Ecommerce Fulfillment Growth Strategies: 10 Proven Ways To Scale Smarter

Table of Contents

Some links on The Justifiable are affiliate links, meaning we may earn a small commission at no extra cost to you. Read full disclaimer.

Ecommerce fulfillment growth strategies matter most when sales start rising faster than operations can comfortably handle. More orders should create momentum, not stockouts, late shipments, rising labor costs, or a flood of “Where is my order?” tickets.

The challenge is that fulfillment rarely fails in one dramatic moment; it becomes less efficient as order volume, SKU count, channels, and customer expectations increase.

This guide shows you how to build a fulfillment operation that can absorb growth without unnecessary complexity, protect margins, and maintain a reliable customer experience as your business moves from manageable volume to repeatable scale.

How Ecommerce Fulfillment Becomes a Growth Constraint

Fulfillment is the operational bridge between a paid order and a satisfied customer. When that bridge is designed for yesterday’s volume, growth exposes weak inventory rules, manual steps, carrier dependence, and poor visibility very quickly.

Understand Where Fulfillment Starts to Break

A small ecommerce business can often fulfill orders with simple processes because the owner or a small team remembers where products are stored, which orders are urgent, and what to do when something goes wrong. The first warning sign is usually not total failure. It is friction. Picking takes longer. Inventory counts become less trustworthy.

A few orders miss the daily carrier cutoff. Customer service starts checking tracking pages manually. The team creates more spreadsheets because the existing system no longer answers basic questions quickly enough.

Look for four forms of strain: capacity, complexity, variability, and visibility. Capacity strain means you simply cannot process enough orders per day. Complexity strain appears when bundles, subscriptions, wholesale orders, special packaging, or multiple channels create too many exceptions. Variability comes from launches, promotions, seasonality, and unpredictable demand. Visibility strain occurs when your storefront, warehouse, purchasing, and shipping data disagree.

I recommend treating these symptoms as system signals rather than isolated employee mistakes. Hiring another picker may temporarily increase throughput, but it will not fix inaccurate stock levels or inefficient routing rules. Before adding people, warehouses, or software, identify which form of strain is actually limiting growth.

Define What “Scaling Smarter” Means for Your Business

Smarter fulfillment does not mean building the fastest possible operation at any cost. It means increasing order volume while controlling cost, service quality, inventory risk, and operational complexity.

A useful way to define your target is to choose a service promise and an economic boundary. Your service promise might be “ship stocked orders within one business day” rather than “deliver everything in two days.” Your economic boundary might be a maximum fulfillment cost per order or a minimum contribution margin after pick-and-pack, packaging, shipping, and returns.

This distinction matters because two stores with the same order volume can require very different fulfillment models. A brand selling lightweight skincare can distribute stock across several locations more easily than a furniture seller with bulky, slow-moving products. A subscription brand values predictable recurring volume, while a trend-driven apparel store may need flexibility around launches and returns.

Write down what you are actually optimizing for: speed, cost, accuracy, inventory efficiency, flexibility, or some weighted combination. Then rank those priorities. That simple exercise prevents expensive decisions driven by competitor promises rather than your own economics.

I suggest scaling the service promise your customers value, not the fulfillment complexity your competitors happen to advertise.

Build a Fulfillment Baseline Before You Add Capacity

Before changing warehouses, systems, or partners, document how orders move today and what each step costs. A clear baseline makes it easier to distinguish a real bottleneck from an assumption.

Map the Complete Order-to-Delivery Workflow

Start with one ordinary order and follow it from checkout through delivery. Record every system, handoff, manual decision, and waiting period. If you sell through Shopify, marketplaces, wholesale, or social channels, map how each order source reaches your inventory and fulfillment process.

Your workflow should cover order capture, payment or fraud review, inventory allocation, picking, packing, label creation, carrier handoff, tracking communication, delivery, and returns. Add exceptions such as address corrections, split shipments, backorders, damaged stock, gift notes, and oversold items.

Then mark each step as automated, manual, or exception-driven. Manual work is not automatically bad. A manual quality check on a high-value order may be intentional. For example, if a team member chooses a carrier service manually for every order, doubling sales may nearly double that workload. If the choice can be governed by weight, destination, delivery promise, and cost rules, the same team can process more orders without making hundreds of additional micro-decisions.

