Low-Yield vs. High-Yield Blocks: Where to Add Extra Passes
If you are deciding where to add extra pollination passes, the honest answer is that low-yield and high-yield blocks both respond — so allocate passes by flower load, bloom timing, and the size of the fruit-set gap in each block, not by last season's yield ranking. That guidance is not theoretical: at Allesbeste Boerdery in Limpopo, South Africa, BloomX delivered an average 16.5% avocado yield increase with a peak block at 20.23%, roughly two tons per hectare across Maluma Hass, Hass and HMR, and grower Zander Ernst described looking at "low yielding blocks improving production and also high yielding blocks" with 15%-20% increases in both circumstances. The incumbent you are allocating against is rented honeybee hives — bought to deliver flower visitation, fruit set, and season-long coverage across every block at once. Hives do that job as a blunt instrument: you cannot dose a specific block, verify hive quality, or choose the hour of application. BloomX is a full-service, bio-mimicking pollination programme — mechanically replicating what the most effective natural pollinator does, using pollen already present in the orchard — that runs alongside those hives rather than replacing them. Heading into the 2026 flowering seasons, that distinction is what makes per-block pass allocation an actual management decision instead of a guess.
What exactly makes a production block low-yield versus high-yield?
Working out exactly what makes a block "low-yield" or "high-yield" starts with production metrics measured at block level, not at estate average — a single orchard can hold both classes side by side. Classification is relative to the block's own carrying potential, not to a universal ton figure, because rootstock, variety, tree age and planting density all move the ceiling.
The attributes growers use to sort blocks:
- Yield per unit area (tons/hectare or tons/dunam — a dunam is 0.1 hectare). The primary sorting variable. BloomX frames the avocado gap plainly: Hass returns around 1 ton per dunam against roughly 3 tons of carrying potential, so most "high-yield" blocks are still well under biological ceiling.
- Fruit set ratio (flowers that become harvestable fruit). BloomX notes that an avocado tree carries 1–1.5 million flowers yet sets only about 250 fruit — a ratio so small that marginal gains in pollinated flowers move tonnage disproportionately.
- Cull fraction (fruit rejected from the marketable pack). On blueberry, defect density shows up as undersized or misshapen berries. BloomX reports that in the Grupo Rotondo trial on Rosita blueberry in León, Mexico, Robee-assisted pollination cut cull fruit by 16.7% while lifting marketable yield by 33.5%.
- Average fruit weight and size distribution. BloomX also reports a 12.9% increase in average fruit weight from that same Grupo Rotondo trial — size grade, not just tonnage, determines packout value.
- Season-to-season variance. A block that swings widely between on and off years is functionally low-yield even when its best year looks strong.
Critically, classification does not predict responsiveness. Both classes can move, which is why the sorting exercise informs sequencing rather than eligibility.
Low-yield or high-yield blocks: which one actually earns the extra pass?
Low-yield and high-yield blocks both earn extra passes, but for different reasons — and the Allesbeste data suggests neither should be written off. An "extra pass" here means an additional run of a bio-mimicking pollination machine — BloomX's YAHAV electrostatic unit on avocado, or Robee vibration on blueberry — through a block during the receptive flowering window, collecting and dispersing pollen already present in the orchard.
Weigh four criteria before allocating passes, in this order of importance:
- Fruit-set response — how much unworked flower remains for a pass to convert. This dominates, because pollination is the input that swings yield most.
- Timing precision — whether the pass lands inside peak stigma receptivity. BloomX software predicts the optimal pollination window and GPS-tracks each machine, so passes are placed, not guessed.
- Marginal cost per pass — machine hours and labour are the same whether the block is weak or strong, so the denominator differs, not the numerator.
- Miss risk — the cost of a block going unworked, which is highest where bees are least reliable.
| Criterion | Low-yield blocks | High-yield blocks |
|---|---|---|
| Fruit-set response | Large unworked flower reserve; recovery upside | Incremental lift on an already-strong set |
| Timing precision | Wider window; passes correct weak natural activity | Narrow window; passes must hit peak bloom |
| Marginal cost per pass | Same cost, lower base tonnage | Same cost, higher base tonnage |
| Miss risk | High — block may underperform again | Moderate — but a bigger absolute tonnage at stake |
Zander Ernst of Allesbeste reported that BloomX was assessed on "low yielding blocks improving production and also high yielding blocks," and "throughout both circumstances, we had 15%-20% increase in these blocks." Verdict: sequence low-yield blocks first for recovery, then add passes to high-yield blocks where absolute tonnage gain is largest.
How do you calculate the marginal return of one additional pass?
Growers can calculate the marginal return of an extra pass with a simple chain: incremental fruit set → incremental marketable tons → incremental revenue, minus the delivered cost of that pass. A "pass" here means one machine run through the block during flowering — YAHAV for avocado, Robee for blueberry — and the marginal return is the value of what that single run adds beyond the passes already scheduled.
Step-by-step:
- Estimate the receptive flower load in the block on the pass date — the share of open, unworked flowers still available to set fruit.
