Let’s say your company has suddenly begun to lose sales or your website is simply not performing up to expectations.

You are the one put in charge of fixing the problem by bringing in outside help. How do you ensure that the web analytics firm you hire is going to give you value for money AND cure your conversion headaches?

There’s a world of difference among the various web analytics providers and what they provide. A full-service web analytics company will use your traffic statistics to help you locate the source of any problems and provide ongoing solutions.

So before you sign on the dotted line with any analytics company, the first two steps are to determine what you want them to do and how much you can afford to spend to achieve that.

Set Your Expectations

First, decide what specific services you want and what results you expect. List the metrics you want tracked, analyzed and monitored. If you don’t know what certain metrics mean or how to interpret them, you’ll need customer support as part of the package. Will you need round-the-clock support by phone or would email support be sufficient?

Does the analytics company need to train your team in conversion tracking so they can understand and participate in the analytics process? Do you also want website usability consulting? What do you hope to achieve with your web analytics investment? All these questions should be addressed before you begin the search.

Your Web Analytics Budget and Costs

Next, you need to tackle the budget issue. Let’s put things in perspective with an example. If your company spends $1,000 a day on PPC advertising that results in $5,000 a day in sales, that’s advertising money well spent, right? However, what if your advertising costs remained steady but for some reason your daily revenues started dropping and wound up below $3,000. Your company is now losing $2,000 in sales a day and you have no idea why.

Maybe click fraud is the culprit or someone in marketing tried to “improve” the sales copy. Or it could be your competitors nipping at your heels. In most cases, the only way to solve the mystery is to bring in a web analytics professional. Their fees can run anywhere from $___ a day to $___ a day. But if they do their job well, the payoff for you means regaining $2000 a day (or more) in sales.

Here it’s important to note that some web analysts simply provide statistical reports that you must analyze yourself. This is the cheaper solution. Others interpret the stats plus help you apply the necessary changes to improve your ebusiness’s performance. You will have to pay more for this sort of professional consulting but you’ll also get more.

Bottom line: if their fee buys you a solution that helps you find the culprit in a matter of hours (not days or weeks), brings sales back up to pre-crisis levels, and then ensures you maintain your solid position in the market, you have a strong case for hiring Firm X.

You’ll have an even stronger case if you can back it up with numbers.

ROI Calculation

Now you need to estimate the incremental cash flow that can reasonably be expected from a web analytics program used over time. There are two key components that lead to increased ROI:

1. The direct increase in sales resulting from improved traffic quality and website usability.

2. The savings resulting from discontinuing non-performing advertising and preventing click fraud.

ROI is defined as the ratio of gains to cost, usually expressed as a percentage:

ROI = (Gains – Costs)/Costs

Consider the following hypothetical example. You can easily insert your numbers in place of ours and see how your potential ROI turns out. First, we need to collect the following data:

1. Site information:

A: Monthly sales ($100,000)

B: Monthly ad budget ($15,000)

2. Sales increase resulting from improved…

C: Traffic quality (2%)

D: Website usability (2%)

3. Savings resulting from…

E: Discontinuing non-performing ads (10%)

F: Detecting click fraud (10%)

4. Web analytics fees:

G: Monthly investment ($1,000)

Now, insert that data into the formula:

Formula: ROI = (A*(C+D) + B*(E+F) – G)/G

Formula with sample figures inserted: ROI = ($100,000*(0.02 + 0.02) + $15,000*(0.10 + 0.10) – $1,000)/$1,000 = ($2,000 + $1,500 – $1,000)/$1,000 = 600% ROI

This example uses a conservative estimate for the improvements that you can reasonably expect from a consistent web analytics deployment. In addition, chances are your monthly costs for a web analytics solution will be much lower than $1,000. If so, your ROI would be even greater.

You can use the formula above to project your own ROI or feel free to download our handy ROI worksheet in Microsoft Excel.

Next… narrowing down the list of prospective web analytics providers and selecting a winner.

Web Analytics Vendor Selection Checklist

Apply the following checklist to each web analytics firm in the running and it will help you zero in on the best one:

1. Only the numbers you need. Don’t be impressed by long lists of reports that some vendors think will turn your head. In this case, less is more: you only need a few report types, but they have to be the right ones.You also need to consider the amount of manual work your team will have to do to generate reports not provided by the vendor. Very important: Ask the vendor if they provide just raw data or if they help interpret and apply the data they supply.

2. Click fraud detection. Fraud artists who deliberately click on PPC ads with no intention of buying can seriously undermine your SEM and PPC marketing efforts. If you suspect click fraud is an issue for your online business, ask the web analytics vendor if they provide click fraud detection and auditing services.

3. Drill down capabilities. You need to be able to “slice and dice” the data, for example, compare average revenue per visitor originating from Google AdWords versus revenue generated by MSN’s organic search results, or compare conversions generated by different landing pages.

4. Checkout tracking. Will your prospective web analytics firm allow you to track visitors each step along the way of the checkout process, view drop-off rates throughout the process, and identify problem areas and exit points? They should.

5. Conversion tracking. You want to be able to measure specific conversions, such as filled-out inquiry forms, newsletter subscriptions, file downloads, and so on. Your analytics firm should also be able to tell you the traffic sources and keywords used to reach the landing page.

6. Data management. If you switch vendors down the line, will you be able to take your data with you? How big a problem will it be to export all historic account activity into XML or Excel?

7. Support and professional services. Will there be extra costs for email support and guaranteed turn-around times? What about phone support? Does the vendor have specialists on staff who can provide services such as web analytics deployment consulting or website usability improvement?

8. Vendor and product longevity. How long has the vendor been in business? Do they seem competitive? Do they have a track record of innovation? Evaluate the vendor’s ability to stay abreast of industry trends and to adapt to the rapidly changing competitive landscape.

Summary

Following these basic steps before hiring a web analytics vendor will help ensure you get the best possible return on your investment in a web analytics package:

1. Set your expectations.

2. Determine your budget.

3. Calculate a reasonable return on investment.

4. Subject each candidate firm to the vendor selection checklist.

One final note: A successful web analytics deployment begins not with the vendor but with you and your company. To ensure success, name a web analytics point person within your firm who can act in a leadership role to implement the changes recommended by the vendor. That will bring your project full circle.

Dmitri Eroshenko is CEO of Clicklab, a leading web analytics service with custom Key Performance Indicators (KPIs), click fraud detection, and usability testing tools geared to improve your online marketing ROI and conversion rates.