Builds the social media ROI model (lifetime value, acquisition cost by channel, break-even and a 12-month projection) or the board business case, in the client's currency (default UGX for Uganda).
Use When
- Finance wants the return per campaign or channel from attributable value and acquisition cost, with a 12-month projection.
- Do customers who came through TikTok, Facebook or Google stay and buy again for longer? Group them by channel and month joined: Week-4 retention, decay curves and lifetime value from GA4 Cohort Exploration.
- Directors or a sceptical board think social media is a waste of money and want the business case for investing: revenue lost to growing competitors, follower value, an A&U study, NPS and budget tiers.
- The team asks when content spend breaks even and whether customer value covers acquisition cost.
Do Not Use When
advertising-attribution-and-measurementfor credit models, allowable CPA and holdout or geo tests on ads.meta-budget-plannerfor dividing the budget across channels.marketing-mix-modellingfor a multi-year model of each channel's contribution to sales and cross-channel budget allocation (Meridian or Robyn).meta-reportingfor the periodic performance report.- Stop when attributable revenue or cost data is missing; state the assumption and never present projections as results.
Required Inputs
| Artefact | Source/provider | Required? | If absent |
|---|---|---|---|
| Average transaction value, transactions per year and relationship length (per segment where values differ) | Client CRM or finance; consultant intake | Yes | Use a conservative single-transaction value as TLV and state that it understates the return. |
| Monthly investment split: retainer, content production, paid ad spend, total | Client finance or agency invoices | Yes | Stop the COCA and break-even steps; request the figures. |
| New customers per month by digital channel, or enquiry volume and conversion rate | CRM records, enquiry log or GA4 | Yes | Use blended COCA, or COCA per enquiry times conversion rate, labelled an estimate; recommend enquiry source tracking. |
| Gross margin % | Client finance | Yes | Mark the break-even step not assessed; do not assume a margin. |
| Channel attribution data and chosen attribution model | Enquiry log, UTM reports, client | If channel COCA is wanted | Report blended COCA only and set up the three-step attribution method. |
| Client name, industry, country/city, goal and currency | Client brief | Yes | Default to Uganda / Kampala and UGX; ask for the goal before framing the retainer. |
The full intake list is in ROI model method § Required Input.
Workflow
- Collect the intake; if revenue, cost or customer data is missing, get a reasonable estimate and log it as an assumption. Stop and route to
advertising-attribution-and-measurement,meta-budget-plannerormeta-reportingwhen the request is attribution modelling, allocation or a periodic report. - Calculate TLV (Bodnar and Cohen, 2012) step by step with the client's figures, per segment where high- and low-value customers differ; use per-channel cohort CLV from retention-cohorts-and-ltv when channels retain differently.
- Calculate COCA per channel (blended where attribution is missing), test the CAC Cap Rule and read the TLV:COCA ratio against the bands below.
- Set up or check the attribution method (enquiry source log, UTM links, monthly source report) and fix the attribution model before the campaign starts.
- Calculate break-even customers per month, build the 12-month projection and, where the client forecasts pipeline, the bottom-up model with stage weighting and deal velocity.
- Write the talking points for the client's actual ratio; for a board that has not yet agreed to fund social media, build the eight-section case in investment-business-case.
- Check every formula against its inputs and label projections as estimates; correct any mismatch and rerun the affected sections, then run the anti-slop gate before release.
Section templates, worked examples, FRAT scoring and the $20 Rule are in ROI model method.
Formulas and ratio bands
- TLV = average transaction value × transactions per year × relationship length (years) (Bodnar and Cohen, 2012).
- COCA = monthly social media spend ÷ new customers acquired per month via that channel.
- ROI = (TLV − COCA) ÷ COCA (Bodnar and Cohen, 2012).
- Break-even customers = monthly investment ÷ (TLV × gross margin %), rounded up.
- Monthly ROI = (TLV × customers acquired − monthly investment) ÷ monthly investment × 100.
- CAC Cap Rule (Kahan, 2022): CAC ≤ (CLV × 0.25).
| TLV:COCA | Interpretation | Action |
|---|---|---|
| 10:1 or above | Excellent | Scale the best-performing channel. |
| 5:1 to 9:1 | Healthy and sustainable | Maintain; optimise underperforming channels. |
| 3:1 to 4:1 | Borderline viable | Cut spend on the highest-COCA channel; improve conversion. |
| Below 3:1 | Acquisition cost too high | Grow TLV (price, upsell, retention) or reduce COCA (targeting, funnel). |
Outputs
| Artefact | Consumer | Acceptance condition |
|---|---|---|
| Seven-section ROI model (TLV, COCA by channel, ratio, attribution method, break-even, 12-month projection, talking points) | Consultant and client finance lead | Every calculation is shown step by step with the client's figures and every assumption is listed. |
| Break-even statement in plain language | Client decision-maker | Reads "If social media brings in just [X] new customers per month, the investment pays for itself" with the client's numbers. |
| Per-channel cohort CLV or retention report | Consultant; meta-budget-planner |
Built with the retention reference; Week-4 comparators labelled provisional. |
| Eight-section investment business case | Board, finance committee or funder | Follows the template order and shows conservative and realistic ROI scenarios. |
Evidence Produced
| Evidence | Format | Acceptance condition |
|---|---|---|
| Assumption register | Table: input, value, source, estimate or measured | Every estimated figure (customers, margin, relationship length) is named as an assumption. |
| Attribution method record | Enquiry source log fields and UTM parameters used | The model named is the one fixed before the campaign; blended COCA is labelled an estimate. |
| Calculation workings | Step-by-step tables per section | A reviewer can recompute TLV, COCA, ratio and break-even from the inputs shown. |
Capability and Permission Boundaries
Read and search only; analysis is read-only. Publishing, spend, live account changes, outreach and personal-data processing need explicit, action-specific client authority. Using a customer list for FRAT scoring, NPS broadcasts or direct campaigns needs a lawful basis and consent records (register UG-DPPA-2019).
