Data Governance Consulting Firms
How to choose between data governance consulting firms
Data governance is the easiest consulting engagement to fake. A policy document, a council with a monthly meeting, a tool licence, and nothing measurable changes. The firms worth hiring are the ones whose deliverable is a control that holds up in an audit, not a framework. This page covers what the work actually is, the questions that separate the two kinds of firm, and what a good engagement produces.
What separates a strong Data Governance consulting company
The deliverable is a control, not a policy
Ask what will be running at the end that was not running before. A certified metric with an owner. Lineage a regulator can follow. An access rule that is enforced rather than documented. A policy nobody enforces is the most common output of a governance programme and the least useful.
Tool-neutral, because the tool is not the problem
Collibra, Alation, Purview and the rest are all capable, and none of them resolves who owns a definition. A firm that opens with a platform recommendation is selling an implementation. Ask what they would do if you bought no new tooling at all.
They will name the owner, and make it stick
Almost every governance failure is an ownership failure wearing a technical costume. The useful consultant sits in the meeting where two directors disagree about a definition and does not leave until one name is written down against it.
Audit evidence as a first-class deliverable
In a regulated business the real test is an examiner asking where a number came from. Ask to see the artefacts a previous client handed to an auditor. Preparation time falling from weeks to days is the honest measure of whether governance worked.
Scoped to the metrics that matter
A firm proposing to govern the whole estate is proposing something that will not finish. The engagements that land start with the handful of metrics that appear in a board pack or a regulatory filing and expand from there.
A council that outlives the engagement
Ask what the ongoing cost of running the governance looks like after handover, and who on your side chairs it. If the answer requires the consultancy to stay, that is a retainer rather than a handover.
Proof
- 3 wks to 2 days: Call report preparation cycle at a regional bank, with no MRAs in the following exam
- $1.8M: Potential penalties avoided after resolving 23 audit deficiencies
- $2.8M a year: Reporting labour automated across 42 certified metrics in 8 university units
What data governance consulting actually involves
The work splits into ownership, definition, lineage, access and evidence. Ownership is the unglamorous core: a named person accountable for each metric and each critical table, agreed by the people who disagreed about it. Definition is turning four regional versions of a number into one certified version, which is where most of the political work happens.
Lineage is the traceable path from a figure in a board pack back to the system that produced it, which is the thing an auditor asks for and the thing most estates cannot produce. Access is who can see what, enforced rather than described. Evidence is the artefact set that proves all of the above was working on a given date, and it is the deliverable that turns a governance programme from an overhead into something that visibly shortens an audit.
What good looks like, in numbers
Governance work is unusually easy to measure once it is real, because the before state is almost always someone spending weeks assembling something by hand. Call report preparation falling from three weeks to two days. Audit preparation falling from six weeks to four days. A federal reporting cycle falling from three months to two days. Twenty three audit deficiencies resolved and no repeat findings in the following examination.
If a firm cannot produce a figure of that shape from a previous engagement, ask why. Governance that changed nothing measurable usually means a framework was delivered and no control was.
Where Thinklytics fits
We treat governance as an ownership problem with a technical tail, not a tooling purchase. The engagement starts with the metrics that appear in a board pack or a filing, establishes a named owner for each, and builds the lineage and evidence underneath them. The same person runs the definition meeting and writes the layer that enforces the outcome, so the agreement does not evaporate between workshops and implementation. We are tool-neutral, and on more than one engagement the recommendation has been to buy nothing.
Frequently asked questions
What do data governance consulting firms do?
They establish who owns each critical metric and table, turn competing definitions into one certified version, build the lineage that traces a number back to its source, enforce access rules, and produce the evidence an auditor asks for. The firms worth hiring deliver controls that run after they leave. The rest deliver a policy document and a council.
How do I choose a data governance consultant?
Ask what will be running at the end that is not running now, and insist the answer is a control rather than a document. Ask what they would do if you bought no new tooling. Ask to see the artefacts a previous client gave an auditor, and the before and after on preparation time. A firm that opens with a platform recommendation is selling an implementation rather than governance.
Do we need a data governance tool like Collibra or Alation?
Often later, rarely first. Those platforms are capable and none of them decides who owns a definition, which is the actual blocker. Start with the handful of metrics that appear in a board pack or a filing, establish ownership and lineage for those, and you will know what you need a catalogue to do before you pay for one. Several engagements have ended with a recommendation to buy nothing.
How long does a data governance engagement take?
A scoped programme on the metrics that matter runs in weeks. Fourteen regional definitions of a patient encounter became one certified definition in 11 weeks, with clinical reporting running uninterrupted throughout. A programme proposing to govern the entire estate before delivering anything is a programme that will not finish.
How much does data governance consulting cost?
The drivers are how many metrics are in scope, how many systems have to agree, how contested the ownership is, and what evidence a regulator requires. Thinklytics prices by deliverable rather than by an hours bucket, so scope and cost are agreed before work starts. Request the audit for a fixed quote.
What is the difference between data governance and data management?
Data management is the operational work of running pipelines, storage and quality. Governance is who decides, who owns, who may see, and how you prove it. They are often sold together and the distinction matters when things go wrong: a management failure produces a broken pipeline, a governance failure produces two numbers that are both defensible and an organisation with no way to choose between them.