Short answer: true-to-target (TTT) tasting asks “does this batch still taste like our brand?” — a conformance question, not an in-spec or a difference question. Its natural visualization is a radar/profile chart with the brand’s target band drawn in, so a batch reads at a glance as a shape that fits the band or pokes outside it. Add a deviation chart (batch minus target, attribute by attribute) and a conformance-over-time trend, and a panel’s verdict stops being a number in a logbook and becomes a picture the brewer, the panel and the brand owner all read the same way.
This is the flagship of the brewing data-visualization series. True-to-target tasting is something most breweries do — a panel checks the beer is “right” — but rarely visualize well, so the verdict lives as a tick in a logbook and its signal is lost. Here’s how to turn it into a picture.
A different question from “in spec”
Your analytical specs check parameters one at a time: ABV in range, IBU in range, colour in range, diacetyl below threshold. A batch can pass every one of those and still not taste like your flagship — because brand character lives in the combination and balance of attributes, not in any single number staying between two limits. True-to-target tasting (also called true-to-type or true-to-brand) asks the combined question directly: a trained panel rates the batch on the brand’s defined attributes — malt, hop aroma, bitterness, ester, body, finish — and the verdict is whether the whole profile matches the target. It is a conformance question, and it is distinct from a difference test, which asks only whether two samples differ at all.
The target band is the foundation
Before any visualization, the brand needs a defined target profile with a tolerance band — the agreed intensity for each attribute, plus how far a batch may deviate and still be “on brand.” This is set once from reference batches the brand owner signs off, and it’s the unglamorous prerequisite the whole method rests on; without an agreed target, “true-to-target” has no target, and the chart is just one batch’s opinion. This is the sensory equivalent of the data foundation that every analysis needs.
Three views of one verdict
With a target band defined, the panel’s scores become three complementary pictures:
1 — The radar/profile chart. Plot the brand’s attributes as spokes, draw the target band as a ring, and overlay the batch. Conformance becomes a shape: a batch whose profile sits inside the band is true-to-target; one that pokes outside on “hop aroma” and caves in on “body” shows you how it’s off, not just that it is. The radar is powerful here precisely because brand character is a shape, and humans read shapes instantly.
2 — The deviation chart. A simple bar chart of batch minus target per attribute, centred on zero. It’s the radar’s information in a form that’s easier to rank and track — which attribute deviated most, and in which direction. Where the radar shows the gestalt, the deviation bars show the priorities.
3 — The conformance-over-time trend. One point per batch — an overall conformance score, or the count of attributes outside band — plotted across batches. This is where the method earns its keep beyond a single QC pass: it catches brand drift, the slow collective creep where each batch passes but the brand quietly moves over a year. That’s a slow-drift problem, and the next post’s CUSUM thinking applies directly to it.
Making the panel’s work visible
The quiet payoff is organisational. A logbook tick says “panel approved batch 412.” A radar inside its band, a deviation chart, and a conformance trend say why, how confidently, and whether we’re drifting — in a picture the brewer, the panel lead and the brand owner all read identically. It turns sensory from a gate people argue about into evidence people act on, the same shift that makes any tasting-panel data worth digitising.
Where this breaks
The honest section. The radar chart has real flaws — the enclosed area changes with attribute order and axis scaling, so it can exaggerate or hide deviations; keep the attribute order and scales fixed forever, and read it alongside the deviation bars rather than trusting the shape alone. It’s only as honest as the panel — true-to-target rests on a calibrated, aligned panel; an un-calibrated panel produces a confident radar built on noise, and averaging disagreeing tasters hides the disagreement. The target itself can drift or be wrong — brands evolve, reference batches age, and a target set years ago may no longer reflect the beer you actually sell; revisit it deliberately rather than treating it as permanent. And conformance is not quality — a batch can be perfectly true-to-target and still be a beer with a dull target; this method guards consistency, not excellence, and the two are different conversations.
The bottom line
True-to-target tasting answers “is this still our brand?” — a conformance question that numeric specs miss because brand character is a balance, not a set of independent limits. Visualize it three ways: a radar profile against the target band for the gestalt, a deviation chart for the priorities, and a conformance trend for brand drift across batches. Define the target band first, keep a calibrated panel, fix your axes, and read the radar with its caveats — and a panel’s verdict becomes a picture the whole brewery trusts. Next, the SPC that makes the drift trend rigorous: capability, CUSUM and EWMA.
Frequently asked questions
What is a true-to-target (or true-to-type) tasting? True-to-target tasting is a brand-conformance sensory assessment: a trained panel judges whether a production batch still matches the established target profile for that brand, attribute by attribute. Unlike a difference test, the comparison is against the brand’s defined profile, not another sample — the question is “is this still our beer”, and a pass/fail or a deviation score is the output.
How do you visualize brand conformance in brewing? A radar (spider) chart overlays the batch’s attribute intensities on the brand’s target band, so conformance reads as a shape that fits inside the band or pokes outside it. Pair it with a deviation chart (batch attribute minus target) and a batch-over-time conformance trend, so you see both where this batch deviates and whether the brand is drifting across batches.
Is true-to-target the same as a triangle test? No. A triangle (difference) test asks whether two samples are detectably different. True-to-target asks whether one batch matches the brand’s established profile. They answer different questions and use different visuals — difference tests visualize correct-identification counts and significance, while true-to-target visualizes a profile against a target band.