Mail acquisition cost savings tips

Now that you’ve driven some costs out of your mailings with postage tips from yesterday, let’s talk about driving costs out your acquisition efforts.

One of the big costs of acquisition mailings is the lists themselves.  Chances are fairly good that if a list works for you, your list works for them.  Thus, make sure you have list exchange agreements set up with these nonprofits, as you can both save each other money.

(Note: if you have a source of names that are not traditional direct mail donors, you may want to take them off of both the rental and the exchange markets.  Most donors are loyal to doing good works and perhaps less so to your organization.  However, if you had a source of donors that are loyal specifically to your organization like your most loyal volunteers, you may not want to exchange these with everyone and their brother.)

Speaking of unique sources of names, another way to save on costs are to try to “acquire” your own names.  That is, you have some lists of people who are constituents of your organization who haven’t yet donated – volunteers, advocates, white paper downloaders, service recipients, etc.  These names may not all make money on the first go-around, but you can usually acquire these at a higher response rate and/or average gift than someone off the street and there are no list costs.

While it will be an additional investment, I remember customizing your copy for that person’s status with your organization.  Indicating that you know they are a volunteer and thanking them for their service will not only cut down on your complaints; it will also very likely increase your response rate.  After all, you don’t want to do business with someone who doesn’t know who you are – why would your donors?

Even if you don’t have these names in your active acquisition efforts, be sure to keep them in your database of record.  When you rent or exchange a list with another nonprofit, it goes through a process called match back.  This is where they run your names against the incoming list – any matches are returned to the origin nonprofit and you don’t pay for them since you already had them on your file.

However, that ping from an outside list can indicate to you that the person that you hadn’t been mailing because they only downloaded one white paper three years ago is giving to other organizations and could be a quality acquisition candidate.  Thus, you can put them into acquisition at no cost.

Another source of “free” names are your lapsed donors.  You likely have a list of people whose RFM analysis (or hopefully modeling or RFM+ analysis) indicates that they shouldn’t be getting donor pieces anymore.  That’s fine and natural.  However, that doesn’t mean you should stop trying to re-acquire them.  Now they are in a limbo between donor and, as the song goes, just somebody that you used to know.

Mailing them with the same piece that attracted their attention at first can be a good way of getting them back into the swing of donating.  And it’s likely a good investment – reacquired lapsed donors tend to have better retention rates than newly acquired donors.  Just don’t take them for granted – they’ve told you with their behavior that they aren’t head over heels for you.  Once you’ve reacquired them, make sure they get your love so that they won’t lapse again.

When you do rent or exchange with outside lists, look at the many variables you can select: recency of donation, amount of donation, demographics, location, etc.  This is a case where you don’t want to scrimp on costs.  If a list is not performing for you as it had, you may want to spend more on it (counterintuitively) by asking for more recent donors, larger donors, only females (if your organization skews this way as most do), or adding a ZIP code select (asking for only donors in your top performing ZIP codes).

On the flip side, if a list is performing very well, you may be able to relax these variables, helping you get additional productive names and saving on costs.

Also, remember that just because you bought the name doesn’t mean you have to mail it.  If your analysis/modeling of an outside list name indicates that they won’t pay for themselves, don’t throw good money after bad money.  The list cost is sunk; you need not be.

It’s important you have a good relationship with your list broker.  They should know your strategy and whether you are driving for quality or quantity of donor.  Chances are if they are doing a good job for you, they will already be optimizing for many of these things, but it never hurts to ask.

Mail acquisition cost savings tips

Online-offline translation guide for acquisition

When I was an exchange student in Japan, I carried a pocket-size English-Japanese dictionary with me.  (Pocket sized to make sure that it never had quite the word I needed, causing me to resort to “large bald person’s religion’s house” when I wanted to find the Buddhist temple.  I quite possibly offended the entire nation and thus apologize here for my adolescent self.*)

Recently witnessing a conversation between two people– one an older direct mail veteran; the other, a digital native online community builder who may never have seen a piece of paper – put me in mind of those days of mistranslation and bumbling.  They never seemed to grasp that one man’s teaser copy was another woman’s pre-header (or close enough to be getting along with).  Thus, they talked past each other and went their separate ways thinking the other was an idiot, even though they seemed to my ears to be agreeing.