Degraded Mode
Without attributable revenue or lifetime value and complete costs, return the narrowest qualified result and mark the affected checks not assessed. The attribution method, intake assumptions and a break-even target at a stated margin can still be delivered, labelled as projections and never as results.
Decision Rules
| Condition | Action | Failure or risk avoided |
|---|---|---|
| Channel-level attribution is not available | Use total monthly investment ÷ total new digital customers as blended COCA, label it an estimate and recommend enquiry source tracking. | A per-channel figure the data cannot support. |
| New customer numbers are unknown | Use enquiry volume as a proxy (COCA per enquiry) and apply the client's conversion rate; state every assumption. | Invented customer counts. |
| CAC exceeds 25% of CLV | Treat the programme as structurally unprofitable regardless of gross revenue and say so in the recommendation. | Scaling a programme that loses money on each customer. |
| TLV cannot be calculated (relationship length unknown) | Use a conservative single-transaction value and note that it understates the true return; recommend retention tracking. | An inflated TLV. |
| The client needs per-channel retention or cohort LTV before the ratio can be trusted | Build acquisition or behaviour cohorts with retention-cohorts-and-ltv and use per-channel CLV in the TLV:COCA ratio. | A single average CLV hiding one-time-buyer channels. |
| Leadership has not yet agreed to fund social media and asks for a business case | Build the eight-section case with investment-business-case, using this skill's ROI formula for its two scenarios. | An ROI model presented to a board that first needs the risk and investment argument. |
| COCA is above the benchmark | Acknowledge it and name the specific action (grow TLV, improve enquiry-to-sale conversion, or move spend from the underperforming channel). | Deflection that loses the client's trust. |
| The requested outcome is budget allocation or a periodic report | Route to meta-budget-planner or meta-reporting and hand over the verified inputs already collected. |
Neighbour collision and duplicated work. |
Quality Standards
- Uses the correct Bodnar and Cohen (2012) formula: ROI = (TLV − COCA) ÷ COCA.
- All calculations are shown step by step with the client's actual figures, not only the final result.
- COCA is calculated per channel where attribution data exists; blended COCA is used and labelled as an estimate where it does not.
- Ratio interpretation names the specific action required at the client's ratio level, not generic advice.
- The attribution section gives the client a practical tracking system to implement, not a conceptual explanation.
- Break-even is expressed in plain language the client can use in a conversation.
- The 12-month projection table is populated with the client's figures and carries a clear caveat about projection assumptions.
- Talking points are specific to the client's ratio and figures, not a generic script; currency follows ROI model method § Quality criterion kept from the HEAD checklist.
Anti-Patterns
- Presenting projections as results. Fix: label every forward figure an estimate and revisit the projection quarterly with actual acquisition data.
- Choosing or switching the attribution model after the campaign has run. Fix: select it before launch and apply it consistently for the strategy period (Hanlon and Tuten, 2022).
- Presenting the retainer as a standalone cost. Fix: set it against TLV and the break-even customer count.
- Using one average CLV for every channel. Fix: calculate CLV per acquisition cohort so one-transaction channels show up.
- Counting all pipeline at full value. Fix: weight opportunities by stage (10%, 30%, 60%, 85%, 100%) before forecasting.
- Claiming accounting-level precision from social attribution. Fix: state that the method gives directional understanding, because social content influences sales tracked to other channels.
- Absorbing budget allocation (
meta-budget-planner) or periodic reporting (meta-reporting) into this workflow. Fix: route the neighbouring output and hand over verified inputs.
References
- ROI model method: read when collecting the intake, building any of the seven sections, modelling revenue bottom-up, or applying FRAT and the $20 Rule.
- retention-cohorts-and-ltv: read when comparing acquisition or behaviour cohorts, Week-4 retention or LTV by channel.
- investment-business-case: read when a board or funder needs a business case for social media investment.
meta-budget-planner: read when the question becomes spend allocation or the bottom-up revenue plan.meta-reporting: read when the output is periodic performance reporting.- Measurement proof pack: read when assembling measurement evidence for release.
- Anti-AI slop production gate: read when drafting talking points and the business case.
ai-slop-audit: read at the release checkpoint.