How a Global Financial Firm Runs a 1,600-Award Program, Year After Year
One recognition program on Awards.com has run every year since 2008. It honors about 1,600 people annually, across a global footprint, and in eighteen years it has never skipped a cycle.
It belongs to the US VP of a global financial services firm, and the recipients are not employees. They are clients, recognized each year for milestones in their relationship with the firm. That one difference shapes almost everything about how the program runs, from who approves the list to how a single misspelled name gets handled.
I asked him how a program that size holds together year after year. His answers are below, largely in his own words, because the parts worth reading are the ones a vendor would never think to say. They come down to four things: partnership between the teams, the automation behind the list, communication when something goes wrong, and the proof process. He spends the most care on the last of these, so that is where this ends up too.
Partnership: A Program Is a Relationship Between Teams, Not a Purchase
Ask him what a lasting program depends on and he starts with people, not software.
Build and sustain strong relationships with every team that touches the program, from data owners to Sales, Legal, Operations, and the external vendor. A repeatable, year-over-year program depends on a professional working environment grounded in clear expectations, consistent follow-through, and transparency. Just as important, take time to recognize the effort each team contributes at key milestones, for example winner list approval, order placement, delivery completion, to reinforce engagement and make future cycles smoother.
Two things stand out. The first is who he counts as a team. Sales is on the list, but so are Legal, Operations, and the data owners, and so is Awards.com. He treats the vendor as one of the teams that runs the program, not a supplier he hands a file to. A program at this scale touches contracts, records, and fulfillment, and any one of those can stall a cycle if the relationship behind it is thin.
The second is the part almost nobody says out loud. He recognizes the internal teams at the milestones that matter, and he names them: winner list approval, order placement, delivery. A recognition program that never recognizes the people running it tends not to last. The work is real, it repeats every year, and the people doing it notice whether it is seen.
Automation: The Winner List Is Where Programs Break
The winner list is the single point where a large program is most likely to fail, and it is the thing he automates first.
Automation is essential to reduce manual workload, improve accuracy, and enable the team to scale. It should support both the creation of the winner list, with consistent eligibility logic, validation checks and version control, and end-to-end tracking of each award, from initial record creation to correct personalization details to ordering, shipping, and delivery confirmation. The goal is to minimize rework, reduce errors, and ensure every winner can be tracked with clear ownership and status at any point in the process.
Read it closely and there are two separate jobs in there, not one. The first is building the list: eligibility logic that stays consistent year to year, validation checks, and version control so there is one true copy instead of five. The second is tracking each individual award from the moment its record is created through personalization, ordering, shipping, and delivery confirmation.
Most programs automate the first and leave the second to email and memory. At 1,600 awards, that is how pieces go missing or ship with the wrong plate. His standard is that every winner can be located, with a clear owner and a current status, at any point in the process. That is a higher bar than a clean spreadsheet, and it is the one that keeps a cycle from unraveling in the final week.
Communication: Escalate Early, Especially the Mistakes
He is unusually direct about surfacing problems, and specific about why.
Maintain a clear, consistent communication structure that highlights progress but also escalates issues early, especially mistakes or data discrepancies that could create confusion. With around 1,600 clients awarded globally each year, even small errors can have outsized impact and often require the list to be pulled back for review. A disciplined communication cadence, a defined escalation path, and a single source of truth for updates are critical to keep all teams aligned and to prevent late-stage surprises.
The detail worth sitting with is that a single data discrepancy can force the whole list to be pulled back for review. At thirty recipients, one wrong name is a quick fix. At 1,600, a small error rate stops being a rounding issue and becomes a re-run, and a re-run late in the cycle is expensive in the one currency nobody has at that point, which is time. So the discipline is to raise the small things early and keep one source of truth for status, with a known path for every problem to the person who needs to hear it. The goal is not to avoid mistakes, which is impossible at this volume. It is to catch them while they are still small.
The Proof Process: Approve What's Right, Fix What Isn't, Skip the Blame
This is the part of the interview that does not turn up anywhere else, and it is the part he has clearly thought about the most.