The goal is not to automate everything. It is to see where time, errors, and delays enter the process so you know where intervention will create the largest operational return.

Establish Cost, Speed, Accuracy, and Inventory Benchmarks

Growth becomes difficult to manage when “fulfillment is getting expensive” is the only diagnosis available. Replace broad impressions with a small baseline of operational metrics.

Track fulfillment cost per order, shipping cost per order, average order processing time, order accuracy, on-time shipment rate, return rate, stockout frequency, inventory days on hand, and the percentage of orders that require manual intervention. If you operate your own warehouse, include direct labor and packaging. If you use a third-party logistics provider, separate storage, receiving, pick-and-pack, shipping, and exception fees where possible.

Do not chase a universal benchmark. Compare performance primarily against your own historical baseline and the margin available in each order.

Segment the numbers too. A blended shipping cost can hide the fact that one region, product family, or oversized SKU is destroying margin. The same applies to returns and fulfillment errors.

Once you can see which order types are expensive or unreliable, you can prioritize changes that remove the most friction instead of improving averages that were never the real problem.

Identify the Bottleneck That Will Break First

A scalable plan starts with the next constraint, not every possible future constraint. Ask what would fail first if order volume increased by 50% without changing your current operation.

ALSO READ: Ecommerce Marketing Platform Review: What Actually Works for Sales Growth

It might be receiving capacity because inbound inventory already sits unopened for two days. It could be picking because your fastest-selling products are scattered across the warehouse. It may be purchasing because reorder decisions depend on someone checking a spreadsheet. In another business, the warehouse is fine but carrier collection capacity or customer support cannot absorb a promotional spike.

Use a simple stress test. Estimate peak daily orders, average items per order, available labor hours, packing stations, carrier cutoff times, and required throughput per hour. Then examine whether inventory, systems, and people can support that demand without excessive overtime or service failures.

If packing is the constraint, adding another sales channel before improving packing capacity may magnify the problem. If inventory accuracy is the constraint, distributing stock across multiple warehouses too early may make the data problem harder.

Fix the constraint with the highest growth impact first, then repeat the test. Fulfillment scaling works best as a sequence of controlled upgrades rather than one large transformation project.

Build Inventory Efficiency With Strategies 1–3

Inventory decisions determine how much cash you tie up, how often you stock out, and how difficult fulfillment becomes. These first three ecommerce fulfillment growth strategies create a stronger inventory foundation before you add locations or partners.

Strategy 1: Forecast Demand by SKU, Channel, and Region

A single storewide sales forecast is not precise enough for fulfillment planning. You need to understand what will sell, through which channel, and where the demand is likely to originate.

Start with SKU-level order history and separate normal demand from promotions, launches, stockouts, and unusual events. A product that sold 300 units last month may not have true monthly demand of 300 if it was unavailable for ten days or heavily discounted for a weekend. Add known future factors such as planned campaigns, seasonal peaks, wholesale commitments, and supplier changes.

Regional forecasting becomes more important when you consider distributed inventory. If 45% of orders for one product consistently come from the western United States, that pattern may support placing more units closer to those customers. But if demand is volatile and the SKU sells slowly, splitting inventory can create stranded stock.

Use forecast ranges rather than pretending demand is perfectly predictable. Create a base case, an upside case, and a downside case for important SKUs. This helps purchasing and operations prepare for uncertainty without treating the most optimistic sales scenario as guaranteed.

It is to make replenishment, labor, and inventory placement decisions with better evidence than last month’s total sales alone.

Strategy 2: Set Reorder Points and Safety Stock From Real Lead Times

Reorder points are most useful when they reflect how your supply chain actually behaves. A basic formula is expected demand during replenishment lead time plus a safety-stock buffer. The difficult part is using realistic inputs.

Measure lead time from the moment you decide to replenish until units are actually available to pick. That can include supplier production, freight, customs, receiving, inspection, and warehouse put-away. If your purchase order arrives on Monday but stock is not available until Thursday, those three days matter.

Safety stock should cover uncertainty, not compensate indefinitely for poor planning. Fast-moving, high-margin SKUs with variable supplier lead times may justify a larger buffer. Slow-moving products with reliable replenishment may need less. If you use an inventory system such as Cin7, centralized stock visibility can help support more structured replenishment decisions, but software cannot rescue inaccurate inputs. Lead times, units of measure, bundles, and receipts still need disciplined data.

