Investment case: opening a store, a line, IT (NPV, IRR, payback)
A calculation of whether an investment pays back: project cash flows discounted to today’s money and compared with the cost of capital. It drives decisions on whether to open a store, launch a new line or buy a system, and in what order to run projects when money is short.
Why this document
A rouble today is worth more than a rouble in three years, and the money put into a store or a system comes back slowly and unevenly. Simple payback ignores this, so net present value and internal rate of return are used. The investment committee compares projects on one scale and picks those that create more value, not just more revenue. The case matters after launch too: a year later actuals are checked against plan, so assumption errors are not repeated in the next project.
At a glance
- What is this document and why is it needed?
- A calculation of whether an investment pays back: project cash flows discounted to today’s money and compared with the cost of capital. It drives decisions on whether to open a store, launch a new line or buy a system, and in what order to run projects when money is short.
- Who owns the document and how often is it rebuilt?
- A financial analyst prepares, the finance director checks, an investment committee or the owner decides; before each significant investment and after the first 12 months (a project post-audit).
- What mistakes are the most common?
- The whole new store revenue is taken → the effect is overstated → count the net increment after cannibalization. Working capital is ignored → payback overstated → include stock and deduct payables; return it at the end. Plan revenue from day one without a ramp → NPV overstated → assume 3 to 12 months to reach plan.
For whom
- Brand
- A brand evaluates lines, factories, tooling, own stores and IT. The key points are cannibalization (part of new-line sales is taken from old styles) and the return of working capital at the end. The Model B capsule line at a rate of 22.5% gives NPV RUB 17.362 million, IRR 94.3%, payback 1.0 years. Reality check: 14 to 16 thousand capsule units equal 18 to 20% of the annual Model B volume (80,000 units), so such sales are acceptable only if wholesale preorders confirm them and cannibalization stays at or below 12%.
- Retail
- A retailer evaluates stores, refurbishments, a warehouse and IT. The key points are the ramp to plan sales, cannibalization of neighboring stores (8%) and the halo effect (online sales growth in the region, 4%). The store in the example is the average Model R store: revenue after markdowns and returns 30,940 a year (RUB 257,833 per m²), four-wall EBITDA 3,219 (10.4%); the same figures stand in the “Store P&L” document. At a rate of 19.8% NPV is −RUB 2.512 million, IRR 8.2%, payback 4.5 years: in the network format the case fails, a mature revenue of at least 34,278 (11% above the average store) or lower rent is needed.
What it consists of
What the case contains
- A project description and the “do nothing” alternative: only incremental cash flows against that alternative are evaluated.
- Investment in year zero: CAPEX (fit-out, equipment, tooling, IT), deposits and rent payments, the initial stock and working capital, pre-opening costs.
- Cash flows by year: revenue ramping to plan, gross profit, variable and fixed project costs, corporate income tax net of depreciation.
- Residual value and release of working capital in the final year.
- The discount rate: company WACC for projects of typical risk (a store in the network format, IT); for a new format, region or line add 2 to 4 pp for project risk. The cost of debt in WACC is the price of new funding (see “Treasury”), not the rate of the existing loan; show sensitivity to the rate.
Metrics
- NPV (net present value): the sum of discounted cash flows; positive means the project earns more than the cost of capital.
- IRR (internal rate of return): the rate at which NPV equals zero; compared with the cost of capital.
- Payback: in how many years the accumulated flow covers the investment; it ignores the time value of money. DPP (discounted payback): the same on discounted flows.
Cannibalization and halo
- Cannibalization: a new outlet or line takes sales from your own. The case takes the net increment: new sales × (1 − cannibalization share), and the gross profit lost in the old is deducted.
- Retailer: Halo: a store raises awareness and online sales in the region; include only a measured effect (from test regions), otherwise the case is “painted”.
The “store opening” case
- Retailer: Drivers: area, revenue per m², conversion and traffic, gross margin, rent (fixed or percentage), staff, utilities, opening marketing, time to reach plan (3 to 12 months).
- The decision is made on NPV and on a safety margin: by how much revenue may fall and rent may rise before NPV hits zero.
Brand and retailer: what is computed differently
- Brand: a brand evaluates lines, tooling and own stores: the key points are cannibalization of old styles, wholesale preorders as demand confirmation and the return of working capital at the end.
- Retailer: a retailer evaluates stores, refurbishments, a warehouse and IT: the key points are the ramp to plan sales, cannibalization of neighboring outlets, rent and its indexation; store revenue decides, as the sensitivity shows.
