Pricing Strategy and Revenue Models Questions

Headline: Recommend +8% price increase on our Core SaaS plan to maximize 12‑month ARR while preserving retention

Executive summary (1 sentence): Modeling shows an estimated incremental net ARR of +$3.6M over 12 months from an 8% price increase on Core, assuming a modest 3.0% uplift in churn and a 2% conversion/upsell effect—see charts & sensitivity table.

Three supporting charts (described):

1. Current ARR Waterfall — stacked bar by plan (Core, Pro, Enterprise) showing base ARR, churned ARR, upsell ARR (last 12 months) to show Core’s revenue concentration.
2. Price vs. Revenue curve (modeled) — x-axis: % price change (-10% to +20%); y-axis: projected net ARR change at 12 months using base elasticity; highlights optimal region and breakeven point.
3. Churn & New MRR projection by month (12 months) — two-line chart: modeled monthly churn rate after price change vs. baseline churn, and cumulative Net New MRR to show time-phasing of impact.

Estimated net ARR impact (12 months): +$3,600,000
Key assumptions:

• Base ARR (Core plan) = $45,000,000 annualized
• Proposed price increase = +8%
• Price elasticity of demand (Core) = -0.4 (i.e., 0.4% loss in volume per 1% price increase)
• Immediate gross churn increase = 3.0 percentage points for affected cohort in month 1, decaying linearly to +1.0 pp by month 6
• No change in new sales pipeline velocity except 2% fewer conversions; 1.5% uplift in ARPA from upsells
• Discounting and one-time refunds negligible

Sensitivity table (net ARR delta at 12 months):

Price elasticity (ε) | Net ARR change
-0.2 | +$6.0M
-0.4 (base) | +$3.6M
-0.6 | +$1.2M
-0.8 | -$0.9M

How ARR impact is calculated (data & formula):
Data required:

• Subscriber counts by cohort & plan (monthly)
• Monthly ARPA by plan
• Historical churn and new MRR by cohort
• Pipeline conversion rates and average contract length
• Elasticity estimates (from past price tests / cohort analysis)

Formulas:

• New ARPA = ARPA_base * (1 + price_change)
• Expected retention factor ≈ 1 + (ε * price_change) where ε is elasticity (negative)
• Projected ARR_t = Σ_cohorts [Subscribers_base_cohort * retention_factor_cohort_t * New_ARPA_cohort]
• Net ARR delta = Σ_t (Projected ARR_t - Baseline ARR_t) over 12 months (or simulate monthly and annualize)
• Include churn dynamics: subscribers_t = subscribers_{t-1}*(1 - churn_t) + new_signups_t

Most important assumptions:

1. True elasticity for Core customers (drives volume loss)
2. Churn uplift magnitude and duration after change
3. Impact on new sales/pipeline conversion and upsell behavior

Risks & mitigations:

• Risk: Higher-than-expected churn → Mitigate with grandfathering (existing customers protected for 6 months), targeted retention offers for high-risk cohorts, and proactive CS outreach.
• Risk: Sales slowdown → Mitigate by pausing price increase for new signups until pipeline clears; offer time-limited grandfathered pricing for deals in negotiation.
• Risk: Competitive reaction → Monitor competitor pricing; prepare promotional campaigns and value messaging; run A/B test in a region before full roll-out.

Recommendation: Pilot +8% in a test cohort (10–20% of Core ARR) for 6–8 weeks, measure elasticity and churn uplift, then roll out with mitigations if observed metrics are within modeled ranges.

Overview: Treat DoorDash as a two-sided marketplace where price changes affect consumer order frequency/size and restaurant acceptance/pricing decisions. Goal: maximize long-run GMV and profit while keeping platform take-rate stable and aligning incentives (higher consumer conversion vs. restaurant coverage/earnings).

Measuring elasticity

• Consumers: estimate own-price elasticity of demand (ε_c = %Δorders / %Δprice). Use log-log regression on order counts or probability of purchase at listing-level, controlling for restaurant, time, and user fixed effects. Segment by distance, time-of-day, cuisine, and user price-sensitivity (RFM).
• Restaurants: estimate elasticity of supply/participation (ε_r = %Δacceptance or %Δhours-active / %Δcommission or %Δper-order payout). Model acceptance probability and menu-price adjustments with panel regressions and discrete choice (restaurant chooses whether/when to be available).

Experiment types to estimate elasticities

1. A/B price experiments (randomized): show different consumers slightly varied delivery fees or menu-level discounts to measure conversion and order size elasticities.
2. Price ladder (multi-arm): expose multiple price points to estimate a demand curve non-parametrically.
3. Randomized encouragement / voucher: give subsets coupons to isolate demand responses while controlling for confounding.
4. Restaurant-side randomized commission pilots: randomly vary take-rate or guaranteed payouts for matched restaurant cohorts to observe changes in acceptance, prep times, and menu prices.
5. Instrumental variables / regression discontinuity: use exogenous shocks (e.g., fuel surcharge changes regionally) to identify causal effects where randomization not feasible.
6. Repeated-measures and adaptive experiments: use bandits for revenue-optimizing price discovery, but complement with randomized arms for valid elasticity estimation.

Telemetry & instrumentation (must be precise and queryable)

• Experiment metadata: experiment_id, cohort, assignment timestamp, seed, exposure reason.
• Consumer-side events: impressions (price components shown), click-to-restaurant, add-to-cart, checkout_shown, order_submitted, order_value, coupon_applied, delivery_fee_displayed, effective_price (total paid).
• Restaurant-side events: menu_price_displayed, menu_price_effective, acceptance_offer, accept_timestamp, reject_reason, prep_time, availability_status_changes, payout_received, aggregated daily hours/active_minutes.
• Contextual fields: user_id (hashed), restaurant_id, market_id, location, distance, ETA, time_bucket, device_type, marketing_channel, promo_id.
• Financial fields: platform_fee, delivery_fee, driver_payout, gross_order_value, cost_components.
• External signals: weather, competitor price snapshots, local events.

Analysis & dashboards (BI responsibilities)

• Core dashboards: real-time take-rate, GMV, orders, conversion funnel by price bucket, restaurant acceptance rate, average payout per order.
• Elasticity view: demand curve plots (price vs. conversion/order size), log-log regression outputs, segment elasticities, CIs.
• Experiment dashboard: per-experiment KPI panel (power, randomization balance, primary/secondary metrics, SATE & HTE), data quality checks (assignment leakage, missing fields).
• Longitudinal monitoring: LTV impact, churn by price cohort, restaurant retention vs. commission.
• Alerts: sudden shifts in elasticity estimates, imbalance in experiment assignments, drop in restaurant availability in markets.

Implementation notes & trade-offs

• Use randomized experiments where possible for causal estimates; complement with observational econometrics to generalize.
• Account for cross-side externalities: run joint experiments or include interaction terms (consumer price × restaurant commission) to capture feedback loops.
• Guardrail metrics: restaurant earnings, delivery times, consumer NPS. Ensure short-run price optimizations don’t degrade supply or long-term retention.
• Sampling & power: pre-calc minimum detectable effects; stratify by market/segment to ensure representativeness.

This design lets BI provide accurate elasticities, run causal experiments, and maintain dashboards/alerts to keep incentives balanced while protecting platform take-rate and long-term marketplace health.