Analytics · The path to value

From data to value in six steps. Most analytics stops at the third.

Data becomes information, information becomes insight, and that is where most dashboards end. The value arrives three steps later: a decision a named person makes, an action the system takes or routes, and a result measured with your Finance team. This page walks the path step by step, with a real case on it.

Steps on the path
Six
Where dashboards usually stop
Step three, insight
Where value is measured
With your Finance team
Worked example
APAC margin agent, $240k recovered
Median to first production artefact
6 weeks
On this pageThe six steps, and what each one needs
  1. 01The six steps, and what each one needs
  2. 02Where the path breaks
  3. 03One path, drawn from a real case
  4. 04Where each part of woodfrog sits on the path
  5. 05Questions buyers ask
01

The six steps, and what each one needs

Each step has to hold before the next one is worth building. Skip one and the steps after it are built on sand.

  1. 01 DataRecords, as the systems hold them

    Rows in the ERP, the CRM, the plant systems and the spreadsheet somebody keeps. Nothing is wrong with them yet. They just do not agree with each other.

  2. 02 InformationJoined, reconciled and owned

    The same customer, order or batch agreeing across systems, with lineage back to source and a named owner for each measure. This is where two teams stop arguing about the number.

  3. 03 InsightA finding that deserves attention

    Not a chart, a finding: this supplier's landed cost is drifting, this group of accounts is going quiet. An insight names what moved, by how much, and why it matters now.

  4. 04 DecisionA named person chooses, with the evidence

    An insight worth having reaches someone who can act on it, with the evidence attached and a threshold agreed in advance. Insights nobody owns are where most of the value leaks.

  5. 05 ActionThe system acts, or routes it to whoever does

    A held purchase order, a reassigned account, an alert to the floor with the fix attached. The action is logged with what triggered it, so it can be traced later.

  6. 06 ValueMeasured, and signed off

    Money recovered, margin held, hours avoided, confirmed by the people who own the number. A figure nobody on the customer side has checked is not published on this site.

02

Where the path breaks

The symptom, and the step that is missing

What people say on the first call usually names the step that broke, even when they describe it as a reporting problem.

Two teams bring two different numbersInformation: no reconciled, owned measure
240 dashboards and still no answerInsight: charts without a finding
The dashboard showed the fall, not the causeInsight: the drift was never watched
Everyone saw it, nobody owned itDecision: no named person or threshold
The report lands after the week it describesAction: the finding arrives too late to act on
Nobody can say what the project was worthValue: never measured with Finance

The second and third lines come from our published cases: a fintech estate of 240 dashboards cut to eight, and an APAC grocery business whose dashboard reported the margin fall without a cause.

What each step has to hold before the next is worth building

These are the checks we run as the path is built, one decision at a time.

  1. 01Information: counts reconcile to source

    By entity and by period, with the differences listed rather than summarised into a percentage.

  2. 02Insight: one definition and one owner per measure

    If two leaders define revenue differently, the finding will be argued with instead of acted on.

  3. 03Decision: a named person and a threshold agreed in advance

    Below the threshold the system routes the case to that person with the evidence. The threshold is a business decision made with you.

  4. 04Action: every action logged with what triggered it

    So a decision can be followed back months later, by someone who was not in the room.

  5. 05Value: checked by the people who own the number

    Your Finance team confirms the figure before anyone claims it, ours or yours.

03

One path, drawn from a real case

The APAC margin agent, step by step

A retail grocery business above $500M in revenue, six weeks, fixed scope and fixed price. Every fact below is on the published case.

  1. 01 DataFour cost fields, read at source

    Contracted price, actual invoice, fuel surcharge and landed cost, read directly instead of waiting for them to reach a report.

  2. 02 InformationEvery field traced to its source

    Lineage came before the agent. It is the part that later let Finance sign the number off.

  3. 03 InsightA fuel-surcharge anomaly

    The gap between what was contracted and what was invoiced, which nobody had been watching across suppliers.

  4. 04 DecisionActed on, on day 14

    The first flag reached the procurement team two weeks before close, with its evidence attached, and they acted on it.

  5. 05 ActionFinance traced the claim

    Finance followed the claim back through every number the agent had used, rather than taking it on trust.

  6. 06 Value$240k, signed off in 30 days

    Annualised procurement savings, signed off by the customer's Finance team within 30 days, and 2.4 points of gross margin recovered before the Q4 board call.

The case in four figures

Money figures were signed off by the customer's own Finance team.

$240kRecovered

Annualised procurement savings, signed off by Finance within 30 days.

+2.4ppGross margin

Recovered between Q3 close and the Q4 board call.

14 daysTo first flag

From kickoff to the first agent flag the procurement team acted on.

6 weeksEngagement

Fixed scope, fixed price.

04

Where each part of woodfrog sits on the path

The path is why the four verticals exist as one firm: each one owns a stretch of it, and the value only shows up when all of them hold.

  1. Steps 1 and 2Data foundations

    Pipelines, integration and governance that turn records into information both teams accept.

    Data foundations ↗
  2. Step 3Analytics

    Dashboards, command centres and Apache Superset, built around the finding rather than the chart.

    Analytics ↗
  3. Step 4AI

    Agents and evaluation that bring a decision to a named person, with the evidence and a confidence score.

    AI ↗
  4. Step 5Digital engineering

    The applications and automations where the action happens, and where it is logged.

    Digital engineering ↗
  5. Step 6Measured with your Finance team

    A figure goes on this site only after someone on the customer side who owns that number has confirmed it.

    The case studies ↗
05

Questions buyers ask

Is this a data maturity model?+

No. A maturity model scores the whole organisation. This is the path one decision takes, and we build it one decision at a time, starting with the one costing you the most.

Do we have to build all six steps before we see any value?+

Value is only measured at the sixth step, so yes for one decision, and no for the business as a whole. The first engagement takes one decision the whole way, narrow on purpose.

Where do most companies get stuck?+

Between insight and decision. The dashboard exists, the finding is on it, and nobody owns acting on it. That gap is usually cheaper to close than building another dashboard.

How long does one path take?+

Across 20+ companies, the median time to a first production-grade artefact is six weeks. On the APAC margin agent the first flag was acted on at day 14 and Finance signed off within 30 days.

Where does AI fit on the path?+

At steps four and five, and only once step three holds. An agent that decides on data nobody reconciled makes the wrong call faster.

Ready when you are

Start with the decision costing you the most

Two calls, a fixed fee, and a one-pager you keep either way: which decision to take the whole way first, and which step on its path is missing today.