What "greenhouse operations software" should cover

• Planting and harvest planning anchored to confirmed sales
• FIFO lot-level inventory with atomic write operations
• Mobile-first daily worker tasks with carry-over rules
• Real-time greenhouse map with maturity, disease, and spray-lockout overlays
• Waste, productivity, and planting-success reporting
• Sales order to FIFO harvest fulfillment automation
• Role-based access (worker, manager, sales, owner)

The 10 questions to ask every vendor

1. Does the data model treat the gutter as the unit of inventory, or just the variety?
2. Are inventory writes atomic when two workers harvest the same line simultaneously?
3. What is the worker UX — phone-first, or a desktop dashboard with a mobile afterthought?
4. How are unfinished tasks handled at end-of-day? Auto-carryover or manual rebuild?
5. Can sales orders auto-generate FIFO harvest tasks, or does someone manually allocate?
6. How is Pre-Harvest Interval (PHI) enforced on sprayed lots — code, or honor system?
7. Does waste tracking require a reason, and can you report waste per customer per line per variety?
8. What languages does the worker app support natively?
9. How is access controlled — single shared login per role, or per-user accounts?
10. What does data export look like — full CSV per report, or PDF-only?

Red flags

• "It works for any farm" — generic ERPs that haven't been domain-modelled for greenhouses.
• No mention of lot codes or FIFO — that's a spreadsheet with extra steps.
• Per-worker pricing — punishes you for scaling exactly when you need the system most.
• No free trial — software you can't pilot on your real data is software you can't evaluate.
• Worker app requires app-store install — guarantees IT friction at every new hire.

Where GutterTrack fits

GutterTrack is purpose-built for moving-gutter hydroponic lettuce and herb production. It covers all seven scope items above, scores well on all ten vendor questions, and ships with English and Slovenian worker UX out of the box. It's not the right fit for soil farms, static-bench greenhouses, or non-leafy crops — but if you're growing lettuce in moving gutters, run a free trial on your own data and see for yourself.

The operating ranges

• Seedling / first 7–10 days: EC 0.8–1.2 mS/cm, pH 5.8–6.2.
• Vegetative / week 2–4: EC 1.2–1.8 mS/cm, pH 5.6–6.0.
• Pre-harvest / final 5–7 days: EC 1.6–2.2 mS/cm (slightly higher pushes weight); pH 5.6–6.0.

These are leaf-lettuce ranges; butterhead trends lower, romaine slightly higher. Herbs vary widely — basil tolerates EC up to 2.5; cilantro prefers under 1.8.

Why pH drifts (and what it tells you)

Healthy plants pull cations from the solution, which pushes pH up over the day. A pH that's stable or drifting down usually means the plants aren't feeding well — root health, oxygen, or temperature problem. The drift direction matters more than the absolute number.

Daily check protocol

Measure EC and pH at the same time of day, same point in the loop, with the same probe. Log it. A spreadsheet works at single-greenhouse scale; integrate it into your ops software once you're managing multiple lines or shifts.

When to dump and refill

When pH refuses to hold its set point even after acid/base dosing, or EC creeps up despite adding water — accumulated salts are throwing off micronutrient ratios. Full dump and refill, every 2–4 weeks depending on water source and crop intensity.

The trap to avoid

Treating EC as a feeding amount. EC measures total dissolved salts — it tells you nothing about ratios. A solution at the right EC with wrong nutrient ratios produces stunted, hollow-stemmed lettuce that hits weight target but fails quality. Test full nutrient profile monthly, not just EC.

The formula

Labor cost / kg = (all-in labor cost for the period) / (kg of saleable lettuce shipped in the period). All-in means wages, taxes, equipment, training, and management overhead allocated to production. Saleable means net of waste and quality rejection — gross harvest weight overstates productivity.

What good looks like

Well-run moving-gutter operations target €1.80–€2.80/kg in EU labor markets, €2.20–€3.50/kg in higher-wage zones. Below those bands usually means under-counting management overhead; above means real operational drag.