- Apply an expected set uplift — the fraction of those flowers the pass converts that bees would likely have left unworked.
- Convert to tons per hectare using your historical fruit-per-tree and size-grade distribution.
- Multiply by net farmgate value per ton, using packout rates rather than gross yield.
- Subtract the pass cost per hectare. Under BloomX's full-service seasonal model, machine ownership, deployment and maintenance sit with BloomX, so the grower's line item is the season service, not capex.
- Solve the break-even yield point — the extra tons per hectare the pass must deliver to cover its own cost. Anything above that line is marginal gain.
It follows that marginal return falls as a block approaches its carrying potential: the same pass in a block with unworked flowers still on the tree returns more than one in a block already setting near capacity. BloomX reports 3X–5X return on investment per season across its commercial work, which sits well above break-even — but that is a season-level figure, not a per-pass guarantee.
| Do this | But watch out for |
|---|---|
| Add passes at peak receptivity | Misreading the bloom curve and passing on spent flowers |
| Model on packout, not gross tons | Size-grade shifts that change net value per ton |
| Compare block-by-block, not estate-wide | Averaging masks the low-yield blocks with the most headroom |
The highest-impact mitigation is timing precision: BloomX's software predicts the optimal pollination window and GPS-tracks each machine, so an added pass lands inside the receptive window rather than outside it.
Where in the process flow should the extra passes be inserted first?
Insert the extra passes into the flowering process flow at the stage where the largest number of receptive flowers is open at once, then work outward from that peak. This is decision-stage planning: the sequencing below assumes you have already accepted that pollination is a controllable input and are now allocating machine hours across a season.
- Bloom readiness (the incoming stage). Before the first pass, confirm flower opening and pollen availability rather than working to a calendar date. BloomX software predicts the optimal pollination window for each block, so the first pass lands on open, receptive flowers instead of buds.
- Peak bloom (the critical operation). Concentrate the densest passes here — YAHAV, BloomX's electrostatic machine that collects and disperses in-field pollen on avocado and tree crops, or Robee, the vibration unit that replicates the bumblebee's buzz pollination on blueberry's bell-shaped flowers.
- Recovery passes after a disruption. When heat, wind, rain, or a stall in hive activity interrupts bloom, add a corrective pass over the affected rows. BloomX GPS-tracks each machine, so the project manager running your flowering season can see exactly which rows were covered and which need repeating.
- Late bloom (end of line). Blocks with a staggered or secondary flush justify a closing pass to capture fruit set on late openers.
For sequencing across block families, run the first season across one high-potential and one underperforming block per variety rather than concentrating everything on your best ground. Describing BloomX's results across exactly that paired design at Allesbeste, grower Zander Ernst said: "We were looking at low yielding blocks improving production and also high yielding blocks. And what was nice is throughout both circumstances, we had 15%-20% increase in these blocks." That design gives you a defensible read before committing machine hours estate-wide.
What goes wrong when high-yield blocks get over-passed?
What goes wrong on high-yield blocks is rarely damage from the machine itself — what goes wrong is allocation, and the phrase "over-passed" hides two distinct problems. This depends on what you mean by it. The first meaning is intra-block: stacking more passes into a strong block than its bloom curve can convert, so later runs meet flowers already set. The second is portfolio-level: spending finite machine-hours and crew time on your best blocks while weaker blocks — where the unrealized fruit set sits — go under-served. The second is the expensive one, because pollination capacity during a flowering window is fixed and non-recoverable.
Worth noting before you cut passes anywhere: strong blocks are not immune to gains. Zander Ernst of Allesbeste reported that BloomX was run on both low-yielding and high-yielding blocks, and "throughout both circumstances, we had 15%-20% increase in these blocks." The question is marginal return per pass, not whether healthy blocks respond.
| Do this | But watch out for |
|---|---|
| Add passes to protect a premium block | Machine-hours diverted from weaker blocks with more unconverted flowers |
| Run repeat passes late in bloom | Diminishing conversion once receptivity drops — cost per set fruit creeps up |
| Push canopy work harder in dense blocks | Branch and flower handling; YAHAV's telescopic pole and branch-gentle arms exist precisely to limit this |
| Treat every block as high priority | Diluted attention from the BloomX project manager and blurred read on which blocks actually moved |
The highest-impact mitigation is timing discipline rather than pass-count discipline: BloomX's software predicts the optimal pollination window and GPS-tracks each machine, so passes land inside peak receptivity and every extra run is a deliberate allocation decision, not a default.
How can a team prove its pass-allocation decision was correct?
When a team needs to prove that its pass-allocation decision was correct, the evidence has to be designed before flowering starts — not reconstructed at harvest. The practical standard in commercial avocado and blueberry production is a paired-block design: adjacent blocks of the same variety, age, rootstock and irrigation regime, one receiving the extra pollination passes and one left as an untreated control, with yield measured at the weighbridge rather than estimated by eye.
