This guide is for anyone whose business runs on software they did not build themselves — SaaS subscriptions, hosting providers, third-party APIs, and managed services — and who is trying to get genuine value from those relationships rather than just paying invoices. By the end you will know how to structure the relationships, what to track, how to escalate when things go wrong, and how to approach renewals so you do not end up locked into vendors who have stopped earning the price.
Who This Guide Is For
Operations leads, IT managers, COOs, and finance leads with a portfolio of software vendors that the business depends on. Most modern businesses have between five and fifty active SaaS subscriptions plus hosting and infrastructure providers. The question is whether the portfolio is managed actively or just renewed reflexively each year. Active management produces lower costs, better service, and fewer surprises.
Before You Start
You should have a vendor inventory: every active subscription, what it costs, when it renews, who owns it internally, and what it does. If you do not, the first task is producing one — managing a portfolio you cannot list is impossible. A spreadsheet with vendor name, contract value, renewal date, owner, and notes is enough.
You should also have a rough sense of which vendors are critical (the business stops if they go down), important (significant disruption but workable), or peripheral (replaceable, low impact). The management intensity scales with the criticality.
Categorise Vendors by Criticality
Not every vendor relationship deserves the same attention. A useful three-tier categorisation:
- Critical: the business stops if this vendor fails. Examples: primary hosting provider, payment processor, core SaaS that the team works in daily. These need active management — defined SLAs, named contacts, escalation paths, regular reviews.
- Important: significant impact if this vendor fails, but the business can keep operating for hours or days. Examples: secondary integrations, reporting tools, internal team tools. These need standard management — annual renewal review, awareness of pricing changes, contract held by a named owner.
- Peripheral: replaceable easily. Examples: one-off tools used occasionally, niche services for specific projects. These need light touch — review at renewal, do not over-invest in management overhead.
The mistake is to manage every vendor with the same intensity. A serious vendor management process applied to a £15/month tool is wasted effort; the same process under-applied to a £15,000/year hosting contract is negligence. Match intensity to criticality.
Establish Real Contact Points
For critical and important vendors, the difference between a relationship and a subscription is named contacts. The vendor should have a named account manager you can reach, and you should have a named internal owner who manages the relationship.
The named contacts matter most when something goes wrong. A support ticket that goes into the general queue is treated as a general ticket. The same issue raised to your named account manager with the right context tends to move faster. This is not about preferential treatment; it is about not starting from scratch with someone unfamiliar with your account when the stakes are high.
For the largest contracts — typically anything over £20,000 annual value — quarterly check-ins with the account manager are worth the time. The check-in is short. Status of the relationship, anything they should know about your business that might affect usage, anything you should know about their roadmap. This is the conversation that surfaces upcoming features, pricing changes, and product directions before they land in your inbox as surprises.
A real example. A client’s annual hosting contract review surfaced — only because they asked — that the provider was deprecating their current product line in 18 months, with a more expensive replacement. Without the conversation, the migration would have been a fire drill in month 17. With it, the migration was planned over three months and ended up on a competing provider with better pricing.
Track What You Need at Renewal
For each major vendor, the renewal is the moment of leverage. The negotiation happens once a year (or less), and it depends entirely on what you bring to the table. The information you need:
- Usage data: are you using all of what you are paying for? Many subscriptions are oversized — paying for 50 user seats but actively using 22, paying for a tier that includes features you do not use.
- Performance data: has the vendor met their stated SLAs? Any incidents in the last year that affected the business?
- Market alternatives: what does the competing landscape look like? Even if you have no intention to switch, knowing what comparable products cost is leverage.
- Strategic position: is this vendor becoming more important to the business or less? More important means you have more to lose from a poor deal; less important means you have more freedom to push back.
Take this information into the renewal conversation. Vendors generally have flexibility on pricing, terms, and feature inclusion when there is a real conversation. A renewal that consists of “we will continue paying what we paid last year” leaves money and concessions on the table.
Define SLAs and Track Against Them
For critical vendors, the contract should include service level commitments — uptime targets, response time targets for support, escalation paths for serious issues. These need to be in writing, not assumed.
Common SLA elements: uptime target (99.5%, 99.9%, 99.95% depending on tier), planned maintenance windows (when and how often), incident communication (how quickly you are notified of a major outage), service credits (what compensation you get if SLAs are missed).