Brand: how the document differs and what to add
- Brand: brand investment cases are lines, capsules, new channels and collaborations; a brand project pays back faster than a store (capsule 1.0 years) but cannibalizes the old line.
- Brand: the capsule starts with investment of RUB 5 million and working capital of 8,000: working capital for stock exceeds capex and returns only with sales.
- Brand: line cannibalization (12% of units at the old price of RUB 2,950) is deducted from the flow: the project is valued by incremental profit, not line revenue.
- Brand: the discount rate of a brand is higher with a long cycle and customer concentration: the brand’s cost of capital is used, not the retailer’s.
- Brand: a case for a new channel (marketplace, own store) includes acquisition cost, returns and logistics: payback is computed per customer and per unit.
- Brand: a collaboration case includes license payments, minimum guarantees and one-off launch costs; the effect on awareness is noted separately, without a monetary value.
- Brand: brand sensitivity is computed for price, volume, FX and returns: a small error in price or FX is often stronger than an error in volume.
- Brand: the project decision is recorded with a stop condition: a weekly sales threshold below which the remaining batch is not bought.
Retailer: how the document differs and what to add
- Retailer: the store case is built on the average store of the network: revenue RUB 30.94 million a year, gross margin 48.3%, variable 11.1%, four walls RUB 3.219 million.
- Retailer: in the base NPV is −RUB 2.512 million, IRR 8.2%, payback 4.5 years: the average store does not earn its capital, and the decision depends on the revenue that the format proves.
- Retailer: ramp-up is given by a curve (70%, 90%, then mature revenue): putting mature revenue in from year one overstates NPV.
- Retailer: cannibalization of neighboring stores (8% of zone revenue) and the online halo (4%) are shown as separate lines so that the decision does not hinge on a disputed effect.
- Retailer: rent is the main assumption: indexation, turnover rent, term and exit conditions change NPV more than margin.
- Retailer: capex includes fit-out, equipment, signage, initial stock fill and opening costs; stock goes into working capital, not capex.
- Retailer: a pilot of a store or a new format is first valued with LFL analogs and site traffic, then scaled; a closure case compares store cash flow with exit payments.
- Retailer: an IT project case (for example loss control or RFID) computes loss savings, working-capital release and implementation costs; the effect is verified by a pilot.
How to read the result and which decisions follow
- NPV is positive but payback is longer than the lease term: the decision is reconsidered because the flow after the lease belongs to nobody; negotiate the term, not margin.
- NPV is negative at base assumptions but positive with revenue 10 to 15% higher: a measurable argument for that revenue (traffic, analog, test) is needed before approval.
- The result rests only on cannibalization or halo: exclude them and compare; a project that survives without a disputed effect is more reliable.
Control checks
- Year zero contains all payments before launch, including working capital; NPV is recomputed independently (the sum formula and the NPV function give the same figure).
- The rate equals the participant cost of capital stated in the cost-of-capital section; comparing projects at different rates without explanation is not acceptable.
- Typical traps: revenue with VAT, corporate income tax ignored, depreciation subtracted as a cash cost, working capital not returned in the final year, residual value taken without justification.
Where the data comes from
| Data source | Fields |
|---|---|
| Data of operating stores or analog lines | revenue per m², conversion, basket, margin, returns, time to plan |
| The premises or project proposal | rent and indexation, rent holidays, deposit, fit-out estimate |
| Financial model and the cost of capital | WACC, taxes, working capital assumptions |
| Tests and measurements | pilot results, control stores, regional halo effect |
How to build it step by step
- Step 1: Formulate the project, the alternative and the horizon (a store 5 years, a line the collection life, IT 5 years).
- Step 2: Collect year-zero investment: CAPEX, deposit, stock, working capital; VAT separately.
- Step 3: Build revenue from drivers with a ramp to plan; compare with operating analogs.
- Step 4: Compute gross profit, variable and fixed project costs.
- Step 5: Take cannibalization and halo into account; deduct the lost profit of old outlets.
- Step 6: Add corporate income tax with depreciation and the release of working capital at the end.
- Step 7: Compute NPV, IRR, payback and DPP at the cost of capital.
- Step 8: Run sensitivity: revenue ±20%, rent ±15%, rate ±3 pp, a delay in reaching plan; find the zero-NPV point.
- Step 9: Compare projects if money is not enough for all (profitability index NPV ÷ investment).