Thus, this week, I’d like to try for some peace, love, and understanding between the often warring nations of offline and online.  Or at least the understanding part.

We’ll start with one of the simplest areas of cultural differences: acquisition.

Those who have been weaned on online will find the offline acquisition culture strange and terrifying.  Most notably, they trade and rent acquisition lists from each other!

For those in the offline world, don’t suggest this as a tactic for your online brethren.  Not only is it illegal (sending unsolicited emails is called spam and it’s even less appetizing than its namesake), but it is culturally not done in the online world.  (Yes, the cultural taboos are even worse than the legal ones.)

Despite, or because of, these differences, however, there is a lot that each tribe can learn from the other.

For online folks, just because you can’t and shouldn’t exchange or rent lists online doesn’t mean that you can’t create mutually beneficial relationships.  You can do this through shepherded emails.  Let’s say another nonprofit has a similar constituents or issue area to you.  You might consider sending an email to your list saying, in essence, that if you like us, you might like them.  And vice versa (of course; there is no quo within the quid).

Similarly, you might try engaging your corporate partners to see if they will run a shepherded email for you to their constituents, urging them to engage with you.  This has its own built-in incentives — the for-profit looks like the valuable philanthropic member of the community they are and you reap the list building benefits.

For offline folks who think the no-list exchange or retail rules are overly puritanical, know that an opt-in model for mail is on the visible horizon.  For those in the US, our friends in Europe are facing this by virtue of EU/UK regulations.  (How politicians justify themselves being able to send mail as they wish with opt-in only for nonprofits baffles me, but I suppose that’s what happens when you write the laws yourself.)

And opting in does provide a stronger bond between you and the donor or potential donor.  Thus, you can learn from your online partners how to build that bond.  Some tips:

  • As we’ve advocated, make sure you are setting expectations for what communications a person will receive in your welcome series.
  • Make it easy for a person to change the frequency, timing, and/or nature of their communications.  One tactic smart online folks will do is have multiple lists for which someone can subscribe.  If a communication is not to the person’s liking, they can be removed from those emailings without losing a constituent.  If a person does not want (for example) premiums, they should be able to request that and have it be honored.
  • Make it easy to opt-out with clearly visible instructions.  A person who asks to be taken off of your mailing list is doing you a favor (not as much of one as they might have done, but a favor nonetheless).  They could simply let you mail away and waste your money, but instead, they are helping you save it.  Help them help you.
  • Get your list through organic means.  Online and offline content can help you build a subscriber and constituent list.  This content marketing isn’t good for just online activation — it can be used for mailing as well.

Hopefully, these will help you discuss acquisition fluently across channels.  Tomorrow, we’ll talk about the cost implications from offline and online, using fun and exciting terms like “marginal costs.”  You won’t want to miss it.

 

* Of course, if I’m apologizing for my adolescent self, we’re talking about way more people than just the entire nation of Japan…

Online-offline translation guide for acquisition

Let there be lightbox

Come, children.  Let’s gather ‘round the fire and I’ll tell you a tale of the Old Web.

Once upon a time, when you went to a site, you would be confronted with “pop-ups”.  These were new browser windows that would open when you would go to, interact with, or try to exit a site.  Sometimes, when you closed them, they would automatically open a new pop-up and so on.  

I’m told that the more ethically dubious the site, the more likely they were to have an endless loop of these.

We fought them with all of our might and hatred, until they became a casualty of the Second Browser Wars, as web browsers realized that they could make us happy by purging these from us once and for all.  Now only a few scattered pop-up survivors live among us, surviving on the barest scraps of attention.

Unfortunately, some of the stigma of pop-ups rubs off on lightboxes (or sha220px-fionaappleshadowboxerdowboxes, which I prefer (while less common) because I think of the Fiona Apple song)

You made me
A shadowboxer, baby
I wanna be ready
For what you do

People think that a light box* (slash shadow box) is just a glorified pop-up.  But when done properly, it has several advantages: it is easily closed, doesn’t cause a loop of pop-ups, and provides a quick guide to the first thing someone might want to do or know on a page.