The SLA matters less when everything is going well and more when something goes wrong. A vendor that has missed their SLAs three quarters running is in a different conversation at renewal than one that has consistently met them. Tracking gives you the data to have that conversation.
Service credits are usually modest — a percentage off the next month’s bill, capped at the monthly fee. They are not financial compensation for the business impact; the real value is the principle that the vendor has committed to a service level and can be held to it.
Plan for Vendor Failure
Critical vendors fail occasionally. Hosting providers have outages, SaaS products get acquired and changed, third-party APIs deprecate endpoints. The question is not whether — it is what you do when it happens.
The patterns that help:
- Know your dependencies clearly. Which workflows depend on which vendors? If one goes down, what stops?
- Have a fallback for the most critical. Not a fully parallel implementation — that is usually too expensive — but enough that the business can operate for a defined period without the vendor. A read-only export, a manual workflow, a backup tool you can spin up.
- Know the migration cost. For each critical vendor, what would it take to switch? Days, weeks, months? The answer informs how much you can push back at renewal.
- Watch the vendor’s signals. Layoffs, acquisitions, executive departures, product direction changes. None of these guarantee a problem; all of them are signals to pay attention.
A concrete example. A client’s customer support tool was acquired by a larger vendor. The acquisition itself was fine; the migration to the new product was painful — fewer features, higher prices, longer support response times. The team had been watching the signals (decreasing release frequency, account manager leaving) and had started evaluating alternatives six months before the acquisition closed. By the time the new pricing landed, the team had already prototyped the replacement and was able to migrate in a controlled window.
Avoid Lock-In Where You Can
Vendor lock-in is not always avoidable, but the degree matters. The dimensions of lock-in:
- Data lock-in: can you export your data in a usable format? How easily? With what completeness?
- Process lock-in: how much of your team’s workflow is built around this vendor’s specific approach?
- Integration lock-in: how many other systems are connected to this vendor through their proprietary interfaces?
- Contractual lock-in: multi-year contracts with cancellation penalties
Some lock-in is the cost of getting value. A CRM that perfectly fits your sales process will inevitably have process lock-in. The question is whether the lock-in is proportionate to the value, and whether you have explored the alternatives.
The hygiene that helps: periodic exports of your data in portable formats (CSV, JSON), documentation of how the vendor is used in your business so the institutional knowledge is not trapped in screenshots in a Slack channel, and an honest assessment at each renewal of how dependent you have become and whether that dependency is paying back.
Common Mistakes
- Reflexive renewal. Auto-renewing a contract you have not reviewed is paying for what you used to need rather than what you need now.
- No named contacts. The vendor is a faceless queue; the response time when something matters is whatever the queue gives you.
- Treating every vendor with the same intensity. Wastes effort on peripheral vendors and under-manages critical ones.
- Skipping the renewal negotiation. Vendors expect to negotiate. Coming to the table with usage data and market alternatives produces better terms.
- No fallback plan for critical vendors. When the vendor fails, the business stops. Even a modest fallback — a manual process, a backup tool — is worth the planning time.
- Ignoring vendor signals. Layoffs, acquisitions, and executive churn are leading indicators. Watch them.
- Lock-in by drift. Each integration added is a step deeper. Periodic honest assessment of how dependent you have become prevents the surprise.
What Good Looks Like
A well-managed software vendor portfolio has every active subscription inventoried, categorised by criticality, with named internal owners and (for critical vendors) named external contacts. Renewals are approached with usage data, performance data, and market alternatives. SLAs are tracked and reviewed. Critical vendors have fallback plans that have been at least sketched out. The portfolio cost is reviewed quarterly and renegotiated annually. Unexpected vendor changes — acquisitions, deprecations, pricing shifts — are surprises that the team responds to in a planned window, not crises that derail a quarter.
Next Steps
If the broader operational discipline is not yet in place, How to Run a Software Review Quarterly covers the cadence that surrounds vendor management. If the conversation is about how much vendor dependency to accept versus building in-house, Choosing Between Zapier and Custom Integration covers one important sub-case. For structured ongoing partnership with one technical vendor (us), see Software Support Retainers.