The four operational levers that actually move the number

• Walk-pattern optimization. Sorting tasks by physical line — not by variety — cuts walking time by 20–40% in most greenhouses. Workers harvest one line, then move on.
• Surplus tracking per worker per order. Visible over-harvest shrinks within days. Make it ungame-able by anchoring it to confirmed pieces, not "how many we cut".
• Carry-over discipline. Unfinished Mon–Thu tasks roll forward; Friday tasks don't. Stops the "weekend overtime to catch up" trap.
• FIFO at the lot level. Workers never have to decide which lot to cut first — the system always points them to the oldest mature one.

What to measure weekly

Labor minutes per harvested gutter, per worker. Compare not to absolute targets but to the worker's own 4-week rolling average. Trending up means training, equipment, or assignment — not a discipline issue.

1. Heads harvested per m² per week

The single number that reflects everything else — cycle time, hole density, harvest discipline. Track weekly, compare against the rolling 8-week average, investigate any 10%+ drop within 48 hours.

2. Sales coverage ratio (Covered / Immature / Short)

For the coming 7 days of confirmed orders: what percent is harvestable today (Covered), in the maturity window soon (Immature), and not coverable from current inventory (Short)? Short above 5% means a re-plan now.

3. Average growing days vs target

Per variety per season. A 3-day drift on a 35-day cycle is an 8% yield difference annualized — investigate sooner rather than later.

4. Waste percentage by source

Over-harvest, bolted, quality reject, wrong-variety. Break it down — the aggregate "waste %" is useless without the source attribution.

5. Labor minutes per harvested gutter

Derived from real task-completion timestamps. The metric that drives most P&L decisions and that nobody tracks honestly without software.

6. Planting success rate

Plants delivered to the line vs plants planned. Green between 90–110%; red outside that band. Most growers discover their distribution is systematically off-target.

7. Spray PHI compliance

Should be 100%. Anything less is a recall risk. Hard-code it in software, don't track it on a clipboard.

8. Order fulfillment accuracy

Pieces shipped vs pieces ordered, per customer, per week. The number sales actually cares about — and the proof of whether your production loops are tight.

Review cadence

Monday morning, 20 minutes, KPIs 1–8 in that order. Anything that moved a band gets a named owner and a follow-up date. Anything stable gets skipped. That's the meeting.

NFT (Nutrient Film Technique)

A thin film of nutrient solution flows through gutters housing the root mat. Moving-gutter NFT is the modern variant: gutters shift along the line as plants grow, opening space between mature heads. Higher hole density, faster turnover, lower per-head water mass.

• Strengths: Highest yield per m² (~50–80 heads/m²/cycle). Tight harvest cadence. Lower water volume to manage.
• Weaknesses: Pump-failure risk — a stalled film dries roots in hours. Capex higher than DWC for the same footprint.

DWC (Deep Water Culture / raft systems)

Plants float on rafts above a tank of oxygenated nutrient solution. Slower to harvest, more thermal mass, more forgiving of equipment failures.

• Strengths: Equipment-failure tolerance (large water volume buffers temperature and oxygen). Simpler plumbing.
• Weaknesses: Lower yield per m² (~30–50 heads/m²/cycle). Heavier infrastructure. Algae and pathogen management is its own discipline.

The decision matrix

If you're chasing tight harvest schedules for restaurant or retail contracts, moving-gutter NFT is the right answer almost every time — provided you invest in pump redundancy and a real ops layer that catches lot-level drift. If you're at smaller scale, have a single operator, and prioritize equipment simplicity, DWC is forgiving in ways NFT is not.

What software changes

Moving-gutter NFT lives or dies by its operations data — lot codes, FIFO discipline, daily worker tasks. DWC tolerates spreadsheets longer because the cadence is slower. By 5,000 heads/week either system needs purpose-built software; below 1,500/week DWC growers often get by with paper plus a calendar.