But most of all, you should be trying light boxes because they work.  One report found that the average conversion rate for email acquisition light boxes is 1.66%.

So how do you make a lightbox work for you?  A few tips:

  • With a few exceptions, I would not suggest a donation ask lightbox.  Like we talked about yesterday, getting the first microconversion is easier and potentially more profitable that a straight ask on the home page.
  • Those few exceptions are:
    • End-of-year fundraising, when people are far more likely to be coming to your site with the express purpose of donating.  A light box will make it significantly easier for them.  (If there are similar campaigns for you, like Giving Tuesday or a special anniversary.  For your non-profit, not your spouse, although you should get them something nice.)
    • Any ask with urgency associated.  For example, if you have a matching gift campaign and can count down to the end of the match in your lightbox, I strongly recommend it.
    • When you are using the lightbox not on your homepage, but after some other action.  For example, shadowboxing the confirmation page of an advocacy action with a campaign-specific ask is a great idea.
  • Make them easy to get out of.  If this offer isn’t for the person, you want them to still enjoy browsing your site, rather than hating your guts.
  • Make them easy to get out of in mobile.  It is a little known fact that hell has expanded to a tenth level to accommodate new technology-associated crimes.  These include:
    • Making a lightbox that is unclosable on a mobile device because you can’t get to the X
    • Taking mobile phone calls in a public restroom
    • Replying all to company wise emails
    • Putting a conference call on hold so that 70 other people have to listen to your hold music before they hang up on you and call back in
    • Making a lightbox that is unclosable on a mobile device.  Oh, I mentioned this one twice?  That’s because it’s a special hell within the special hell.  Like child molesters in prison or Kourtney Kardashian**.
  • Make it smart.  Don’t show a repeat visitor the same shadow box over and over; use cookies to either vary your offer or leave it off altogether.  Research shows that you get the same number of sign-ups showing your light box every month rather than every time.  Also, customize your ask to the content a person is seeing.  If they go to your statistics page, give them an offer for your statistics white paper.  This will increase your conversions and their satisfaction.
  • Give a reason. You don’t want a lot of text on your light box, but you do want to have at least one statement that says why a person would want to take the action you want them to take.
  • Wait about five seconds.  That gives the person enough time to look at (and more than enough time to judge) your site, but not enough time to lose them to another offer.
  • Make sure you are layering in other techniques.  In particular, social proof XXX (“join the other 123,456 people who enjoy our email newsletter”), color theory (having a button color that stands out), and having a no option that doesn’t fit with a person’s self-image (“yes, I’d like to join the fight” versus “no, I don’t care about the environment”) can all be very effective.

So, pop-ups didn’t really die that day, kids.  They evolved to be less intrusive, more useful, and something that both users and nonprofits can value.  Now, alla you young’uns get some sleep, and dream of increased conversion rates.

* No, not a typo.  There does not appear to be a consensus on whether people say light box or lightbox or shadowbox or shadow box.  Thus, I’ve included all of the words in this piece so that they can be found by search engines.  I suggest you do the same in your content marketing.

** I don’t actually understand this joke myself, but I’m told it’s funny.

Let there be lightbox

Web metric basics

We talked yesterday about email metrics; now it’s Web site metrics’ turn.

We start here with the most generic of all online metrics: traffic.  No less an authority that FiveThirtyEight says that we still don’t know how to measure Web traffic. The difference is how unique visitors are measured versus total visits.  If you are an advertiser, you want to make sure the 1,000,000 visits a person is claiming to her/his site aren’t just a guy hitting reload over and over again.  This can be done by cookie, by IP address.  

My advice on this is sacrilegious for a metrics guy: don’t worry too much about it as long as you are using a consistent system for measurement.  I’ve used mainly Google Analytics for this, because it’s free, but any system will have its own way of determining this.

From this number, you can derive revenue per visitor by simply dividing your annual online revenues by your number of visitors to determine revenue per visitor.  This is a nice benchmark because you can see what all of your optimization efforts add up to; everything you do to try to get someone to a donation page, what you do to convert them, your average gift tweaking, the value you derive from your email list — all of it adds up to revenue per visitor.

