Forge ROI for Saudi SMEs

A Saudi SME can publish more content and still make a poor investment. The extra material may require more review, attract enquiries the business cannot serve or consume time that nobody has budgeted. A useful Forge ROI model therefore starts with the work and commercial outcome you want to change, then separates cash spending, staff capacity and additional contribution from sales.

This guide builds a hypothetical model for a small commercial plant-care provider. It is an arithmetic exercise, not a Forge customer case, productivity benchmark or forecast of Saudi business performance. Replace every assumption with your own records before using the model to make a spending decision.

Choose the comparison before calculating a return

Write down what would happen without the proposed change. The baseline might be an existing employee-led publishing process, a contracted service or no regular publishing programme. Those are different comparisons. A platform can reduce the effort required for an existing workload while adding cash expenditure; a new programme can add both work and spending while pursuing a commercial opportunity.

Use the same period, accepted output and quality standard on both sides. Comparing an old process that included Arabic review, evidence checking and reporting with a new process that only counts draft generation understates the new cost. Include rejected drafts, factual corrections and publication checks.

For the plant-care example, assume the team wants to maintain the same agreed publishing workload. It will assess whether Forge changes the effort needed to complete it. A separate question is whether the resulting content helps create additional profitable orders. Keeping these questions separate makes the conclusion easier to defend.

Establish the platform usage assumption

The public Forge offer, checked on 21 October 2026, lists USD1 per credit, 5 free credits for a new account, no monthly platform subscription and no expiry for unused credits. The pricing and plans guide explains the commercial options and how to count actions. Optional managed services require their own budget and are excluded from this example.

The owner's operating guide gives reference costs of 1 credit for a Manual publication, 1 per Auto-generated post, 1 per scheduled Batch post and 2 for a Content Studio article. Check current Billing before committing. Auto generation uses credits even if a reviewer does not approve the draft; approval is not the charging event.

For the illustrative month, budget 24 Auto generations and 4 article generations. The calculation is 24 × 1 + 4 × 2 = 32 credits, valued at USD32 at the stated rate. This is an action budget, not a promise of 28 approved publications. Any additional generation or other billable action must be added to the actual record.

The model excludes the initial 5 free credits so the recurring estimate is not temporarily improved by a starting allowance. It also makes no currency-conversion or tax assumption. A Saudi business budgeting in SAR should use the conversion and applicable charges shown in its actual purchasing records, then express every model input in that same currency.

Worked example: cash spending and economic cost differ

Assume the provider currently spends 12 internal hours per month on the agreed content work. The proposed process is estimated to need 6 hours, including source maintenance, editing, language review, publication checks and measurement. Value internal time at a hypothetical loaded rate of USD25 per hour. This is a planning assumption, not a Saudi salary benchmark.

Both routes also use USD50 of external design work. Assume the employee's pay does not change, there is no overtime reduction and the design cost remains payable. The table separates the value of internal effort from the cash expenses that actually change.

Hypothetical monthly comparison for the same accepted workload

Cost or effort	Existing process	Proposed Forge process
Internal hours	12	6
Internal effort valued at USD25/hour	USD300	USD150
External design spending	USD50	USD50
Forge usage	USD0	USD32
Total economic cost	USD350	USD232
Cash expenses shown, excluding unchanged payroll	USD50	USD82

The proposed process releases 6 hours, valued at 6 × USD25 = USD150. After the additional USD32 usage cost, the model shows an economic improvement of USD118 per month. Cash spending, however, rises by USD32. The business has not saved USD118 in cash merely because the employee finishes the work sooner.

Record what happens to the released time. It might help the employee answer enquiries, prepare proposals or complete overdue operational work. That can be valuable without becoming a payroll saving. If paid overtime or an avoidable external fee really falls, record the actual avoided amount and the conditions under which it can be avoided.

Include setup and test the fragile assumption

Assume initial setup takes another 6 internal hours: collecting approved knowledge, resolving offer boundaries, connecting the appropriate channel and preparing the first review process. At the same hypothetical rate, that is USD150 of setup effort. Allocating it across a 3-month assessment period adds USD50 to each month's economic cost.

The proposed monthly cost during that assessment is therefore USD232 + USD50 = USD282, compared with the USD350 baseline. The modelled economic improvement becomes USD68 per month. Allocation is a management comparison; it does not turn internal setup time into a cash invoice.