The target ranges

• Day temperature: 18–22 °C for most leaf lettuce; herbs trend warmer (22–26 °C for basil).
• Night temperature: 12–16 °C — the day/night swing matters as much as the absolute number.
• Relative humidity: 60–75% in the canopy. Above 85% invites downy mildew and tip burn.
• CO₂: 800–1,200 ppm during photoperiod. Below 400 ppm caps photosynthesis hard.
• DLI (daily light integral): 14–17 mol/m²/day for lettuce; supplement in winter.

Daily checks

Walk the greenhouse mid-morning and mid-afternoon. Look for condensation on plastic (humidity too high), wilting at midday (light or temperature stress), and stratified airflow dead zones. Five minutes a day catches problems a sensor dashboard misses.

Weekly checks

Calibrate one sensor per zone against a handheld reference. Sensor drift is the silent killer — a humidity probe reading 8% low for three weeks is how disease outbreaks start.

Seasonal pivots

Winter: shorten day temperature ramps, raise CO₂ ceilings, add supplemental light to hit DLI. Summer: shade nets, evaporative cooling, harvest earlier in the day to protect quality. Track growing-day variance per variety per season — that's the data that proves climate changes are working.

The integration question

Most climate-control systems (Priva, Hortimax) handle environment well but can't see your harvest plan. The fix isn't ripping them out — it's making sure your operations software (like GutterTrack) consumes the same lot data so a sprayed lot or a slow-maturing line gets flagged before it costs you a harvest cycle.

The four sources of harvest waste

• Over-harvesting to fulfil orders. A 400-head order gets 460 cut "to be safe" — the surplus 60 sit until they bolt.
• Bolted lots. A line passes 100% maturity and gets harvested two days late. Heads are unsellable.
• Wrong-variety planting. Sales forecast says romaine, the planner sows butterhead. Three weeks later you have inventory nobody ordered.
• Quality rejection. Disease pressure or spray-interval errors push a whole lot below grade.

The four fixes that recover most of it

• Track over-harvest in pieces and percent. Make the surplus visible per order, per worker, per day. What gets measured shrinks.
• Maturity-based harvest triggers. The 85% rule, applied per lot, kills bolted-lot losses almost entirely.
• Anchor planting plans to confirmed orders. Forecast layers go on top of confirmed sales — never in place of them.
• Pre-Harvest Interval (PHI) lockouts. A sprayed lot cannot be harvested until its PHI clears. Hard-coded, not memorised.

The reporting habit that compounds

A weekly waste report broken down by customer, line, greenhouse, and variety surfaces the one or two recurring losses that account for most of the total. Fix those, re-measure, repeat. Most operations halve their waste within a quarter once the data is visible.

The three problems with the WhatsApp model

• No source of truth. When the harvest plan changes at 7am, half the team is already mid-task on the old version.
• No completion signal. A manager has no idea which gutters have actually been cut until someone replies in the chat.
• No carry-over logic. Tasks not finished today silently disappear. Tomorrow they have to be re-created from memory.

What a good worker task system looks like

One screen per worker, sorted by physical line so they walk the greenhouse once. Each task shows the lot, the gutter count, and the action — harvest, plant, move, or spray. Completing a task is one tap, with a confirmation dialog so accidental completions don't corrupt inventory.

The carry-over rule that matters

Mon–Thu, any task not completed by end of day rolls forward to the next day. Friday tasks don't carry into Saturday — that prevents weekend surprises. This single rule eliminates almost all "we forgot to harvest line 7" incidents.

Why mobile-first beats a desk dashboard

Workers don't sit at desks. The phone in their pocket is the only interface that survives a full shift in the greenhouse. Anything that requires logging into a laptop will get bypassed within a week.

The benchmark range

For a well-run moving-gutter system growing leaf lettuce, expect 50–80 heads per m² per cycle, with 8–11 cycles per year. That puts annual yield in the 450–800 heads/m² range. Romaine and butterhead trend lower per cycle but heavier per head; loose-leaf trends higher but lighter.

