How to spend your Microsoft Co-op in FY27 (without wasting a penny of it)
What's changed, why it's changed, and a long list of practical things you're actually allowed to spend the money on.
Jump to a section
Every year, partners hand back millions.
Every year, thousands of partners hand back millions of pounds worth of Co-op funding. Not because they don't want it, but because they get to the end of the usage period, look at what's sat there untouched, and still haven't got a clear answer on what to spend it on, or the types of activities that will even be approved by Microsoft.
This year, that's a more expensive mistake than usual. Microsoft has tightened the rules for FY27, and a lot of the things partners have leaned on for years have quietly disappeared off the list.
So, this guide does two things. It explains what's changed and why, and then it gives you a long list of practical, permitted things to actually spend the money on. There's a checklist at the end you can actually take back into your business and use to hit the ground running.
Why Microsoft has tightened the screws.
Let's be honest about what's happened here. Co-op has been used in some creative ways over the years. Partners have used loopholes to recoup funds through sales incentives, and plenty of it has been spent on things that never generated a single conversation with a new customer.
The reality is that these Co-op funds were never there to line your pockets. Their primary purpose is to help you grow your practice and support Microsoft's growth. And now, Microsoft has rewritten the rules to make sure that's what happens.
There's a bigger shift underneath it too. For years, partners were rewarded for hanging onto business they already had. But when a customer is on Microsoft 365, or Azure, or Dynamics, or Copilot, they tend to stay - they like it, it works, and there isn't a lot out there they'd easily move to. Unless they get acquired or merge with someone and a technology conversation opens up, they're staying put.
So, paying a partner to maintain that base was never great value - and in some ways, it rewarded complacency. Microsoft has now said as much, and FY27 is where it shows up in the rules. Grow your Microsoft business, and you'll be rewarded. Renew it, and you won't be. At least not in the same way.
The clearest signal of all? Paid media is now a hard requirement for multi-touch campaigns. If that's not Microsoft telling you to go after net new, I don't have a clue what is.
The allocated resource cap - your funded heads - stays at 50% for FY27, but Microsoft has confirmed it's expected to come down in FY28 and told partners to plan accordingly. They haven't published a number, but my gut says somewhere around 25-30%, which would be a serious dent if you've built your marketing function around it. Don't say you weren't warned.
The rules of the game, in plain English.
Before the ideas, here's what you're working within.
Paid media is required
Any multi-touch campaign claim needs a paid element. Not optional, not nice to have.
Two further activities, minimum
You need at least two more activities alongside the paid media, from an approved list.
One joined-up campaign
It has to run as a single campaign, not a pile of separate tactics that happen to be running at the same time.
Evidence, all the way through
Impressions, clicks, leads, downloads. All tracked, all claimable.
Co-op is earned and used in six-month halves, so what you earned in the first half of this calendar year is what you're spending between now and December.
Please don't spread it thin.
This is the bit that matters most, so I'm putting it before the fun stuff.
Whether you've got $20,000, $100,000 or $200,000, the temptation is to split it ten ways, so everyone gets a go. Don't do that. Thin budgets spread across lots of small things produce lots of small nothing.
Start this financial year in a strategic mindset instead; treat your total Co-op pot as one sustained programme: launch in the first half, then use what you learn to optimise, expand, cut what isn't working and test the next hypothesis in the second.
Because here's the thing. Every partner I've spoken to over the years - and that's hundreds of them, all over the world - wants to bring in the big deals. You don't land those with a four-week burst of activity, and you certainly don't land them with a bit of PPC. You land them by showing up consistently, across multiple channels, for long enough that people start to recognise you.
higher future earnings from Microsoft business for partners investing in paid media with multi-touch campaigns, according to Microsoft's own analysis.
of partners under-invest in demand generation in the first place, according to Omdia.
So, the money is telling you where to go. Follow it.
Pick a hero campaign theme to run with.
Every good campaign needs one meaty thing at the centre of it. A research study, a white paper, an e-book, a proper guide. Something with actual substance that sets the theme for everything else.
This is where you'll happily burn a decent chunk of budget, and it's money well spent. It builds credibility, it gives you something to say that nobody else is saying, and it gives you the overarching story for ten months of activity.
Keep it to 12-16 pages. Not 200. Nobody reads 200 pages - the Microsoft incentives guide itself is proof of that.
The research study blueprint
If you want hero content that packs a punch, there's no greater way than with primary research. Here's how it works, using accountants as the worked example.
Survey your niche
Survey somewhere between 150 and 250 respondents in your target market. You can either do this yourself, which takes time, or you can use a credible survey house, which comes at a cost but can get quality data back to you quickly. It's a trade-off worth making. With a survey house, the cost goes up with both the number of questions and how niche the audience is, so 10 to 20 questions is a sensible place to land.