But more than that, revenue per visitor also allows you to see what you are willing to invest to get someone to your site.  Let’s say your revenue per visitor is right at the M+R Benchmarks Report 2016 average of $.65 per visitor.  If the average blog post you do results in an extra 1000 visitors to your site, you should in theory be willing to pay up to $650 to write, deliver, and market that blog post (because revenue per visitor is an annual figure, so acquiring someone at cost that you can then engage in the future is a beautiful thing).

I say in theory because revenue per visitor varies based on the type of content or interaction.  I’ll talk about this at the end because we need to go through the other metrics to break this down more efficiently.

A close cousin to revenue per visitor is site donation conversion rate, or how many of the people who come to your site donate.  Instead of dividing your annual online revenues by visitors, you’ll divide the number of donations by visitors.  This is one of two key inputs to revenue per visitor (the other being average gift) and is a good way of testing whether wholesale changes to your site are helping encourage people to give.  

I recently worked with someone who put a thin banner at the top of their site encouraging donation.  He was disheartened because less than half a percent of the people who came to the site clicked on the banner.  I asked him if the total clicks were additive to donation clicks (that is, they represented people who wouldn’t have clicked to donate otherwise) or substitutive (that is, total donation clicks didn’t go up; they just moved from another donate button to this bar).  We were able to tell not only because of the donation clicks went up over baseline, but because the site donation conversation rate went up.  Now we are working on a strategy to test this bar throughout the site and with different context-specific asks.

Drilling down from the site donation conversion rate is the page donation conversion rate.  This is people who donate to a donation page divided by visitors to your donation page.  It’s a standard measure of the quality of your donation page.  This and average donation on a donation page combine to create the revenue per page.  

Revenue per page is not only a good way of measuring which donation form is better — it’s a good way of getting a feel for the valuable content on your site.  See how many of the people who come to a page end up donating directly from the page (you can do sophisticated attribution models to determine this — going directly to a donation is a quick and dirty way of doing it) and what their average gift is.  Divide that by the number of visitors you have to that page and you can see what the revenue per page is on a non-donation page as well.

This is great information to have.  Let’s say the value of a visitor to your home page is 10 cents, to a program page is 20 cents, and to an advocacy page is 40 cents.  This helps you make decisions about your content.  Do you need better calls to action on your program page?  What should be your next home page feature? (Answer: probably something about advocacy)  Where should you direct the bulk of your Google Grant traffic?  Etc.

However, there is one thing missing from all of this.  You will note that I said site donation conversion rate and page donation conversion rate.  Usually metrics folks won’t put donation in there — it’s implied.

But there’s another conversion rate that’s vitally important and that’s conversion to a constituent.  Remember that the conversion to donation process often is a series of smaller steps.  You need constituents who subscribe to your email newsletter, volunteer for your activities, and read your social media posts (OK, maybe not that last one).  A person has given you permission to talk to them is a valuable thing and should not be forgotten about.

So there’s also a site constituent conversion rate and page constituent  conversion rate — how good are your pages at capturing people.  Only when you have this to add to your revenue per page do you have a true measure of page quality.

But wait!  How do you add people converted to revenue?

That’s the topic for tomorrow as we discuss how to value a constituent.

Web metric basics

Escaping fixed ask strings

Most of the science of ask strings that we’ve talked about is related to variable ask strings that depend on who the potential donor is.  However, when acquiring new donors, this is often not possible, since you know little to nothing about who the person is (yet).  Thus, while we’ll talk mostly about variable ask strings or topics that apply to both fixed and variable ask strings, it’s important to discuss fixed ask strings.

Namely, don’t use them whenever possible.  Yes, they are necessary for some acquisition purposes, but the effort to customize them to even what little you know about a donor is worthwhile.  Some tips:

Online donation forms are usually customizable.  CDR Fundraising Group estimates that this simple step can increase your response rate by 50% and your average gift by 40%.  In fact, they’ve posted code for how to do this in Salsa Labs. What if you don’t use Salsa Labs?  Usually searching for “dynamic ask strings XXname of giving platformXX” will get you some tips on how to.

But if these tips are Greek to you, you can always take a shortcut: setting up multiple donation forms with different ask amounts and sending the links to customized segments of your audience.  This isn’t ideal, but it gets you most of the way there.  Even if you take a very shortcut and have a $100+ versus under $100 versions of your donation form to send, you will be customizing the experience for your online donor a little bit.