- Step 10: After launch run a post-audit: actual versus case at 6 and 12 months; feed conclusions into the next cases.
Formulas
| Measure | Formulas | Note |
|---|---|---|
| NPV | NPV = Σ CF_t ÷ (1 + r)^t, t = 0…N | CF_0 is negative (the investment); r is the cost of capital of the project. |
| IRR | IRR: the rate r at which Σ CF_t ÷ (1 + r)^t = 0 | With alternating signs there may be several roots; rely on NPV. |
| Payback and DPP | PB: the year in which Σ CF becomes ≥ 0; DPP is the same for CF ÷ (1 + r)^t | The fractional part uses linear interpolation within the year. |
| Net sales increment | Project contribution = Sales × (1 − Cannibalization) × store GM − Sales × Variable % − Fixed + Sales × Halo × Online contribution | Measure cannibalization on neighboring stores before and after opening. Halo is the region online sales: it goes on a separate line with the online contribution (online GM minus last mile and acquiring), not with the store margin. |
| Profitability index | PI = (NPV + Investment) ÷ Investment | Helps rank projects under a money limit. |
Formulas in notation
- Net present value
NPV = Σ_(t=0..T) CF_t / (1 + r)^t
CF_t is the cash flow of year t (negative at zero: capex and working capital), r is the participant cost of capital.- Internal rate of return
NPV(IRR) = 0
The rate at which NPV equals zero; compared with the cost of capital, not with the bank deposit rate.- Payback
T_pb = t − 1 + |C_(t−1)| / CF_t, C_t = Σ CF
C is the cumulative flow; t is the first year when C becomes non-negative; the discounted variant divides CF_t by (1 + r)^t.- Four-wall cash flow
CF = Rev · GM − V − F − ΔWC
GM is gross margin, V is variable costs, F is store fixed costs, ΔWC is the change in working capital.- Project breakeven revenue
Rev*: NPV(Rev*) = 0
The mature revenue at which project NPV equals zero; found by iteration, then compared with the format’s attainable revenue.
Worked example
Three investment cases of the end-to-end models: cash flows by year, NPV, IRR, payback, RUB thousand
| Case | Y0 | Y1 | Y2 | Y3 | Y4 | Y5 | NPV | IRR, % | Payback, years | DPP, years |
|---|---|---|---|---|---|---|---|---|---|---|
| Retailer: a new 120 m² store in the network format (revenue of the average Model R store) | −7,154 | −729 | 1,269 | 2,078 | 2,078 | 5,303 | −2,512 | 8.2 | 4.5 | no payback |
| Brand: a capsule line | −13,000 | 12,705 | 14,835 | 18,575 | 17,362 | 94.3 | 1 | 1.3 | ||
| Retailer: an inventory accuracy system (IT) | −5,000 | 3,841 | 841 | 841 | 841 | 841 | 30 | 20.2 | 2.4 | 4.9 |
| Store: mature revenue −10%: NPV | −4,898 | |||||||||
| Store: mature revenue +10%: NPV | −182 | |||||||||
| Store: mature revenue +20%: NPV | 2,148 | |||||||||
| Store: rent +15%: NPV | −3,629 | |||||||||
| Store: rent −15%: NPV | −1,394 | |||||||||
| Store: revenue +10% and rent −15%: NPV | 936 | |||||||||
| Store at cannibalization 0%: NPV | 68 | |||||||||
| Store at cannibalization 15%: NPV | −4,821 | |||||||||
| Break-even mature store revenue, RUB thousand (the NPV column holds revenue, not NPV) | 34,278 |
Rates come from the “Capital” document: retailer WACC 19.8%, brand 22.5% (cost of new debt 20%). The store comes from Model R: fit-out 3,929 (network depreciation over 5 years ÷ 14 stores) + working capital 3,225 (net stock minus payables per store), mature revenue 30,940 reaching plan in two years, gross margin 48.3%, fixed store costs 8,282 (rent 3,024, utilities 429, staff with contributions 4,829), variable costs 11.1% (marketing, acquiring, warehouse, packaging from the P&L). Break-even mature revenue is 34,278; revenue +10% gives NPV -182, rent −15% gives -1,394, both together 936. Cannibalization of 8% eats the whole safety margin: NPV is 68 at zero and -2,512 at 8%; the main lever is still the store revenue. The figures are illustrative while the method is transferable; a flagship format (revenue 1.5 to 1.8 times the network) is a separate case that must prove such revenue. The store payback of 4.5 years is computed on the full investment (fit-out plus working capital 7,154) after corporate income tax, with 8% cannibalization and a yearly ramp (70, 90, 100%); in the Store P&L document the same store pays back only the fit-out from the four-wall contribution before tax, without cannibalization and working capital, in about 15 months. These are two definitions of one store, not two stores: the opening decision is made on the first, the second shows how fast the point returns the money spent on fit-out and equipment.