The proof process can be challenging with so many recipients and so many specific recognition KPIs. Garbage in, garbage out. It's not worth complaining during the proof process about where the bad personalization came from, whether it was introduced while Awards.com was setting up the art proof or if we gave them bad data on a spreadsheet. The important thing is simply to approve all that is correct, and provide quick corrections for those that are not.
There is a lot packed into a short answer. Start with the phrase he borrows, garbage in, garbage out. With 1,600 recipients and a set of specific recognition criteria behind each one, the proof is only ever as good as the data that fed it, and at this scale the data is never perfect.
What makes the advice credible is what he does next. He allows, plainly, that a bad personalization could have come from either side. It might have been introduced while Awards.com was building the art proof, or it might have been sitting in the spreadsheet his team handed over. He does not spend the proof round trying to prove which. His rule is simple: approve everything that is correct, and send quick, precise corrections for everything that is not.
That even-handedness is not just good manners. It is the faster path. Arguing about the origin of an error during proofing does not fix the proof, and it burns the time the correction needs. A vendor who feels blamed slows down; a team that feels blamed gets defensive. Approving the good and fixing the bad, without the detour through fault, is how a proof round of this size actually closes on schedule.
What This Looks Like on Our Side
From where we sit, two things make a program like this repeatable, and he names both.
Each year we select a piece that remains within the scoped budget and stays on brand. Awards.com offers such a wide selection that fulfilling both those requirements is never an issue. Inventory is not an issue either since they manufacture the inventory for the program when I have my selection and estimated quantity in advance of the order and personalization.
The selection has to clear two bars at once, budget and brand, and it has to clear them every year. Awards.com carries enough range to make that routine, from crystal awards to deal toys that mark a closed transaction and sales awards for incentive programs. Because the quantity is known ahead of time, the pieces are made to order for the program rather than pulled from a shelf, which is why supply is not a variable he has to manage. That is most of what our end contributes: a deep catalog of recognition awards that stays on brand inside a set budget, and manufacturing that starts once the count is confirmed.
If Your Program Is 30 People, Not 1,600
Most recognition programs are not 1,600 people. They are twenty, or forty, run by one person who also has another job. The scale is different, but the four ideas translate down cleanly.
Partnership becomes knowing who owns the data before you need it. You may not have Legal in the loop, but someone controls the list of who qualifies, and finding that out in January is easier than finding it out the week of the event.
Automation becomes one spreadsheet with a single owner and real version control, instead of five people editing copies and emailing them around. You do not need a system. You need one file that everyone agrees is the file.
Communication becomes telling people early when a name is wrong, rather than hoping it resolves itself. It rarely does.
And here is the honest part the big program cannot claim. At thirty recipients, you can read every proof line by line, and you should. The VP's team cannot check 1,600 awards by eye, which is exactly why they lean on validation logic and disciplined proofing. You can, so use it. It is the one real advantage a small program has. If you are building a program from scratch, our guide to launching a recognition program walks through the first cycle.
Running a Recognition Program at Scale FAQ
How do you keep a recognition program running year after year?
Treat it as a relationship between teams rather than an annual purchase. The programs that last keep clear expectations and consistent follow-through with everyone who touches the list, from data owners to fulfillment, and they recognize the internal teams at key milestones so the people running it stay engaged.
What is the hardest part of running a large recognition program?
The winner list and the proof process, both because of volume. A small data error that would be trivial at thirty recipients can force a list of 1,600 to be pulled back for review, so most of the effort goes into building the list carefully and catching problems while they are still small.
How far in advance should you order awards for a large program?
As early as you can confirm your selection and estimated quantity. For a large program the awards are made to order once the count is known, so giving the manufacturer your piece and quantity ahead of the order and personalization keeps supply from becoming a constraint.
How do you handle personalization errors across hundreds of awards?
Approve everything that is correct and send quick corrections for everything that is not, without spending the proof round arguing about where the error came from. At high volume the error could originate on either side, and sorting out fault during proofing only costs the time the correction needs.
Does this apply to smaller recognition programs?
Yes. The same ideas scale down: know who owns the data, keep one master file with version control, flag wrong names early, and check proofs closely. A smaller program has one advantage a large one does not, which is that you can read every proof line by line.