Use intelligence from your outside list selects.  If you are like many organizations, your outside list selects will feature a minimum threshold below which you won’t accept donors (often $5 or $10).  Chances are you have tested into these amounts:one list is productive without a threshold, so you haven’t incurred the cost; another had subpar performance, so you asked for a more select group of donors.

Chances are, your $10+ donors from one list will behave differently from your $5+ donors from other and from your “anything goes” donors from list number three.  Thus, you can use this threshold as a customization point for your fixed ask, making sure to ask people who give more for more.

Make sure your ask string testing doesn’t select just one winner.  When you test an ask string in acquisition, there’s a temptation to treat it like a traditional control and test, where a winner is chosen and rolled out with.  Here, however, you may find that even though the majority of lists performed best with your control ask string, there were a few lists that had demonstrably better results with your test version.  Since different lists have different donor characteristics, you may get better results by keeping with an ask string that better fits those donors.

Use modeling to determine your ask.  List cooperatives will be only too happy to create models for you.  Chances are, they can do a response model that maximizes response and another that maximizes average gift.  The folly is when both of these groups get the same ask strings when they were set up with different goals in mind.

However, you don’t have to use a co-op or pay a PhD to run a basic model.  Simply take the average gifts from your current donors at acquisition by ZIP code, standardize them (rounding to the nearest five or ten for fluency), and use that as the basis for your fixed ask strings.  After all, there’s no reason you have to treat 90210 as the same as 48208 in Detroit.

Make sure you are using information from multichannel giving when running a conversion program.  Sadly, walkers, event donors, volunteers, online donors, and e-newsletter subscribers are often dropped into an offline acquisition with nary a thought as to ask string.  Please don’t do this.  You could be asking your $500 online donor or your gala chair to sign a $20 check.  It’s debatable whether it would be worse if they didn’t give a gift or if they did.

Instead, make sure all giving, not just channel-specific giving, is taken into account when formulating your asks.  Additionally, even if someone has not given, you can apply filters like ZIP code or historical data (e.g., last time, your volunteers’ average donation was twice that of your e-newsletter subscribers; why not ask for twice as much?) to your ask string.
Hopefully, these tips help make even your fixed ask string more customized.


This is a special bonus Sunday blog post.  As I was writing my mini-book on ask strings, I realized this was a topic I hadn’t covered yet on the blog, so I’m putting up a draft version of the content here.  Please let me know what you think at nick@directtodonor.com so I can improve it.  And, if you would like a free copy of the book when it is ready, sign up for my weekly newsletter here.

Escaping fixed ask strings

Building awareness versus actually doing something

OK, that headline is harsher than I meant it.  Awareness is a necessary and useful precondition for many nonprofits.  Using an example I know well, drunk driving was a late-night joke just a few decades ago.  It took awareness activities to alert a nation to the fact that it is an unnecessary, tragic, and violent crime.

But does raising awareness sell?  That is, do people want to donate money to raise awareness about an issue or organization?  Or do they want to fund efforts to remediate wrongs directly?  Robert Smith and Norbert Schwarz wanted to find out.

Actually, being good scientists, they wanted to analyze donor’s metacognition about awareness activities vis-à-vis whether the cause was already in the donor consideration set.  Which means the same thing when you translate it into English.

They found three major things:

  • When people knew more about a charity and its work, they were more likely to donate to it and the more they were likely to donate.  The researchers actually manipulated this knowledge in a cool way. They asked subjects questions about what they had read about a charity, but there were two sets of questions: an easy one and a hard one.  The people who got the easy set of questions and thus thought they knew more about the subject were more likely to donate.
  • This result reversed when the charity was engaging in awareness activities.  That is, if people thought they knew all about the charity and its aims (that is, they got the easy questions), they were less likely to want to invest in the charity’s efforts to raise awareness.
  • Looking at actual donations (not just intent to give), people gave far more to help than to raise awareness when they knew a lot about a cause.  They gave slightly more to help raise awareness when they didn’t think they knew a lot about a cause.

