Skip to content
Menu
All work

Client projectPrivate

A market-wide price study, checked three ways before anyone saw it

For a European parcel operator, 5,093 competitor prices across 226 destinations, turned into a branded dashboard and a live workbook the pricing team could trust.

Industry
European parcel operator
What it shows
Competitor intelligence, OSINT, Data collection, Dashboards
Year
2026
Status
Private
Competitor price studyExport parcel prices, SeptemberDummy data226 of 237 destinations compared · 15 providers · 5,093 prices collectedYour price against the market10 of 226 destinations, most above the median firstYour priceCompetitor rangeMedianAbove every competitorvs median€10€20€30€40€50€60Price per parcelDestination J+€19Destination G+€16Destination D+€12Destination B+€6Destination I+€2Destination A−€1Destination F−€1Destination C−€2Destination E−€3Destination H−€3Destination KNo competitor quote–FindingsAbove every competitor in 3 destinationsJ, G and D. Review these first.Below the median in 5 of 10Room to raise price and stay under the median.Gaps are shown, not estimatedDestination K has no competitor quote, so it stays blank.37 headline figures checked three waysThree independent routes, zero disagreements.
Recreated for this page with dummy prices and names. The coverage counts and the verification note are the project's real figures.

In numbers

competitor retail prices collected
5,093
destinations compared, gaps marked rather than guessed
226 / 237
providers mapped to one list
15
workbook formulas recalculating with zero errors
4,808

Figures come from the project’s own files and records.

What changes

The steps people did by hand, and the route the work takes now.

The pricing team asks where it stands

Before

  1. Visit every comparison site
  2. Copy quotes into a spreadsheet
  3. Match carrier names by hand
  4. Re-add the totals

With AI

  1. An agent collects and cleans 5,093 prices
  2. Every headline figure checked three ways

A dashboard and live workbook the team trusts

The problem

The pricing team needed one answer: where do our export prices sit against every competitor, destination by destination? The data existed, but it was scattered across comparison sites and a competitor’s own price calculator, in different formats, with carriers named a dozen ways. Collecting it by hand would have taken weeks and still left doubts about the numbers.

What we built

A collection and analysis pipeline that gathered 5,093 retail prices from 15 providers, cleaned them into one consistent table and produced two deliverables in the client’s own language and branding: a dashboard for reading the market at a glance, and a workbook with live formulas for the analysts who wanted to test their own assumptions.

Where a destination had no quote, it stays visibly blank. The report shows the coverage gaps rather than filling them with estimates.

Why the numbers can be trusted

A price study is only useful if nobody has to wonder whether a figure is wrong. Every headline number was calculated three separate ways: from the raw data, through the workbook formulas and through the dashboard pipeline. All 37 matched. The workbook’s 4,808 formulas recalculate with zero errors.

What it shows

Most of the work in a market study is collection and cleaning, and that is exactly the part AI-assisted tooling does well when it is checked properly. The same pipeline can be pointed at another market, or run on a schedule to become a standing monitor that tells the pricing team what moved each week. Below, sort the price view yourself, or switch to the weekly monitor: an illustration of where the same approach goes next, reading more open sources than prices, from job ads to tariff PDFs.

Interactive · 6 journeys

Interactive system map

Every step of the study, and how they connect

16 parts and 6 journeys, from a public quote to the workbook, the dashboard and the three-way check. The weekly monitor chapter is an illustration.

Play 6 journeys, or click any of its 16 parts.

competitor retail prices collected
5,093
headline figures checked three ways, all agreed
37
workbook formulas, recalculating with zero errors
4,808

Explore the map

InteractiveSort the prices, show only the problem rows, or switch to the weekly monitor.
View
A sample of 20 priced destinations and 2 gaps, from 237. 4 are above every competitor.
Your price against competitor prices per destination, dummy data. Sortable.
Price chart
J+€19, above every competitor
G+€16, above every competitor
D+€12, above every competitor
O+€12, above every competitor
B+€6
M+€5
T+€5
Q+€4
I+€2
A−€1
F−€1
L−€1
N−€1
P−€1
S−€1
C−€2
U−€2
E−€3
H−€3
R−€3
KNo quote, blank–
VNo quote, blank–
Illustrative. The same approach, run every week as a standing monitor. Competitors and signals are dummy data, not from the client project.