A useful practice is to flag SKUs by risk: high stockout impact, high overstock cost, or high lead-time uncertainty. Manage each group differently instead of applying one blanket days-of-stock target across the catalog.

Strategy 3: Improve Slotting, Pick Paths, and Packaging

Before adding warehouse space, make the space and labor you already have more productive. Slotting means placing inventory according to how frequently, together, and easily products need to be picked.

Move fast-selling SKUs closer to packing stations and place commonly purchased combinations near each other where practical. Keep heavy items at safe, accessible heights. Separate visually similar products that are frequently mispicked. Review slotting periodically because the “A” products that drive most picks can change with seasonality or product launches.

Next, examine pick paths. If employees repeatedly cross the warehouse for common orders, reorganizing locations or using batch and zone picking may improve throughput without additional headcount. Batch picking works especially well when many orders contain the same popular SKU; zone picking can help larger warehouses divide work by area.

Too many box sizes can slow packers, while oversized packaging can increase dimensional shipping costs and damage risk. Standardize where possible, but keep exceptions for fragile, premium, or unusual products.

A hypothetical example illustrates the point: a store processing 800 orders on launch day may not need eight extra people if most orders contain the same two products. Pre-positioned inventory, simple batch picks, and ready-to-use packaging can remove much of the congestion before labor becomes the answer.

Increase Shipping Leverage With Strategies 4–6

Once inventory can move predictably inside the operation, the next opportunity is reducing repetitive decisions and improving how orders travel from the warehouse to the customer.

Strategy 4: Automate Repetitive Fulfillment Decisions

Automation creates leverage when it removes high-frequency decisions with clear rules. The best candidates are tasks that employees repeat hundreds of times and rarely require judgment.

Carrier service selection, order prioritization, package assignment, label generation, customer notifications, and exception tagging can often be rule-based. A shipping platform such as ShipStation can be useful when you need to centralize orders and automate shipping workflows across carriers and channels. Start with stable decisions.

For example: orders under a certain weight going to a defined region can use the lowest-cost service that meets your promised delivery window. High-value orders can require signature confirmation. Orders containing hazardous, oversized, or temperature-sensitive items should follow their own workflow where applicable.

Do not automate a broken process simply to make it run faster. First confirm the desired decision, inputs, exceptions, and fallback. Then automate the repeatable path while keeping unusual orders visible to a person.

Measure the result in minutes saved per order, reduced error rates, or fewer manual touches. Automation should remove work or improve consistency. If it only creates another dashboard the team must monitor, it has not meaningfully increased fulfillment capacity.

Strategy 5: Diversify Carriers and Route Orders by Service Need

A single-carrier setup feels simple, but it can become a growth risk when rates, capacity, regional performance, or peak-season conditions change. Begin by analyzing your shipment profile: destination zones, package weights, dimensions, residential concentration, delivery promises, and surcharge exposure. Then compare services for the order types that matter most. A lightweight parcel traveling within one region may have a different best option than a heavier cross-country order.

Build routing rules around customer promise first and price second. The cheapest label is not a saving if it routinely misses the expected delivery window and creates refunds or support costs. Likewise, paying for premium speed on every shipment can erode margin where customers would have been satisfied with standard delivery.

ALSO READ: How To Scale An Ecommerce Website Without Killing Your Profit Margins

Keep at least one operational fallback for critical lanes when practical. During peaks or disruptions, that flexibility can be more valuable than squeezing every shipment through the lowest theoretical rate.

From what I’ve seen, the most useful carrier strategy is rarely “always use Carrier A.” It is “use the lowest-cost reliable service that satisfies the promise for this specific order.” That turns shipping selection into a controlled economic decision rather than a habit.

Strategy 6: Distribute Inventory Only Where Demand Justifies It

Placing stock closer to customers can shorten shipping distance and improve delivery speed, but a multi-node network introduces new inventory and replenishment complexity. It should be earned by demand density.

Start by mapping order concentration by region. If a substantial share of orders consistently comes from two or three geographic areas, test whether a second fulfillment node meaningfully changes shipping zones, transit times, and costs. A provider such as ShipBob illustrates the outsourced model: merchants can use a distributed fulfillment network and place inventory in multiple facilities. The strategic question is not whether multiple warehouses are available; it is whether your economics support splitting stock.