Ask about what you're good at
AI readiness, security posture, where they are on cloud - whatever your proposition actually solves. You know your strengths better than anyone, so lean into them and ask open questions.
Analyse the data properly
Map the answers back to real customer challenges and pull out the priorities, the pain points, the things that made you sit up.
Find your theme, and be creative with it
If 90% of respondents say AI is a priority, please don't make your headline "AI is a priority for accountants". It's lazy, boring, and every other partner has more than likely already had a go at bringing that to life. It's also something that ChatGPT can probably regurgitate. If you've invested in research, invest the time to come up with headlines that pack a punch. Find the angle underneath it, give it a bit of bite, and make it something people remember. People buy from people, not from a logo and a stock photo.
Slice and dice to get your message everywhere
A hosted report on your website, blogs pulling on individual findings, webinars unpacking the data, emails, social carousels, ads, roundtable topics, event giveaways, and a press release into the trade titles your audience actually reads.
A quality campaign, compliant with Co-op guidelines
Paid media is used to promote the report, a website landing page hosts it, and a run of blogs off the back of it give you the evidence you need to claim back without any hassle - the paid media requirement plus two supporting activities, ticked without even trying.
Got a bigger budget? Rinse and repeat by vertical
Focus on accountants this half, financial services or retail next. Just don't run three at once and stretch yourself thin - that's the mistake too many partners make.
Not a research person? Other hero plays
Research isn't for everyone, and that's fine. A webinar series works well too - fortnightly or monthly, with your experts front and centre.
And if you haven't got the expertise in-house, or you're a bit shy about hosting, bring someone in. The likes of Illuminate Learning will run the session for you while you handle the promotion. When it comes to Copilot, I'd trust the people doing it every day over having a go yourself and hoping for the best.
Want to drive sign-ups? Eventbrite and similar work well for the S in SMB. They won't work for the top end, and they definitely won't work for an enterprise audience - be honest with yourself about who you're actually trying to reach.
One meaty piece of hero content, promoted properly for ten months, will beat ten small things nobody remembers. That's the whole game.
Nathan Selby, Founder, ResultfulChoosing your channels.
Still the most reliable way to get in front of a specific person. It's not cheap - eight, nine, ten pounds a click is normal, and in competitive periods we've seen it push eighteen. Not often, but it happens. Sharpen your targeting and your message, drive a better click-through rate, and that number comes down.
Google Ads and Microsoft Ads
Reactive channels. Someone's already searching, which means they're already further down the process. Useful, but don't pour budget into them expecting to build awareness - that's not what they're for. Think about intent. "What is Cowork" is somebody exploring. "Microsoft Cowork partner" is somebody looking for help to onboard it and get value out of it. Two completely different people, two completely different journeys, and only one of them is close to buying.
So, use LinkedIn and email to drive the educational content, and keep search engines for the further-down-the-funnel stuff.
Don't forget about the full funnel
The marketing funnel is anchored around three key stages.
Top of funnel
Education and problem awareness - e-books, blogs, white papers. Helping someone realise they've got a problem at all. They're not searching for a solution yet, they're searching for "what is this thing I'm dealing with, and has anyone else dealt with it?"
Middle of funnel
Where you help customers consider if this solution is right for their challenge - webinars, battle cards, demos, free trials. They know they've got a problem, and now you're helping them see that Copilot, Cowork, Azure, Fabric or Business Central might be the answer.
Bottom of funnel
The decision-making or conversion stage. Somebody searching "Business Central partner" has already been through the what-is, the versus-QuickBooks and the versus-NetSuite. They want someone to do the job. Shorter cycle, far less nurturing, much better conversion rates, because they've done the solution qualification already.
What you can actually spend it on.
Microsoft's own recommended split for a non-distributor partner is 50% demand generation, 40% market development and 10% partner capability. On $50,000 of earned co-op, that's $25,000, $20,000 and $5,000. Distributors get a fourth bucket for reseller activation, which takes the biggest share.
Demand generation (50%)
- Paid media with multi-touch campaigns - the big one, and the one with the best return
- Customer events and conferences
- Microsoft first party events
- Targeted prospecting
- Promotional branded merchandise, capped at 1%
Market development (40%)
- Allocated resource for sales and marketing, capped at 50% - and coming down next year
- Customer workshops - demos, briefings, hands-on labs, hackathons
- Customer solution adoption, capped at 5%
- Solution co-innovation
- Internal incentives and SPIFFs, each capped at 2%
Partner capability (10%)
- Internal partner skilling
- Partner programme audit fees and certifications
- Partner AI adoption, capped at 5%
What counts as your two supporting activities
Alongside paid media, pick at least two of these: a landing page on your website, email or direct mail, blogs and videos, SEO or LLMO or GEO work, webinars, or customer case studies. Organic social counts as long as it's genuinely part of the campaign - localisation and boosting are the claimable bits.