This week

  1. Competitor B is pushing into the Baltics: lower prices on 4 routes, a new branch and ads claiming the lowest price.
  2. Competitor C looks set to open a depot: 6 warehouse roles advertised in one new city.
  3. Competitor A’s service is slipping: complaints about late delivery doubled.
For you to decide
Whether to answer B’s Baltic prices, and on which routes.

Changes by competitor

Competitor
Source
11 changes this week
  1. 22 SepCompetitor BPrices
    Cut prices to 4 Baltic destinations, by 9 to 14%
    Evidence: Public price page, read from a screenshot
    Before and after captures of the price page. Destination D: €29.90 → €26.90; three more routes changed the same day.
  2. 22 SepCompetitor EPrices
    New tariff sheet: fuel surcharge up 1.5 points
    Evidence: Tariff sheet, read from a PDF
    Surcharge table on page 3 of the new tariff PDF, compared with last month’s file.
  3. 23 SepCompetitor CHiring
    Advertised 6 warehouse roles in a city where it has no depot
    Evidence: Job boards, read from text
    Six adverts posted within two days, all for the same new address. Often an early sign of a new depot.
  4. 23 SepCompetitor AProducts
    Launched next-day delivery to 3 destinations
    Evidence: Product page, read from a screenshot
    New service tier on the product page, with its own price column.
  5. 24 SepCompetitor BAds
    Running 12 new ads that claim the lowest price to the Baltics
    Evidence: Public ad library, read from images
    Twelve new ad creatives; the agent read the claim from the images themselves, not from captions.
  6. 24 SepCompetitor AReviews
    Complaints about late delivery doubled week on week
    Evidence: App store reviews, read from text
    41 new reviews; 17 mention late delivery, up from 8 the week before.
  7. 25 SepCompetitor BFilings
    Registered a new branch address in the Baltics
    Evidence: Business register, read from text
    New branch entry in the public company register, filed on 19 Sep.
  8. 25 SepCompetitor CProducts
    Removed the 30 kg option from checkout
    Evidence: Checkout page, read from a screenshot
    Weight selector captured before and after: the top band is gone.
  9. 26 SepCompetitor DHiring
    Hiring a head of pricing
    Evidence: Job boards, read from text
    Senior pricing role advertised; the advert mentions “dynamic pricing”.
  10. 26 SepCompetitor EFilings
    Won a public tender for parcel lockers
    Evidence: Public tender portal, read from a PDF
    Award notice published as a PDF on the tender portal.
  11. 26 SepCompetitor DAds
    Started advertising weekend delivery
    Evidence: Public ad library, read from images
    Four new ad creatives, all built around Saturday delivery.
Dummy data throughout. The prices view mirrors the delivered dashboard; the weekly monitor shows where the same approach goes next.

Under the hood

  • Claude
  • Python
  • Playwright
  • openpyxl
  • SVG charts

  1. 01Collect (OSINT)

    Open-source intelligence: quotes from public price-comparison sites and a competitor's public pricing service, gathered through a real browser session.

  2. 02Normalise

    Python parses every quote into one shape and maps dozens of carrier spellings onto a single list of providers.

  3. 03One pipeline

    A single pipeline feeds both deliverables, so the workbook and the dashboard cannot disagree.

  4. 04Workbook

    Built with live formulas, so the pricing team can change an assumption and watch it flow through.

  5. 05Dashboard

    A standalone report with charts, in the client's branding, readable on any device.

  6. 06Verification

    The 37 headline figures were recomputed by three independent routes before delivery. They agreed every time.

Want something like this built for your team?

Tell us about the process or the data. We'll reply within two business days with a first view of what we'd build.

Start a project