Extension revenue share
The apportionment of third-party extension revenue by PAID USAGE: for each extension (and its developer), per period, how much was charged and how it splits between the DEVELOPER (the dev's share, 70% by default) and the PLATFORM (the remainder). The total always reconciles: dev share + platform sh…
What you find on this screen
A portrait of the screen, read from the screen itself — it changes when the screen changes.
Form fields
Nome*Como o teste é chamado na operação (ex.: preço do plano básico).
Métrica de desfechoO que decide o vencedor (ex.: conversão da proposta).
Variante A (controle)O que é feito hoje.
Variante B (nova)O que se quer testar.
Efeito mínimoFração com ponto: 0.05 = 5 pontos percentuais. É a menor diferença que vale a pena detectar.
Alfa (significância)O risco aceito de um falso positivo (o usual é 0.05).
PoderA chance de detectar o efeito, se ele existir (o usual é 0.80).
n necessárioCalculado pelo botão "Calcular poder": quantas observações o teste precisa no total.
n atualQuantas observações já entraram.
Sucessos ASucessos observados no controle.
Amostra AObservações do controle.
Sucessos BSucessos observados na variante.
Amostra BObservações da variante.
SituaçãoRascunho → Rodando → Concluído, pelo teste sequencial.
ResultadoO que o teste sequencial disse na última avaliação (efeito, IC e n).
DecisãoA leitura descritiva registrada quando o teste concluiu — a escolha comercial é sua.
Criado emQuando o experimento foi criado.
Start here
O que esta tela responde
How to test a change in your operation so that the result means something: what sample is needed, and when to stop.
What this screen is
Test price, cadence and discount policy in YOUR operation with correct statistics — something nobody delivers to SMEs. The POWER calculation says how many observations are needed to detect the minimum effect you care about (with the chosen alpha and power); the SEQUENTIAL TEST lets you follow the result over time WITHOUT inflating the error (the boundary only declares a conclusion at the right moment — no "I peeked and it was significant"). 🔴 The decision is descriptive: the effect with interval and n; the commercial choice is yours.
A proper A/B test, no statistician needed
Describe the two variants and the metric, state the smallest effect that matters and click "Calculate power": Tenka tells you how many observations the test needs. Then "Evaluate (sequential test)" — in the form or in the row’s ⋮ — tells you whether you can already conclude.
How to
"n necessário": por que a conta vem antes do teste
You declare the smallest effect worth detecting, and the screen calculates how many cases are needed to see it. Running a test with an insufficient sample is spending weeks to end up with no answer — and concluding "it made no difference" when what was missing was size.
Efeito mínimo, alfa e poder: os três números que definem o teste
Minimum effect is the smallest difference worth detecting, as a fraction (0.05 = 5 percentage points) — the smaller it is, the more observations the test requires. Alpha is the accepted chance of declaring an effect where there is none (false positive). Power is the chance of detecting the effect when it really exists. All three feed the calculation of the required sample (required n), which the table compares with the sample already collected (current n) while the experiment runs.
Standard list and form
The screen became the standard list: search, filters, saved views, columns, export and trash with Undo. Click the record (or New) to open the form, with the detail tabs when there are any.
Save, browse and undo
Save stores without leaving the form (Ctrl+S); Save and close returns to the list; the header arrows move from one record to the next. Delete sends the record to the list’s trash, from where Undo brings it back.
Rules
What this screen answers
Whoever watches the scoreboard continuously and stops when winning finds an "effect" where there is none — the error accumulates with every peek. The sequential monitoring on this screen is designed to allow looking without inflating that error: it tells you when to conclude.
O registro é descritivo
What is recorded is the effect with interval and n. The choice to adopt the change or not belongs to the business owner — the screen neither makes it nor recommends it.
Rascunho, Rodando, Concluído: o ciclo do experimento
The experiment is born when saved, with variants A (control) and B (new) and the outcome metric declared. While it runs, the "current n" column shows the collected sample against the required n. The evaluate button applies the sequential test: if the boundary has not yet been crossed, the experiment keeps running; when it is crossed, it is concluded and the result with the decision appears on the table row itself — no "I peeked and it was significant".
The test only concludes at the right time
Checking the result every day and stopping when it looks good inflates the error. The sequential test corrects that: until the boundary is crossed it says "still collecting" — even if the difference looks large. The recorded decision is descriptive; the business choice is yours.
Frequently asked questions
Which question this screen answers
How many observations I need to test a price, cadence or discount — and when I can conclude without fooling myself. Fill in minimum effect, alpha and power and click "Calculate power" to see the required n before starting; then follow along with the sequential test (evaluate button on the row), which only declares a conclusion when the statistical boundary allows. The recorded result is descriptive: effect with interval and n — the commercial decision remains yours.
Was this article helpful?
In the same module: The company over time5
Purchase ordersThe commitment made with the winning supplier: items, quantities, amounts, and lead time. The NF-e receipt ma…Asset custodyAsset tagging and custody for fixed assets: who has each asset, since when, and under which term. Issuing it …Stock movementsThe stock ledger: receipts, issues, adjustments, and transfers. A receipt recalculates the average cost; an i…Purchase quotesRequest pricing from N suppliers and compare them on a grid (rows = items, columns = suppliers, cell = price)…NF-e receivingImport the purchase invoice's XML: Tenka reads the items and amounts (a namespace-tolerant parser), you check…
Tenka Academy
Explore the AcademyStep-by-step learning paths for your industry, inside the platform.