Model the full cost. Include inbound freight to each node, storage, transfers, receiving, extra safety stock, and the risk of having the wrong SKU in the wrong location. Compare that with expected outbound savings and service improvement.

A sensible first test is often to distribute only the fastest-moving products while keeping long-tail inventory centralized. This reduces the amount of stock you fragment across the network and gives you cleaner data on whether regional placement actually improves margin and customer experience.

More fulfillment locations create proximity, but they also multiply inventory decisions. Add nodes because the numbers support them, not because a larger network looks more advanced.

Add Partners and Systems With Strategies 7–8

Growth eventually forces a choice between expanding internal operations and using specialized partners. The right answer depends on your economics, operational strengths, product requirements, and need for flexibility.

Strategy 7: Use Clear Triggers for Moving to a 3PL

A third-party logistics provider, or 3PL, stores inventory and handles activities such as picking, packing, and shipping on your behalf. Outsourcing can remove warehouse constraints, but it also changes your cost structure and operating control.

Create objective triggers before starting a search. Useful triggers include persistent space limitations, inability to maintain shipping cutoffs, excessive fulfillment hiring, difficulty adding regions, rising management time spent on warehouse issues, or a need for capabilities your current operation cannot economically build.

Then compare total cost, not only pick-and-pack fees. Ask about receiving, storage, packaging, account management, minimums, returns, projects, peak surcharges, inventory transfers, and exception handling. Evaluate technology integrations, accuracy standards, support model, carrier options, geographic coverage, and how quickly inventory becomes available after inbound delivery.

A straightforward apparel brand may have different needs from a heavy-goods seller or a business requiring lot tracking, kitting, serialization, or custom packaging. Providers such as ShipBob, ShipMonk, ShipHero, and Red Stag Fulfillment serve different operational profiles, so shortlist based on requirements rather than name recognition.

Run a cost model at current volume, expected growth volume, and peak volume. The best partner should improve your ability to scale without making unit economics or exception management opaque.

Strategy 8: Create One Reliable Source of Inventory and Order Truth

As sales channels multiply, system disagreement becomes one of the most damaging scaling problems. If the storefront says 14 units are available while the warehouse believes there are eight, growth only accelerates overselling and manual reconciliation.

Decide which system is authoritative for each type of data. One platform might own product information, another inventory, another shipping status, and another financial records. What matters is that each field has a clear source of truth and a documented synchronization path.

Map how orders, cancellations, returns, bundles, purchase orders, and stock adjustments move between systems. Pay special attention to timing. “Integrated” does not always mean instantly synchronized, and delays can matter during flash sales or low-stock periods.

Use exception reporting instead of expecting employees to compare systems manually. Examples include orders that failed to import, inventory that went negative, shipments without tracking, or returns received without a matching authorization. Give each exception an owner and resolution deadline.

Avoid adding software simply because volume is growing. Add technology when it replaces a known manual process, improves visibility, or enables a capability the business needs.

The result should be a cleaner operating model in which people manage exceptions rather than constantly repairing basic data flow.

Protect Customer Experience With Strategies 9–10

Fulfillment does not end when the parcel leaves the warehouse. Returns, delivery exceptions, and demand spikes are part of the same operating system and should be designed before they become emergencies.

Strategy 9: Make Returns a Structured Fulfillment Workflow

A scalable process decides what happens before the first peak-season pile of boxes arrives.

Define return reasons and disposition rules. When a returned product reaches the warehouse, should it go back to sellable inventory, be inspected, be refurbished, be quarantined, or be written off? Make those decisions consistent so employees do not improvise case by case.

Speed matters because a delayed return can become trapped working capital. A sellable item sitting unprocessed is inventory you own but cannot sell. At the same time, returning damaged goods to available stock can create a second customer problem, so quality checks need clear standards.

Tools such as Loop Returns can support structured ecommerce returns, while platforms such as AfterShip can support post-purchase and shipment visibility workflows. Use software only after your eligibility, routing, refund, exchange, and disposition policies are clear.

Track return rate by SKU and reason. If one product generates unusually high “too small,” “damaged,” or “not as expected” returns, fulfillment should feed that information back to merchandising, packaging, product content, or quality control. Returns are not merely a reverse-logistics cost; they are an operational feedback loop.