Figures
- Store (retailer)
- Capsule line (brand)
- IT project (retailer)
Data as a table
| Measure | Store (retailer) | Capsule line (brand) | IT project (retailer) |
|---|---|---|---|
| Year 0 | −7,154 | −13,000 | −5,000 |
| Year 1 | −7,883 | −295 | −1,159 |
| Year 2 | −6,614 | 14,540 | −318 |
| Year 3 | −4,536 | 33,115 | 523 |
| Year 4 | −2,458 | 33,115 | 1,364 |
| Year 5 | 2,845 | 33,115 | 2,205 |
Cumulative cash flow of three projects. The year the line crosses zero is the payback; the retailer store does not reach it on a discounted basis in the base, the brand capsule pays back in 1.02 years (after the first year the cumulative flow is still -295).
Data as a table
| Measure | NPV |
|---|---|
| Store | −2,512 |
| Capsule line | 17,362 |
| IT project | 30 |
NPV of projects at the participant cost of capital. The same formula: a project with a negative NPV has no ground to start without a change in assumptions.
- −6,096 … 0 … 3,170 RUB thousand
Data as a table
| Rent −15% | Rent base | Rent +15% | |
|---|---|---|---|
| −10% | −3,725 | −4,898 | −6,096 |
| Base | −1,394 | −2,512 | −3,629 |
| +10% | 936 | −182 | −1,299 |
| +20% | 3,170 | 2,148 | 1,030 |
Store NPV: revenue and rent. The store turns positive when mature revenue grows by about 11% or when rent falls together with revenue growth; read it as a tolerance map.
Specifics by category
- Apparel
- Apparel: sales are seasonal, compute the store ramp from the opening season; in a line case include the residual stock and price reductions.
- Footwear
- Footwear: higher investment in size stock and fitting areas; returns and stock turnover time shift payback more.
- Accessories
- Accessories: less area and stock, faster payback; a large role for the window, small shrink; for department store corners the key is a percentage of turnover.
- Kids
- Kids: the ramp depends on school and winter; a larger area because of prams and families, a higher share of repeat purchases; certification in the line investment.
- Men
- Menswear: steadier revenue per m², lower risk, a higher core share; more predictable payback.
- Women
- Womenswear: higher revenue per m² but also dead stock and markdown risk; include markdowns and a weak-season scenario in the case.
Common mistakes
- The whole new store revenue is taken → the effect is overstated → count the net increment after cannibalization.
- Working capital is ignored → payback overstated → include stock and deduct payables; return it at the end.
- Plan revenue from day one without a ramp → NPV overstated → assume 3 to 12 months to reach plan.
- IRR is compared with the loan rate, not the cost of capital → weak projects accepted → compare with WACC and project risk.
- Halo without measurement → the case is “painted” → include only a measured effect.
- No post-audit → the same errors in every case → compare actuals with the case at 6 and 12 months.
Russian specifics
- Corporate income tax 25% (Art. 284 of the Tax Code as amended by Law 176-FZ); for USN and reduced regional rates compute differently; take depreciation and tax in the case from tax accounting. Rule: Tax Code Art. 284: Tax Code, Art. 284 (corporate income tax rates) (rule check not completed).
- Leases: under FSBU 25/2018 the lessee recognizes a right-of-use asset and a liability; in an investment case rent cash flows go as operating payments, while on the balance sheet and in debt they are treated differently; do not count rent twice. Rule: Order 208n of 16.10.2018: Ministry of Finance Order of 16.10.2018 No. 208n (FSBU 25/2018 Lease accounting) (rule check not completed).
- Input VAT on capex is deductible under general rules (Art. 171 of the Tax Code); in the case show the investment without VAT and the VAT flow separately. Rule: Tax Code Art. 171: Tax Code, Art. 171 (VAT deductions) (rule check not completed).
- Equipment leasing: the legal basis is in the leasing law; compare buying and leasing by NPV, taking tax effects into account. Rule: Law 164-FZ of 29.10.1998: Federal Law of 29.10.1998 No. 164-FZ on financial lease (leasing) (rule check not completed).