This makes a good bit of sense.  If you think the average person (which people usually consider to be a slightly dumber version of themselves) knows about something, why would donate money to raise awareness?  On the flip side, if you felt there was a story that was undertold, that people needed to hear, you might ante up.

This has a major implication for nonprofits as they mature: what got you here won’t get you where you are going.  In the infancy stage of a nonprofit, it is acceptable simply to point at a problem and say “this is a problem; we need to get more people like you to acknowledge the problem.”  However, as nonprofits mature and people are aware of the issue the cause represents, it needs either to adjust its fundraising efforts to focus on what it is doing to solve the problem or to find more obscure areas of its cause to reenergize its donor base.

This also has implications for donor communications: there’s a difference between what you talk about to acquire a donor and to retain one.  That is, people who are your supporters know you and your issues (or, at least, think they do).  They don’t want to support awareness activities for things they think people already know about.  On the other hand, people who are new to your organization may be willing to chip in to help spread the word.

So remember your audience when you are pitching both helping and awareness activities for greater results.

Building awareness versus actually doing something

Implications of more donors versus better donors

Let’s say you’ve organizationally had the debate that we’ve been following the past three days and you have come down on the side of better donors: you’ve taken into account all of the long-time and non-financial benefits of lower-dollar donors and still can’t make the average $10 or less donor work for you organizationally.

Here are the steps you can take in your program to skew your results toward getting fewer, better donors.  Note that if you decide the other way — neither of these approaches are right or wrong — just do the opposite of everything listed below.

Up your ask strings.  As we’ve seen in two different studies of ask strings (here and here), increasing the bottom number on your ask string increases average gift.  If you are in a Pareto efficient model like we talked about on Monday, there will likely be a resultant decrease in response rate.  

Like this study indicates, I would do this with single donors and not try to get my multi-donors to elevate when they aren’t ready to.  There, I think you would be wise to keep the highest previous contribution as the base donation, but increase your multiply.

Change your defaults.  This can be the default online (where you have the radio button start on $50 instead of $25) or the amount you circle on a direct mail piece with the social proof “Most people give X”.  Moving the default up should get you fewer higher-value donors.

Move up your list selects.  When you rent or exchange with outside lists, even if a list works well for you with no qualifier on it, you can request only $5+ or $10+ donors to that organization.  It will cost a little bit more to get that list, but you will be able to cut some of the potential tippers out of your program.

Incidentally, there is a trick you can do here with a list that performs well and offers a higher-value list select (say, $50+): rent the list twice.  Once, rent it with a $10+ select and the other with a $50+ select.  Then, you can separate out your ask strings to those two lists and mail the $50+ list twice (like multis) with an appropriate ask string.

Work with your modeling agencies and coops.  They will be more than happy to build you a model that maximizes gift instead of maximizes response rates.

Invest in telemarketing upgrades.  Upgrading seems to work better when people talk with other people.  I would counsel doing this with a monthly giving ask with the appropriate audience — it’s literally the gift that keeps on giving.

Shift your lapsed reacquisition selects.  Because you “own” those names, you have the most freedom to play around with who you are trying to reacquire.  You may be able to change the complexion of your file by communicating less deeply (say, moving from 12 months to six months) among under $10 and more deeply (say, moving from 36 months to 48 months) among your $50+ donors.

Use ZIP modeling.  This can work with both acquisition and donor communications.  In both cases, you can get more aggressive about your ask strings with wealthy ZIP codes.  In acquisition, you may even choose to omit the bottom half (or whatever) percent of ZIP codes from some lists.  As with tighter donation selects, you will pay a bit more for those names, but you will get higher average gifts.

Invest in your second gift apparatus.  This is probably a good idea regardless, but if you are going to bleeding donors intentionally, you are going to need a way to make sure you are converting those you do bring on.  This may be an investment you only make for $20+ donors or the like, but a welcome series for this audience will help you keep the donors you want to keep.

Thanks for reading.  Be sure to sign up for my newsletter to keep up with the latest debate.

Also, I’d appreciate it if you’d let me know at nick@directtodonor.com if you like the debate format.  If so, we can try this with some other hot topics in nonprofit direct marketing.  If not, then we need never speak of this again.

Implications of more donors versus better donors