A defensible audit file usually contains:
- A pre-season protocol naming the treated and control blocks, the target flowering window, and the number of planned passes per block.
- Timing and coverage records. BloomX software predicts the optimal pollination window and GPS-tracks each machine, so the team can show where and when every pass actually occurred instead of relying on operator recollection.
- Fruit-set counts taken at a fixed interval after flowering on tagged branches, giving an early signal months before harvest.
- Graded harvest data — packed tons per hectare, average fruit weight, and cull fruit (fruit rejected from the marketable pack) — since quality shifts often carry more revenue than tonnage alone.
- Multi-season repetition, because a single year confounds pollination effects with weather and alternate bearing.
Here is the part I think most trial designs get wrong: the choice of control block, not the treatment, decides the credibility of the result. Comparing a strong treated block against a weak neighbour manufactures a lift that will not survive a buyer's scrutiny — stratify instead, and test both ends of the yield range.
BloomX's reported results at Allesbeste Boerdery in Limpopo, South Africa did exactly that: an average 16.5% yield increase with a peak of 20.23% across Maluma Hass, Hass and HMR varieties. Grower Zander Ernst has described the same pattern holding in low-yielding and high-yielding blocks alike.
Frequently Asked Questions
Where should extra passes go first — low-yield blocks or high-yield blocks?
Extra passes pay off in both low-yield and high-yield blocks, so the first decision is not which block type but which flowering windows inside each. In practice, growers prioritise blocks where flower load is high but fruit set is historically weak — that is where the unrealized gap is widest. BloomX's own framing of the avocado opportunity is blunt: a tree carries 1–1.5 million flowers yet sets only around 250 fruit, and Hass typically returns about 1 ton per dunam (a dunam being 0.1 hectare) against roughly 3 tons of carrying potential. Zander Ernst of Allesbeste described looking at "low yielding blocks improving production and also high yielding blocks. And what was nice is throughout both circumstances, we had 15%-20% increase in these blocks."
How do low-yield and high-yield blocks compare when allocating passes?
The two block types differ in why they respond, not in whether they respond. Use these criteria to weight allocation before you build the season plan:
| Criterion | Low-yield blocks | High-yield blocks |
|---|---|---|
| Main constraint | Poor fruit set; flowers go unworked | Set is decent, but fruit size and pack-out limit revenue |
| Response to extra passes | Larger relative lift off a small base | Smaller relative lift on a much larger tonnage base |
| Risk of over-investing | Low if the limiter is genuinely pollination, not water or nutrition | Low, but marginal passes compete with other blocks |
| Best diagnostic | Flower count vs. historic set ratio | Fruit weight and cull rate at pack-house |
| Typical priority | First, when bee activity is visibly weak | Second, where quality metrics have headroom |
Verdict: allocate opening passes to the weakest-setting blocks with a full bloom, then extend coverage into strong blocks once the peak receptivity window in those blocks is confirmed.
Why do managed honeybee hives alone leave yield on the table in both block types?
The managed honeybee is a generalist, and two high-value crops sit outside its strengths. Honeybees avoid Hass avocado's potassium-rich nectar, so a large share of flowers are simply never worked. Blueberry's bell-shaped flower needs buzz pollination — the rapid flight-muscle vibration a bumblebee uses to shake pollen loose from poricidal anthers — which honeybees perform far less effectively. Bio-mimicking pollination, BloomX's approach of mechanically replicating the right natural pollinator using the pollen already present in the orchard, works alongside hives rather than replacing them, and reduces the workload carried by the colony.
When is staying with hives alone the right call?
Staying put is reasonable when pollination is demonstrably not the limiting factor. If a block is water-stressed, nutritionally deficient, badly alternate-bearing, or too young to carry a commercial crop, extra passes will not fix the ceiling — correct the agronomy first. Very small or fragmented plantings may also not justify a full seasonal deployment. Blocks with reliably strong set and healthy hive activity can stay on the incumbent programme while you prove the model elsewhere on the estate during the 2026 season.
How is the timing of each pass decided?
Timing is driven by software that predicts the optimal pollination window and GPS-tracks each machine, so passes land on receptive flowers rather than on a fixed calendar. BloomX runs a full-service seasonal model: it owns, deploys and maintains the machines and assigns a project manager to the flowering season, then redeploys equipment across territories. YAHAV, the electrostatic unit for avocado and tree crops, and Robee, the vibration unit that replicates bumblebee buzz pollination on blueberry, are scheduled against that predicted window.
What results have growers reported across different block qualities?
On avocado at Allesbeste in Limpopo, South Africa, BloomX delivered an average 16.5% yield increase with a peak of 20.23% — roughly 2 tons per hectare across Maluma Hass, Hass and HMR. On blueberry (Rosita), Grupo Rotondo in León, Mexico recorded a 33.5% increase in marketable yield, a 16.7% reduction in cull fruit and a 12.9% increase in average fruit weight. BloomX reports 3X–5X return on investment per season, and points to more than six years of year-over-year commercial proof rather than a single trial.