Warning: this is a long post…because choosing an ESP is a big deal!
Choosing an ESP is like selecting a spouse
Disclaimer: I have never chosen a spouse.
When you choose an ESP, you’re looking for stability, a good fit, room to grow together and a trust fund…er, the first three. You’re not looking for sexy features that are appealing today but will fade over time. Choosing a vendor is a long term commitment; you might select a provider knowing that you’re going to get divorced in a few years when you grow out of your relationship, but remember, the switching costs are high. Both the financial costs and the time investment in migrating your data and integrating with a new platform warrant serious consideration up front.
Exception: If you’re a startup betting on totally bananas, Groupon-style overnight user growth, you need a different game plan.
Make a list of ‘must-have’ traits
Define your requirements before your get started. I’m a broken record, but there it is. Create a simple vendor requirements doc. to guide you through the process. If it’s thorough, it’ll also knock the socks off potential vendors and let them know you mean business. Neurotic freaks like me can even use this list to create a comparison spreadsheet to log how vendors stack up on each of these items during your sales call! Totally baller tool.
1. Dedicated IP: If the ESP doesn’t provide a dedicated IP, other users with whom you’re sharing the IP could potentially compromise your sender reputation. Likewise, the ESP might reserve the right to police your account and suspend it at any time if they feel that your volume of complaints, etc. put other senders at risk (I’ve heard this about MailChimp…). Update: if your send volume is large enough you may want to use multiple dedicated IPs. Consider starting with at least one for your transactional emails and another for your marketing newsletters. I’ll follow up on this in a future post.
2. Email types: Do you need a vendor who supports transactional real-time ‘triggered’ emails, automated daily emails, one-off Marketing newsletters?
3. API database integration: Is this a must-have? Nice-to-have? Don’t need it? Will it add value to your emails? Will it increase efficiency by freeing up your engineers and letting your marketers create email filters/targets on-the-fly? What integration options are available?
4. Dynamic merge fields: Again, how critical is this? Sure, it lets you throw in some sick personalization, but will it add tangible value to your communication strategy?
5. Conditional content: See above. Does this fall into the bells & whistles category?
6. A/B testing capabilities: How sophisticated? Subject line only? Body content? Specific sections of body content? Delivery times? You get it =)
7. Analytics: Start with the basics. At a minimum, they should be abe to show you # sent, # delivered, # bounce, #complaints, open rate (BS metric=), clicks, click-through-rate (clicks/delivered).
Ask questions on the first date
Remember that list you made above? It’s time to put it to work! Cover all your bases with each vendor, moving down your list to confirm whether they offer your make it or break it features:
- What was their uptime last year? You’re looking for 99.9%.
- What type of backup infrastructure do they have in place for downtime? How is data backed up/recovered?
- What’s their policy regarding the collection of email addresses? Will they allow you to mail to rented lists? Single opt-in? Only double opt-in?
- Who owns the unsubscribe process? Can you manage this on your end, including the link in emails and the page to which you send users?
- What type of performance monitoring do they offer? Is there a way to set alert thresholds for unusually high/low volume, etc.?
- What channels do they offer for customer support? Ticketing system for email? Chat? Phone support?
- What’s the acceptable support response time set forth in their SLA? If you’re having a critical email issue, believe me, you’re going to want an immediate response.
Don’t get in bed together too soon
Start with drinks, then dinner, then maybe a weekend away. Give it at least three dates.
(1) Discovery call
(2) Platform demo
(3) Technical call (top three vendors and the engineering lead from your team and theirs)
Google your potential partner
When you think you’re ready to take the plunge, it’s time to dig up some dirt. Absolutely request references from you top three vendors. But those guys aren’t going to tell it like it is. Check out Quora, ask around within professional organizations, tap into your network and reach out to other clients to find out what people really think.
If you ain’t no punk, holla ‘we want pre-nup’
Ask yourself, what would Kanye do? I’m going to bet he’d negotiate like a mutherf*cker and get it in writing. When you’re ready to sign the deal, put on your poker face and get down to business. Nothing’s off-limits (we convinced our existing vendor to revise our SLA!).
- Highlight your projected growth trajectory. Include supporting examples. Do they really want to miss out on the opportunity to be the ESP to the next Foursquare?
- Bring some hard numbers. Get quotes from several potential vendors, not just your final three. Use this to your advantage when you’re driving down the CPM with prospective vendors. And do not, I repeat, do not, agree to any set up fees.
- Find out when their sales cycle ends. Make sure your closing date is ~1-2 days before the end of the cycle; your bargaining power will increase dramatically.
- Agree to sign on for a longer term in exchange for a lower CPM. And insist that there be no penalty for early termination! I know what you’re thinking- there’s no way they’ll go for that. Wrong! You both know it’s in your best interest for this relationship to be sustainable long-term. You don’t want to have to pay for another wedding…they just want to put you in the mindset of long-term commitment.
And then they lived happily ever after…