7 Questions to Ask Before Hiring a SharePoint Consultant
Most SharePoint projects that go wrong were agreed in a pleasant meeting that nobody examined closely. Seven questions to ask before you sign.

Choosing a SharePoint consultant is one of those decisions whose consequences arrive late. The proposal gets read in a week. The quality of what was decided in that week shows up two years later, when somebody needs to change an approval threshold and discovers they cannot do it without raising a ticket.
The pattern is consistent enough to be useful. Projects that go well tend to start with a buyer asking uncomfortable questions, because those questions force both sides to be specific about scope, ownership and what happens after launch. Projects that go badly tend to start with a pleasant meeting, an impressive demonstration and a proposal nobody examined closely.
Here are seven questions worth asking before signing anything, what a strong answer sounds like, and what should make you uneasy.
First, a word in SharePoint's defence
Before the questions, it is worth addressing something you will hear in procurement meetings across the region: "we tried SharePoint and it did not work for us."
In most cases where those deployments are examined closely, the platform was not the problem.
SharePoint Online is one of the most capable content platforms available to any organization, and most enterprises already own it. It brings enterprise-grade security and compliance, an identity model already wired into the business, versioning and retention, search across the estate, automation through Power Automate and Power Apps, and native presence inside Teams and Office where people already work. It handles very large content estates without complaint, and in 2026 it has become the grounding layer that makes assistive AI genuinely useful on a company's own documents, which is a capability no standalone document platform can match without a significant integration project.
What usually failed was something else. A 2013 on-premises farm that was never upgraded. A deployment with no information architecture, where teams were handed empty sites and told to organize themselves. A migration that lifted a fifteen-level folder tree across unchanged and added nothing. A project with no owner after go-live.
SharePoint is a platform, not a finished product. That is its great strength, because it can be shaped to how a business actually runs rather than how a vendor assumed it should. It is also exactly why the choice of who shapes it matters more here than with a closed system. With a packaged product you are buying someone else's assumptions. With SharePoint you are buying the capacity to encode your own, and the quality of that encoding is the consultant's contribution.
Which is what these questions are for.
1. What have you built that is still being used three years later?
Anyone can show you a demonstration. Almost anyone can get a system live. The interesting question is whether what they built survived contact with reality, staff turnover and a reorganization.
A strong answer names a specific client and system, says what it does now that it did not do at launch, and explains who has been maintaining it. Bonus points if they can describe something that needed rework and why.
Be uneasy if every reference is from the last twelve months, or if the portfolio is a collection of launches with no afterlife. A consultancy that has never had to live with its own decisions three years on has not yet learned which decisions matter.
Ask to speak to one of those long-standing clients. The question to ask them is not whether they were happy at launch. It is whether they can still change the system today.
2. What would you advise me not to put on SharePoint?
This is the single most revealing question on the list, and it is the reason it comes early.
SharePoint is excellent for documents, departmental processes, structured lists, approvals, collaboration and portals. It is the wrong home for transactional systems of record, applications with heavy relational logic and high write volumes, regulated clinical systems, and anything requiring real-time processing. Within the Microsoft stack those belong on Dataverse, Azure SQL or a proper line-of-business system, and a good consultant will say so without being prompted.
A strong answer draws that line clearly, gives examples from their own work where they recommended against SharePoint, and explains what they proposed instead.
Be uneasy if the answer is some version of "SharePoint can do anything." It can host almost anything. That is not the same as being the right place for it, and a partner who will not tell you where the boundary is will eventually build you something on the wrong side of it.
3. Who will actually do the work, and can I meet them before I sign?
The oldest problem in consulting. The people in the pre-sales meeting are frequently not the people who will deliver.
A strong answer names the individuals, describes their roles and experience, and offers a working session with the actual architect or lead developer before contract. Many good firms will happily run a paid discovery workshop, which is also the fastest way to find out whether you can work together.
Be uneasy if the response is about the firm's capabilities rather than named people, or if the team will be assigned after signature. You are not buying a logo, you are buying the judgement of two or three specific individuals.
Also ask the unglamorous version of this question: what happens if that person leaves mid-project?
