Understanding Brand Learning — Sennos
Sennoslink — Brand Learning

How accuracy improves
over time

Brands help Sennoslink understand what "normal" looks like for your fermentation process. The more you ferment, the smarter it gets.

How Brand Learning works

When a Brand is first created, Sennoslink has limited historical context. As more fermentations are completed under the same Brand, patterns begin to emerge — allowing Sennoslink to evaluate fermentations more accurately, compare batches more meaningfully, and provide clearer insights over time.

Early predictions may appear broader or less precise because the Brand is still building context. This is expected. Each completed fermentation provides the baseline information needed to understand how a Brand behaves under real production conditions.

What makes it work

Two things Sennoslink needs from every fermentation

Brand learning happens automatically — no special configuration required. But for the system to learn effectively, every fermentation needs two things committed consistently:

Original Gravity & Final Gravity
Culture (with generation information)

Manual readings between start and end aren't required, but act as checkpoints — helping the system build accuracy faster and giving you a more complete picture of each fermentation.

Brand Maturity

Three stages of Brand Learning

As fermentation data builds, each Brand naturally progresses through three stages. The dot next to each Brand name in Sennoslink shows its current stage — select one below to explore.

Stage 1 of 3 Learning
What's happening
  • Few or no completed fermentations
  • Sennoslink is collecting baseline fermentation data
  • Typical behavior is not yet clearly defined
What you may notice
  • Predictions may vary between fermentations
  • Comparisons are limited
  • Benchmarks, alerts, and KPIs are minimal or unavailable
Sennos is learning this brand's fingerprint. Predictions may be less accurate until more data is available. Consistently recording Original Gravity, Final Gravity, and Culture with each fermentation will accelerate learning.
Stage 2 of 3 Refining
What's happening
  • Multiple fermentations have been completed
  • Common trends and variation ranges becoming clearer
  • Brand-level aggregation begins to reflect real behavior
What you may notice
  • Predictions become more consistent
  • Visualization views show meaningful averages and ranges
  • Benchmarks and alerts start to align with actual production
Sennos is building accuracy for this brand as more readings are added. This is a good time to review and complete your Brand information — gravity targets, style, and yeast profile — to further improve accuracy.
Stage 3 of 3 Trained
What's happening
  • Strong history of consistent fermentations
  • Typical behavior is well established
  • Outliers are easier to identify and manage
What you may notice
  • High confidence in predictions and comparisons
  • Benchmarks closely reflect normal production behavior
  • Alerts focus on meaningful deviations, not noise
Sennos has a strong understanding of this brand based on consistent data over time. This stage supports both day-to-day monitoring and long-term process optimization.
Under the Hood

How Brand data improves accuracy

Every completed fermentation makes the next one more reliable. There's no finish line — the system improves continuously.

Sennoslink automatically aggregates fermentation data across batches, establishes expected behavior and normal variation, and improves comparisons between individual fermentations and Brand-level trends.

Brand-level views and analysis become more reliable as historical data grows. Filling in complete Brand information — including Original Gravity, Final Gravity, and Culture — at every fermentation helps it learn faster.

Example — same Brand over time
Illustrative only. Your Brand may progress faster or slower.
~2 fermentations — wider variation, patterns just beginning
~5 fermentations — trends emerging, benchmarks forming
~8 fermentations — stable patterns, high-confidence comparisons
Best Practices

Improve accuracy faster

Accuracy improves automatically over time, but a few consistent habits make a meaningful difference.

Start here Complete your Brand information Fill in as much Brand detail as possible — style, gravity targets, fermentation profiles. The more complete the picture, the more accurately Sennoslink can establish what "normal" looks like for this Brand.
Assign fermentations consistently Always assign each fermentation to the correct Brand. Misassigned fermentations introduce noise that works against the model.
Record Original Gravity & Final Gravity These are required inputs. Providing them every fermentation gives Sennoslink the complete arc it needs to establish Brand patterns.
Record Culture with generation information Culture data — including generation — is a required input. It helps Sennoslink understand how yeast performance evolves across batches.
Add manual readings as checkpoints Manual readings between start and end aren't required, but act as checkpoints — helping the system build accuracy faster and giving you a more complete picture of each fermentation.
Exclude test runs and outliers Known anomalies and test fermentations can skew Brand learning. Excluding them keeps the model grounded in real production data.
Advanced Features

When to use advanced Brand features

Benchmarks, alerts, and KPIs are most effective once your Brand has enough history to establish a reliable baseline.

Benchmarks

Compare individual fermentations against Brand-level norms. Most meaningful once multiple fermentations have established a clear typical range.

Alerts

Surface meaningful deviations from expected behavior. Alerts reach you wherever you work.

In-app SMS Email Push

KPIs

Track Brand-level performance indicators across batches. Require sufficient history to calculate reliable aggregates and meaningful trends.

Using advanced features too early may result in misleading trends or excessive alerts due to limited data. For best results, allow the Brand to reach at least the Refining stage before relying heavily on these tools.
Key Takeaway

The more you ferment, the more Sennoslink learns

Brand learning is an ongoing process. Every completed fermentation with complete data improves the system's ability to support confident, accurate decision-making.

Define expected behavior
Identify meaningful variation
Support confident decisions
Start fermenting first. Accuracy follows.