Strategy 10: Build Peak Capacity and Exception Playbooks Before You Need Them

Your normal operating day should not determine your peak-season plan. Promotions, holidays, launches, influencer spikes, and wholesale orders can compress several ordinary days of demand into a much shorter window.

Build three demand scenarios: expected peak, strong upside, and extreme but plausible surge. For each scenario, calculate orders per day, units per order, receiving workload, labor hours, packing-station capacity, carrier handoff requirements, packaging supply, and inventory availability.

Then define what changes at each threshold. You might add temporary shifts, pre-build kits, pause nonessential projects, extend carrier pickups, simplify gift options, or reserve overflow storage. If you use a 3PL, confirm cutoff calendars, receiving deadlines, capacity expectations, and peak procedures well before the event.

Create playbooks for the exceptions that create the most customer pain: missed carrier scans, oversold products, delayed inbound inventory, lost parcels, damaged shipments, address errors, and system outages. Each playbook should specify who owns the issue, what information they need, what customer communication is appropriate, and when escalation occurs.

It is a series of predetermined decisions. When demand accelerates, the team can execute those decisions instead of debating them under pressure.

Troubleshoot the Scaling Mistakes That Erode Margin

Even a strong growth plan can become expensive if complexity is added in the wrong order. These mistakes are common because each one can look like progress until the hidden operational cost appears.

ALSO READ: Ecommerce Fulfillment For Startup Brands: What To Set Up First

Avoid Scaling Complexity Before You Scale Process Discipline

Opening more warehouses, adding more carriers, adopting a new warehouse management system, or outsourcing to multiple partners can all be valid moves. They are poor substitutes for process discipline.

If inventory counts are unreliable in one location, spreading stock across three locations makes reconciliation harder. If your team does not use consistent SKU naming, a new integration can synchronize bad data faster. If returns lack disposition rules, outsourcing them simply moves the confusion somewhere else.

Standardize first. Document receiving, put-away, cycle counting, picking, packing, shipping, and return procedures at the level needed for consistent execution. Establish who can change product data, inventory adjustments, automation rules, and service promises.

Then look for the smallest structural change that solves the growth constraint. You may need a new packing station rather than a new warehouse. You may need better carrier rules rather than a second shipping platform. You may need a cycle-count routine rather than more safety stock.

Every node, vendor, workflow, and exception path has to be monitored. Add it only when it creates enough capacity, resilience, speed, or margin improvement to justify that ongoing burden.

Diagnose Late Orders and Inventory Mismatches at the Source

When late orders rise, separate processing delays from transit delays. Processing delay happens before the carrier receives the parcel; transit delay happens afterward. For processing delays, examine order release rules, pick queues, staffing, replenishment from reserve storage, packing capacity, system latency, and carrier cutoff timing. Measure where the order waits longest. A three-hour pick process may not be the problem if the order sat unallocated for 14 hours first.

For transit delays, segment by carrier service, origin, destination region, package type, and day of week. One weak lane can distort an otherwise reliable operation. Avoid changing the entire carrier strategy before identifying the specific pattern.

Inventory mismatches require similar discipline. Trace whether the variance comes from receiving, mispicks, damages, unrecorded adjustments, bundle logic, returns, theft, or synchronization delays. Cycle count the affected locations and compare physical units with the system record.

Do not “fix” chronic variance by adding inventory buffers alone. That hides the symptom while tying up more cash. Correct the transaction or process that causes inventory to become inaccurate, then use safety stock for real demand and lead-time uncertainty.

Balance Faster Delivery Against Contribution Margin

Fast delivery can improve the customer experience, but speed should be purchased deliberately. Calculate contribution margin after fulfillment, outbound shipping, payment costs, expected returns, and other variable order costs. Then examine how shipping upgrades affect that number. A premium service may be sensible for high-value orders or customers paying for expedited delivery but uneconomic as the default for low-margin products.

Use segmentation. Offer free standard shipping above an order threshold if the economics support it. Charge for expedited options. Consider slower services for remote destinations when the delivery estimate remains acceptable. Place high-volume inventory closer to customers only when the network savings offset added inventory complexity.

Be careful with broad promises such as “two-day shipping” when your process actually requires a full business day before carrier handoff. Delivery expectation is the combined result of order processing and transit time.