The stuff that'll trip you up
Worth reading twice, because these are the ones that get claims bounced.
No travel, meals or mileage unless a specific activity explicitly allows it.
No partial claims, and no claiming the same expense across two programmes.
Events have to be clearly educational. Stadiums, sports suites and entertainment venues are out.
You can't claim advertising placement on your own website.
Print and postage aren't claimable under a multi-touch campaign.
Creative and printing on merch - bags, T-shirts, golf balls - is out.
New for FY27: you can't use co-op on Partner of the Year Award nominations or the assets that go with them.
Prove it, or lose it.
Microsoft has sharpened up the evidence side this year, and it's the bit most partners underestimate.
A detailed third-party invoice is now the preferred route, and it has to be detailed enough that someone who knows nothing about your campaign can see what was done, when it happened and where. Vague agency invoices won't cut it anymore.
If you haven't got that, you're onto the certification form with every expense itemised as its own line. For a multi-touch campaign that means at least three lines: a screenshot of the paid media, plus two supporting pieces, plus a platform report showing your impressions, clicks or leads across a date range.
So set your tracking up on day one, not the week before the claim deadline.
Two things nobody's talking about yet
Microsoft now recommends putting 10-20% of your co-op into global or centrally developed campaigns, so you get repeatable, scalable go-to-market while keeping local relevance. If your market generates less than $50,000 in earned co-op, they suggest keeping the money local and using the central assets instead. It's a recommendation this year. It becomes a requirement next year, so start building the habit now.
Your co-op marketing plan needs to be developed with your Partner Marketing Manager where you've got one, and they need to review and sign it off as part of your business plan. That's not a rubber stamp at the end - it means having the conversation before you commit the budget.
Where to start this week.
- 01
Check what you've actually earned in Partner Center, and whether you clear the minimum threshold.
- 02
Pick one audience. One. The niche one you keep talking yourself out of.
- 03
Choose your hero piece and get it briefed before the 16 August refresh.
- 04
Map your paid media plus your two supporting activities.
- 05
Sort your tracking and your invoice detail now, so the claim is boring when it comes.
- 06
If you have one, get your plan in front of your Partner Marketing Manager. If you don't, we can advise whether it's in line with the new guidance - just reach out.
Do that and you'll get to next June with a campaign that built something, rather than a pot of money you never quite worked out how to use.
And if you just want to chew through some ideas over a 15-minute chat before the refresh lands, with no obligation at all, don't hesitate to get in touch.
Treat your Co-op pot as one sustained programme, not ten smaller campaigns or activities. That single decision is usually the difference between a campaign that builds something and a balance you hand back.
Nathan Selby, Founder, ResultfulCommon questions about running Co-op as one programme.
- What does it mean to treat a Co-op pot as one sustained programme?
- It means committing your full Co-op balance to a single campaign theme and audience across the year, rather than splitting it across lots of unconnected activities. You launch in the first half, learn from the results, then use the second half to expand what works and cut what doesn't.
- Why is one sustained programme better than several smaller campaigns?
- Smaller campaigns rarely run long enough to build recognition or generate enough data to optimise. A sustained programme keeps the same message in front of the same audience long enough to move them, and it gives you a single set of numbers to report against when you claim.
- How many supporting activities should sit around the hero campaign?
- Pick one hero piece and plan at least two or more supporting activities around it, such as paid media, email nurture, events or content syndication. Each supporting activity should point back to the hero piece rather than run as a standalone effort.
- Does a sustained programme make Co-op claims harder to evidence?
- It usually makes them easier. One programme means one tracking setup, one set of creative assets and one invoice trail, so the proof of execution you submit is consistent instead of fragmented across many small activities.
- What if part of the programme isn't working halfway through the year?
- That's the point of the mid-year review. Keep the theme and audience, but reallocate budget away from the channels that aren't converting and into the ones that are. Changing the channel mix is fine; changing the theme every quarter is what wastes the pot.
- How much Co-op should be held back for the second half?
- As a rule of thumb, commit enough in the first half to get a real read on performance, then keep a meaningful reserve to double down. Just don't leave the reserve so late that you can't spend and claim it before the deadline.
Want more FY27 Co-op ideas?
Drop your work email and we'll send new FY27 Co-op ideas, guidance and campaign examples as we publish them.