4. How will you find the rules nobody has written down?
In any process that has been running for a decade, roughly a fifth of how it actually behaves is undocumented. The approval that skips a step for one department. The field that is technically optional but which finance treats as mandatory. The exception that exists because the old system forced it.
These rules do not appear in requirements documents. They appear during user acceptance testing, when someone says the new system does not work, meaning it does not reproduce a behaviour they never mentioned.
A strong answer describes a structured discovery approach, talks about interviewing the people who do the work rather than only the manager who owns it, and specifically mentions asking about exceptions: when do you not follow the normal path, and why. The best answers also say that they treat each discovered rule as a decision rather than a requirement, because half of those rules exist only because the old system demanded them.
Be uneasy if discovery is one workshop in week one, or if the proposal moves from kickoff to build with nothing in between. That proposal is cheaper for a reason, and you will pay the difference in change requests.
5. What will this cost me beyond your invoice?
The implementation fee is the visible number. It is rarely the whole number.
Depending on what is being built, there may be additional licensing for Power Apps or Power Automate premium connectors, consumption-based costs for Copilot Studio agents or AI-assisted workflows, Azure resources for custom components, third-party migration tooling, and internal time from your own IT and business teams, which is real cost even though it never appears on a quotation.
A strong answer itemizes these unprompted, distinguishes between what is covered by the licences you already hold and what is additional, and models consumption costs for anything metered rather than leaving you to discover them in month two.
Be uneasy if the answer is that everything is included in your existing Microsoft 365 licences without qualification. Much of it genuinely is, which is one of the strongest arguments for building here. But "all of it" is almost never true, and a partner who will not tell you that at proposal stage will not tell you other inconvenient things either.
6. What does day one after go-live look like, and who owns it?
Automation and document systems are not projects that finish. Approval thresholds change, people move, departments reorganize, new document types appear. If nobody owns the system afterwards, it drifts away from how the business works and people start routing around it. Within eighteen months it becomes the legacy system that the next consultant is asked to replace.
A strong answer covers hypercare in the first weeks, a defined support arrangement with response times, training for both users and the person who will administer the system, and a documented handover. It will also tell you plainly that you need a named owner on your side, and may decline to proceed without one.
Be uneasy if support is an afterthought priced separately with no detail, or if the proposal ends at go-live. The go-live date is the midpoint of the work, not the end of it.
7. Can my team change it without you?
The final question, and the one that tells you what kind of relationship you are entering.
A well-built SharePoint solution should let your team add a document type, change an approval threshold, add a site from a template or adjust a view without raising a ticket. The deep work, architecture, integration, complex automation, is where a consultant earns their fee. Day-to-day configuration should not be.
A strong answer describes what will be handed over: documented architecture, a solution deployed through proper application lifecycle management, admin training, and a clear division between what your team can safely change and what should come back to them. A confident firm is happy to make itself unnecessary for routine work, because the clients who stay are the ones who stay by choice.
Be uneasy if the architecture depends on undocumented custom code, if no training is offered, or if every change however small requires a change request. Dependency is a business model, and it is worth recognizing before you sign rather than in year two.
Two questions that matter less than you think
"How many certified consultants do you have?" Certifications indicate exposure, not judgement. A team of five certified people who have never had to maintain anything is worth less than two who have.
"Have you worked in my exact industry?" Useful, not decisive. Domain knowledge is learnable in weeks by a good consultant who asks the right questions. Architectural judgement is not learnable in weeks by anyone. Prefer the firm that understands document control deeply over the one whose only qualification is a logo from your sector.
What good looks like
A partner worth hiring will tell you what not to build, name the people doing the work, insist that you appoint an owner, be specific about costs you have not thought of, and leave you able to run the thing yourselves.
And a last point, because it is the thesis underneath all seven questions. The organizations that get extraordinary value out of SharePoint are not the ones that bought the most ambitious implementation. They are the ones that got the information architecture right, kept the governance simple enough to actually follow, and worked with someone who was honest about the boundaries. The platform has been ready for years. What varies is the judgement applied to it.
Digitize Flow designs and builds document management systems, intranets and automated workflows on SharePoint and Microsoft 365 for enterprises across the Middle East, with long-running client systems we still maintain today.
Book a discovery session and we will walk through your current estate, tell you plainly what belongs on SharePoint and what does not, and scope what the first phase should be.


