March 16th, 2017

The Sales Tax Audit Struggle Part Two: It's Not House Money

In this two-part series, we're talking about why we eschewed the finder's fee model for sales tax audit services. In Part One, we explained that we wanted to avoid situations where our clients felt like we had a conflict of interest when it came to sales tax auditing.

First and foremost, ZacTax was built as an analytical platform. In our day-to-day lives, we had desperate need for a better way to analyze sales tax data. What we had was a quarterly report provided as a by-product for using a sales tax audit service. It wasn't timely, it was paper-based, and it was difficult to do any real analysis with.

Since it wasn't part of Zac's core mission, we struggled with whether to provide auditing services. The debate centered around a) whether we should offer auditing services at all, and b) if we did, whether or not it should be self-service.

The truth is, there is big money to be made in sales tax auditing. Open data is an amazing thing; all you have to do is dig through a few cities' check registers to find payments of 5 and 6 figures annually to various audit firms.

The prevailing business model in the sales tax audit world is the finder's fee, in which an audit firm takes a cut of whatever they find. On the surface, this sounds like a good deal for both parties. The city gets new revenue, and the firm gets paid with money the city didn't have anyway.

When it came to audit services, we knew we could build a better mousetrap. The process can be largely automated. Sprinkle in some machine learning to make the automation smarter over time and a little elbow grease, and we could make it rain like

Remote Image

It's just not that hard

The problem is, it's just not that hard to audit your sales tax. We routinely find missing taxpayers while demoing our audit tool to new clients. During a recent demo, we helped find a missing taxpayer worth nearly $15,000 a year in sales tax. Under a finder's fee model, that would have netted us $13,500 for about 5 minutes worth of work.

Not a bad haul. And the city gets $31,500 in one-time revenue plus $15,000 a year in its sales tax base. So why is this such a bad thing?

It's real money!

Because even though it's money you didn't have, it's still real money. It could be used for fixing roads, repairing sidewalks so kids can walk to school, replacing a broken slide in a pocket park, updating the media selection at a library, or lowering a property tax rate.

In other words, even though it seems like you're working with house money, it's still a real expense that your taxpayers are bearing in the form of foregoing other essential services.

Breaking down the effort required

Let's think about how much we're actually paying for audit services, not in terms of some percentage of revenue we didn't have, but in terms of the effective hourly rate it costs to get that revenue.

In the example above, our effective rate would be $162,000 per hour. One might protest that this is an outlier, and that would be fair. So let's conduct a little thought experiment together instead.

We can only speak to how easy our audit tool is to use, so for this example we'll consider how much time would be required if we provided audit services using the patent pending tool included in ZacTax. The estimates below are based on our experience using it to help our clients audit their sales taxes. It's possible that other audit firms have less automated workflows that require significantly more manual labor.

Prior to demoing a new client, we set up the entire audit platform for them. Once the automated parts have run, we'll often spend about 30 minutes doing some QA/QC and getting familiar with the businesses in their city.

To fully audit an average city of about 25,000, we would probably spend about 20-30 hours on an initial pass. If we were doing it ourselves, it would involve spending a few hours going through the data in the office and then taking an iPad into the field for follow ups. Once the initial audit is completed, we would probably need to spend less than an hour a month making sure nothing new pops out.

Let's just say we spend 40 hours auditing this city during the first year, and are able to find $50,000 worth of annual taxes. At a 30% cut, our fee would be $45,000, which equates to $1,125 an hour for our efforts.

We can't find any justification to charge that much. And, truthfully, would you accept a contract that charged that much if it were phrased in this way?

Why does it matter?

It's not our goal to put audit firms out of business. We hold data integrity as a very high value, and ensuring that tax revenue is being sent to the right place is of great importance. They do good work.

But if we can disrupt the business model by getting cities to think about these finder's fees as real money, we're perfectly fine with that. Our disagreement lies not in the act of auditing sales taxes, but in the belief that negotiating with vendors to go from a 30% fee to a 28% fee is wrongheaded at a fundamental level.

We chose to provide an audit tool because we thought we could do it better. We chose **not** to take our piece of the (very large) pie, but to try to shrink the pie and keep that money where it belongs: in your city's revenue account. That doesn't make us better people, it's just what *we* thought was the right thing to do.

When you evaluate an audit service, consider the effort that's actually required and the effective cost for the work that's being done. If you're haggling for a few percentage points off of the finder's fee, it might be time to step back and re-evaluate things. And remember that these fees are not paid with house money. They're paid with real revenue that you should have been receiving all along.