several are already priced as assumptions in the model above (rework, gaps, supervision, internal shadow). We discuss the whole topic, all of these risks, and the hidden costs in our post The Real Cost of Salesforce Consulting Services.

Architecture chosen for speed, paid for later

Early choices that skip future-state questions often force multi-month rewrites once scope grows.

Ticket-only delivery

Work matches the written requirement and the happy path - edge cases and design pushback appear only after UAT.

Review cost shifts to seniors

Unstructured or junior-heavy output needs line-by-line senior review, corrections, and re-review - hours that rarely show on the vendor rate card.

Compounding technical debt

Velocity falls over time as fixes and discovery consume capacity that used to ship features - until remediation becomes its own project.

Single-maintainer components

Critical pieces that only one person can safely change create delay and rewrite risk when they are unavailable.

Missing business-technology challenge

Assumptions go unchallenged; wrong requirements get built "correctly," and rework shows up after go-live.

CI/CD support load on your team

Merge conflicts, pipeline issues, and process gaps pull internal seniors into unblocking - time outside the agency quote.

Heavier build than the platform needs

Custom automation where a formula, config, or known library would suffice - more build cost and more long-term maintenance.

Dual QA after trust drops

After repeated defects, internal teams re-test agency work end-to-end - you pay for delivery and for verification again.

Extra coordination layer

A team that cannot own stakeholder communication usually needs an additional PM - extra burn with little delivery output.

Defects shipped as "done"

Tickets close quickly, then reopen as bugs; later sprints spend capacity on fallout from earlier "completed" work.

Stakeholder-call risk

When consultants cannot represent the work on a business call, your leads become permanent proxies - or reputation takes the hit.

Idle time on blockers

Billable hours continue while access, answers, or environment setup are pending - and unblocked slices of work stay untouched.

AI output without ownership

Model-generated changes arrive without understanding; review comments bounce back to the model, so seniors still carry the real delivery risk.

Senior experts only

Need help with the Salesforce implementation?

We don't staff juniors on your project. If you want architecture, delivery, and review owned by true experts - we can help.

Book a 30-min Call

Or email us directly: contact@beyondthecloud.dev