The most uncertain input is often review effort. A draft that takes longer to correct can erase much of the expected capacity benefit. Keep the same USD32 usage cost and USD50 monthly setup allocation while testing different amounts of time released.

Sensitivity to actual monthly time released

Time released	Value at USD25/hour	Net economic improvement before setup	Net improvement during the 3-month setup allocation
2 hours	USD50	USD18	−USD32
4 hours	USD100	USD68	USD18
6 hours	USD150	USD118	USD68

With these assumptions, the ongoing capacity model needs more than 32 ÷ 25 = 1.28 hours released per month to show a positive economic difference. During the setup allocation, the threshold is (32 + 50) ÷ 25 = 3.28 hours. If review time rises, calculate the loss instead of assuming that generation speed guarantees a saving.

Define a commercial outcome that the team can recognise

For the plant-care provider, an impression is not a suitable enquiry, and a suitable enquiry is not a contracted service. Define each stage before the trial. A suitable enquiry could require an office within the service area, a relevant plant-care need and enough information for the team to assess the scope. An order should follow the business's agreed commercial definition.

Keep a simple record of enquiries, qualification decisions, proposals and orders in the system the business already uses. Preserve the source reference when available and record missing information honestly. The Forge reporting guide explains the available publishing metrics; those metrics do not by themselves provide a native sales ledger or revenue attribution system.

Google's attribution guidance describes how credit is assigned across touchpoints. Credit assignment does not establish how many orders would have happened anyway. Its separate Conversion Lift guidance illustrates the use of treatment and control comparisons to estimate additional conversions. This is a distinction in measurement, not a claim that Forge runs those experiments.

A small before-and-after trial can reveal practical issues, but seasonal demand, changes in service availability and sales follow-up can affect the result. If causal evidence is unavailable, describe orders as observed or associated with the content and keep the uncertainty visible. Do not put all of their value into a claimed incremental return.

Use contribution and state the break-even basis

Suppose an illustrative order brings USD500 of revenue and requires USD300 of direct costs to deliver. Its contribution is USD200 before the marketing costs being evaluated. Use a realistic contribution estimate that includes relevant fulfilment, transaction and refund costs. Revenue of USD500 is not USD500 available to pay for marketing.

There are two legitimate but different break-even questions. To cover the additional USD32 cash spending in the earlier comparison requires 32 ÷ 200 = 0.16 additional orders, so 1 whole additional order would cover that amount. To fund the entire USD232 monthly economic workflow cost from additional contribution requires 232 ÷ 200 = 1.16 orders, so 2 whole additional orders would cover it.

During the setup allocation, the full-cost calculation is 282 ÷ 200 = 1.41 orders, still requiring 2 whole additional orders. These are conditional arithmetic results. They neither predict that an order will occur nor prove that a publication caused one. The first threshold evaluates extra cash spending; the second asks the content programme to support its full modelled cost.

For a conventional incremental ROI percentage, use a positive incremental investment denominator: additional verified financial benefit minus additional investment, divided by that additional investment. State the baseline, period and included costs. If the change reduces total economic cost, report the net difference clearly rather than forcing it into a misleading percentage.

Do not count the same capacity twice. If the released hours are used to deliver the additional orders and those delivery costs are already reflected in contribution, review how their value enters the model. Adding the full capacity value and the full sales benefit without checking the overlap can exaggerate the result.

Run a Saudi operating trial with explicit stop rules

Use the languages and commercial conditions your actual customers need. Budget Arabic and English review separately when the workload requires both; check that service coverage, enquiry instructions and scope retain the same meaning. Estimate the time for the real reviewer, including the founder if that person currently approves everything.

Choose a representative workload and record work by activity for both the baseline and trial. Keep standards consistent, note unusual rework and reconcile actual credit use. The platform evaluation guide helps define acceptance criteria; the message-testing guide addresses the separate question of what a fair comparison can tell you about the message.

Agree a decision before seeing the results. Continue if accepted quality holds, the effort fits the available capacity and the economic case meets your chosen threshold. Revise the inputs or scope if factual corrections dominate. Pause an uneconomic publishing rhythm instead of increasing volume to conceal the problem. Use the Saudi automation cost guide when additional tools or services change the budget.

Request a Forge demo with your current workload and cost assumptions. Bring one real offer, the people who review it, a baseline time record and a clear definition of a valuable enquiry. The Forge platform overview explains the product workflow; the investment decision should rest on the difference you can actually measure.