Techbanq

Buyer's guide · testing

Every proposal describes the same methodology

What separates a useful engagement from an expensive one is the document at the end. Ask for a redacted sample before award, and read it against this list rather than against the proposal.

Serving Saudi Arabia from our Riyadh office and our engineering hubs in Lahore and Karachi

Market
Saudi Arabia
Written for
CISO, CTO
Decision
Specify the deliverable before awarding a testing engagement
Languages
EN · AR

Cybersecurity credentials

  • National Cybersecurity Authority, Saudi Arabia

    NCA registration

    Registered cybersecurity service provider

  • ISO

    ISO/IEC 27001

    ISO/IEC 27001 certified

Certifications & registrations

Why this is the specification

Methodology sections are copied. Reports are not

We do not have an office in Saudi Arabia

We serve organisations there from our Riyadh office and our engineering hubs in Lahore and Karachi, travelling for the parts of an engagement that need to happen in person.

Read four proposals for the same engagement and the methodology sections will be close to interchangeable — the same recognised standards named, the same phases listed, the same reassurances about experienced consultants. This is not dishonesty; the methodology genuinely is broadly standard, which is exactly why it does not distinguish anyone.

The variance is in the document that arrives eight weeks later, and it is very large. One supplier returns a findings list exported from a tool with severity labels attached. Another returns a document where each finding carries the request that produced it, the account it was sent as, the pattern behind it and a fix written against the framework in use. Both engagements cost the same and were sold with the same methodology section.

So the useful procurement step is to ask for a redacted sample report before award and read it. A supplier who cannot produce one, or who produces one where the findings are indistinguishable from scanner output, has answered the question. This page is written so it can be used that way against us as readily as against anyone else.

The checklist

Eight things a report has to establish

A stated scope, including exclusions

The systems tested, the accounts used, the dates, and what was explicitly out of scope.

What its absence costs: A clean report reads as coverage of everything, and someone will cite it that way later.

Reproduction steps as text

The exact request, the account, the response. Copy-pasteable, not a screenshot of a tool.

What its absence costs: A developer cannot confirm the issue, so the first response is to dispute it.

The pattern stated separately from the instances

The shape of the defect described once, then the endpoints where it was observed.

What its absence costs: The team patches the listed instances and leaves the rest. This is the single most common incomplete fix.

Impact against the real data

What the reachable object actually contained, named — not a severity label alone.

What its absence costs: Remediation is sequenced by label rather than consequence, and a notification question cannot be answered.

Remediation written for your stack

The change expressed against the framework in use, not a generic control reference.

What its absence costs: The engineering team has to translate before it can estimate, and translation is where fixes stall.

What was not reached, and why

Areas in scope that could not be exercised — blocked by rate limiting, missing credentials, a broken environment.

What its absence costs: The residual unknown disappears, and the report claims more assurance than the work produced.

Two audiences, separated

A short section a non-technical reader can act on, and a technical body that does not repeat it.

What its absence costs: Either the board section is unreadable or the engineering section is padded to reach a page count.

Retest carried in the same document

Each finding with its state at retest and the date it was verified, in one file.

What its absence costs: Answering an auditor a year later means reconstructing the history from an email thread.

Padding

What length is usually hiding

  1. 01

    A methodology chapter

    Twelve pages describing a standard the reader can look up. One page stating which standard was followed and where the engagement deviated from it is more useful and more honest.

  2. 02

    A tool output appendix

    Two hundred pages of scanner results attached to a fifteen-page report. It adds weight rather than information, and it is worth asking whether the findings in the body were verified by hand or lifted from it.

  3. 03

    Informational findings inflating the count

    A missing header, a server banner, a cookie flag on a page with no session. Worth listing briefly; not worth a page each with a CVSS score attached. A report with sixty findings and four that matter has made the four harder to see.

  4. 04

    A compliance mapping nobody asked for

    A table mapping each finding to five frameworks, generated automatically. Useful when you are working to one specific framework and asked for it; otherwise it is length. Ask for the mapping you need and skip the rest.

What a report is

What the record of work establishes

What we do

  • Send a redacted sample report before any commercial conversation, so the deliverable is judged rather than described.
  • Provide a letter of attestation stating scope, dates, methodology and the status of findings at retest.
  • Run the debrief twice where the audiences differ, rather than once at a level that suits neither.

Where the report is evidence for a framework assessment, the assessor sets the format, the required scope and the independence they expect. Establish that before the engagement rather than rewriting the document afterwards.

Primary source

Common questions

What buyers ask

Will a supplier really send a sample report?+

A redacted one, yes, and the reluctance to is itself informative. What you should get is a real engagement with the client, the systems and the specific identifiers removed — enough that you can read one finding end to end and judge whether an engineer could act on it. A generic template with the findings replaced by placeholders is not the same thing and does not answer the question.

Should the report include a risk rating?+

A rating is useful for sorting and useless as a decision. A published scoring system rates a defect against a generic environment; it does not know that the affected object holds identity numbers, or that the system is reachable only from an internal network. Ask for the standard score and a separate sentence on impact against your data, and prioritise on the sentence.

How long should we keep a report?+

Long enough to answer the next assessor, which in practice means at least until two more tests have superseded it, and treat it as sensitive throughout. A penetration test report is a description of how to attack your systems; it belongs under the same access control as the systems it describes, and it should not be circulated to a customer in full when an attestation letter is what they asked for.

Ask us for a redacted sample report

A redacted sample report, sent before any commercial conversation, so the deliverable can be judged rather than described.

Start the conversation