The four levers that actually move the number

• Hole density per gutter. Going from 160 to 240 holes per gutter is a 50% yield jump — but only if the variety tolerates the spacing without losing head weight.
• Cycle time. Shaving 3 days off a 35-day cycle adds nearly a full extra cycle per year. Climate control and seedling vigour matter more here than the variety choice.
• Harvest discipline. Letting lots sit two days past 100% maturity quietly destroys yield — bolted heads are a write-off.
• Empty-line time. The hours between one lot's harvest and the next planting are pure waste. Sub-24-hour turnaround is the target.

Why measuring weekly beats measuring annually

Annual yield is a vanity number. Weekly heads-harvested per m² is the operational metric — it tells you within days when something has slipped, before three months of underperformance compound.

What to track

Heads harvested, gutters cycled, average days from sow to harvest, and the percentage of harvests that fell inside the 85–105% maturity window. Those four numbers explain almost every yield gap we have ever investigated.

The lot is the unit, not the gutter

A lot is everything sown on the same day, on the same line, of the same variety. Track lots, not individual gutters, and FIFO becomes trivial: when an order comes in, the system pulls from the oldest lot first.

Lot codes that workers can read

A good lot code tells a worker everything they need at a glance. We use {Greenhouse}-{Line}-{Variety}-{MONday}. So GH1-L4-Romaine-MAY12 is unambiguous on a printed pick list.

What FIFO prevents

• Over-mature waste. Yesterday's lot ships before today's gets cut.
• Phantom inventory. Sales sees real, age-aware stock — not a flat "you have 800 heads".
• Dispute tracing. If a customer reports quality issues, you know exactly which lot, which line, which day.

Where it gets hard

Manual harvest entries that don't decrement the right lot. That is the single biggest source of inventory drift, and the one place where atomic database operations matter — every harvest must reduce one specific lot, not "the variety in general".

Step 1 — Anchor the plan to confirmed sales

Start every weekly plan with the orders you have already accepted. These are non-negotiable: every gutter dedicated to them must be in the right maturity window on the right day.

Step 2 — Layer in forecast demand

On top of confirmed sales, add the rolling 4-week average for each variety. This is the buffer that catches walk-in orders and last-minute upsells.

Step 3 — Translate pieces to gutters

Customers buy heads. Greenhouses produce gutters. The conversion is variety-specific (holes per gutter × expected yield per hole). Keep this stored per variety, not per planting — that's what makes the math repeatable.

Step 4 — Distribute across lines

Allocate gutters to lines proportionally to their hole count. A 240-hole line gets 1.5× the gutters of a 160-hole line for the same variety.

Step 5 — Build the sowing schedule backwards

From the harvest date, subtract the variety's growing days (seasonal — winter is slower). That's your sowing day. Slot it into the daily worker tasks.

Step 6 — Track maturity, not just dates

A harvest "due" today might be only 75% mature in February. The 85% rule is a far better trigger than the calendar.

What to automate

Steps 3, 4, and 5 should never be done by hand. Steps 1 and 2 are judgement calls — software should surface the data, you make the call.

What "moving-gutter" actually means for software

In a moving-gutter system every gutter has a position, a variety, a sowing date, and a harvest window — and all of those change daily. A static spreadsheet can record any one of them. It cannot keep them in sync without someone manually re-typing the entire grid every morning.

The three failures we see most often

• Lost lots. A worker harvests a gutter early, nobody updates the sheet, the next planting is scheduled on top of plants that were already cut.
• Phantom inventory. Sales promises 400 heads of romaine for Friday. The sheet says you have them. The greenhouse says half are still 4 days from maturity.
• Worker confusion. Tasks live in WhatsApp messages and printed lists. Half the team works yesterday's plan.

What purpose-built greenhouse management software does differently

A real system tracks gutters, not rows in a sheet. Every harvest, planting, or move updates the same source of truth — the dashboard, the worker's phone, the sales view, and the manager's planning screen all read from it.

That is the model GutterTrack is built on. Plant a tray and the line's gutter count updates. Harvest a lot and inventory drops. Promise an order and the system tells you whether it is covered, immature, or short — before the customer is told yes.

When to switch

If you find yourself reconciling the same numbers twice a week, or workers are asking "is this still the latest version" — your spreadsheet has already broken. The cost is just hidden in the time you spend patching it.