RFPs: 3 Lessons for Organizations Issuing Them

Posted on March 8, 2011 Under Management
Sign posted on a fence

By Flickr user supertobor

We’re outsourcing more work these days, so we’re putting more thought into how to best match vendors with projects. Recent RFPs include the full range of big and small — from a complete re-thinking of our web site (still in progress) to a fast-turnaround WordPress theme.

Here’s what I’ve learned so far about issuing successful RFPs. Stay tuned for the next post, which will be about my requests for the firms and contractors who respond. [Updated to add link.]

1. Share your budget range.

This causes more angst than any other aspect of issuing RFPs. I was skeptical at first, but I’m now firmly on the side of sharing. In my experience, a transparent budget doesn’t seem to lead to jacked-up prices, which is what many organizations worry about. But it does seem to lead to more appropriate proposals. (See Confluent Forms and Idea Champions for the vendor point of view.)

A few years ago, I put out an RFP without a budget, and got bids from $14K to $90K. The high bid offered way more service than we needed, and the low bid cut way too many corners. Working from just the list of deliverables — even a really clear list of deliverables — there is just too much leeway for firms to interpret the scope. This happens on small projects, too. On another project, we received bids ranging from $500 to $15K.

For our complete site overhaul, we took a different approach. We didn’t think a vendor could deliver high-quality work for less than a certain fee — and we didn’t want them to try! On the other hand, we only have so much money budgeted. So we decided to share a range rather than just one number.

Happily, the bids did not cluster at the higher number. They were distributed throughout our range, from the very bottom to just over the top. And they all seemed to be fair prices for the specific work offered.

2. Get help writing the RFP.

Sometimes, the reason that you’re cagey with your budget is that you don’t know how much you should pay. Uncertainty can plague other aspects of the RFP, too — you might not know how long the project should take, what skills the vendor needs or what information the vendor might require to assemble an accurate bid.

We’ve used two tactics to get help, and they’ve both worked out well.

In the first case, we issued a project description with as much detail we could, along with the following request:

We’re looking for help integrating data from a Salesforce database into a Drupal site. We plan to hire a contractor or firm to do the integration, and we know what we want the end result to be. But because of our limited experience in this area, we’d like to talk with potential contractors as we put together the RFP.

A note on Progressive Exchange introduced me to this idea, which is sometimes known as a request for skills, or an RFS. We added this line, too:

If scoping the project turns out to be a significant amount of work, we’re happy to treat it as phase one of a larger project.

We didn’t want to take advantage of anyone’s goodwill. I don’t think any of the conversations or notes took a vendor more than an hour — and they were incredibly helpful. This is still in progress, but I’m confident we’ll be able to issue a much more focused RFP. Better yet, we’ve met some vendors we’ll definitely invite to reply.

In the other case, we knew we needed a lot more help than we could get in an hour. So we hired Blenderbox to provide the research, a framework and the sheer number of hours that it took to write a complex RFP. This particular RFP needed to unite requirements from many stakeholders throughout the organization (again, see the advice from Confluent Forms), and we weren’t sure we had either the time or the perspective to do it well.

We came out with an RFP that captured our needs, and I don’t know that we could have done it without the reinforcements that Blenderbox provided.

3. Target who you send the RFP to.

People come down on both sides of this question, too. If you invite only specific firms to reply, you could miss potentially great matches. If you send it to everyone and their brother, you could drown in responses.

Again, I’ll speak from our experience. We got eight complete responses to our RFP for our full site overhaul. That added up to almost 400 pages of reading. It took a full week to digest it all, review portfolios, compare them, discuss them and follow up with all the vendors. It was grueling.

I don’t think the solution is more responses — but how about better responses?

The only way we’ve found to get better responses is to know who’s good, and ask them to respond. Many of the best firms aren’t searching message boards looking for cattle call RFPs — they don’t need to. They choose carefully what RFPs to respond to, because it costs them hundreds, if not thousands, of dollars to put together a good proposal. (Many thanks to Andy at ClearLeft for walking me through the way a firm makes these choices.)

So figure out who you want to respond, and ask them to. You can identify good people over time, as you tune in to who’s doing good work and who’s looked to as a leader by their peers. Or you can do it in one burst, researching and seeking recommendations right before you issue the RFP.

Either way, invest a little extra time (in our case, a day or two) to find people or companies you’d be excited to work with. That’s your best shot at actually getting to work with them.

That’s not the whole story…

We’re just one organization with one set of experiences. If you issue RFPs, what are some of the lessons you’ve learned? Please post in the comments!

And finally, the organization issuing the RFP is, of course, only half the equation. Stay tuned — later this week, I’ll share my three requests for vendors responding to RFPs. [Updated to add link.]