The strongest fulfillment growth strategy is not the one with the fastest headline. It is the one that delivers the experience you promised at a repeatable cost the business can afford.

Measure, Optimize, and Scale What Works

Once the operation is stable, improvement becomes a measurement problem. A small set of connected metrics helps you see whether growth is creating leverage or simply adding volume and cost.

Build a Fulfillment KPI Dashboard Around Decisions

A useful dashboard should lead to action, not simply display numbers. Start with metrics that connect customer experience, cost, inventory, and capacity.

Track order cycle time, on-time shipment rate, order accuracy, cost per order, shipping cost as a percentage of revenue, inventory accuracy, stockout rate, return rate, days of inventory on hand, and orders processed per labor hour where relevant.

Segment these KPIs by warehouse, channel, SKU family, carrier, shipping service, or destination when the averages hide meaningful differences. For example, overall on-time shipping may look healthy while one warehouse misses cutoffs on Mondays. Average return rate may look normal while one size or product color has a serious issue.

Set alert thresholds around operational decisions. A stockout forecast might trigger purchasing review. A rise in processing time might trigger staffing or slotting analysis. A carrier-service deterioration might trigger routing changes.

Review leading indicators as well as lagging outcomes. Inventory cover, inbound delays, backlog, and forecast error can warn you before customers experience the problem. That is what makes a fulfillment dashboard useful for growth: it helps you intervene before volume turns a manageable issue into a service failure.

Test Operational Changes Like Business Experiments

Fulfillment teams often make several changes at once, which makes it difficult to know what actually improved performance. Treat major changes as controlled business experiments whenever practical.

Start with a clear hypothesis. For example: moving the top 20 SKUs closer to packing stations will reduce average pick time without increasing mispicks. Establish the baseline, change one main variable, and measure performance over a representative period.

The same method works for package sizing, carrier routing, cutoff times, staffing patterns, automation rules, and inventory placement. For larger changes such as adding a second fulfillment node, pilot a limited SKU set or region before moving the full catalog.

Watch for downstream effects. A faster pick process that increases packing errors is not a successful optimization. Lower shipping cost that increases delivery complaints may simply move cost from logistics to customer support and refunds.

Document what changed, why it changed, what the result was, and whether the new process becomes standard. This creates institutional knowledge and prevents the team from repeating old experiments.

Add Locations, Channels, and International Reach in the Right Order

Each new warehouse, marketplace, wholesale relationship, or country adds order flows, inventory commitments, service rules, and exception cases.

When considering another fulfillment location, review regional order density, SKU velocity, shipping-zone economics, replenishment requirements, and the amount of duplicated safety stock the network will require. Add the location when the expected cost or service improvement is meaningful and the inventory system can manage it reliably.

For new sales channels, confirm inventory synchronization, order import, cancellations, tax handling, shipping expectations, and returns before launch. A channel that creates revenue but also creates overselling and manual reconciliation may not be operationally ready to scale.

International expansion adds customs, duties, taxes, restricted-product rules, longer replenishment paths, and country-specific customer expectations. Start with markets where demand is already visible, then choose whether cross-border shipping or local inventory makes sense based on volume and economics.

The same principle applies everywhere: scale complexity after the underlying process is repeatable. Growth becomes more durable when each new layer is supported by data, clear ownership, and enough margin to absorb inevitable exceptions.

Choose the Next Fulfillment Upgrade With Confidence

The best ecommerce fulfillment growth strategies do not begin with a bigger warehouse or a longer software stack. They begin with visibility into where orders slow down, where inventory becomes unreliable, and where cost increases faster than revenue.

Build the baseline first. Improve forecasting, replenishment, slotting, packaging, automation, carrier routing, and exception handling before adding unnecessary network complexity. When demand justifies a 3PL or distributed inventory, evaluate the full economic impact rather than focusing on one attractive rate or delivery promise.

Your next step is simple: identify the single constraint most likely to limit the next stage of growth, choose the strategy in this guide that addresses it, and define the metric that will prove whether the change worked. Then scale the process that performs—not the complexity that merely looks like growth.

Juxhin

I’m Juxhin, the voice behind The Justifiable.
I’ve spent 6+ years building blogs, managing affiliate campaigns, and testing the messy world of online business. Here, I cut the fluff and share the strategies that actually move the needle — so you can build income that’s sustainable, not speculative.